Draft
Minimum age for clear-cutting native species with energetic
potential in the Brazilian semi-arid
Journal Canadian Journal of Forest Research
Manuscript ID cjfr-2016-0392R3
Manuscript Type Note
Date Submitted by the Author 02-Dec-2016
Complete List of Authors Santana Otacilio Universidade Federal de Pernambuco Biophysics and Radiobiology
Keyword growth rings firewood Semi-Arid rotation age bioenergy
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Minimum age for clear-cutting native species with energetic potential in the Brazilian 1
semi-arid region 2
3
Otacilio Antunes Santana1 4
Av Prof Moraes Rego ndeg 1235 Departamento de Biofiacutesica e Radiobiologia (DBR) 5
Mestrado Profissional em Rede Nacional para o Ensino das Ciecircncias Ambientais 6
(PROFCIAMB) Centro de Biociecircncias (CB) Universidade Federal de Pernambuco (UFPE) 7
Cidade Universitaria Recife Pernambuco Brazil 50670-901 8
otaciliosantanagmailcom 9
otaciliosantanaufpebr 10
+55 81 997371266 11
1Author for correspondence 12
13
14
15
16
17
18
19
20
21
22
23
24
25
Page 1 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Abstract 26
What is the minimum age for clear-cutting native species with energetic potential in the 27
Brazilian semi-arid region This was the central question of this study which aimed at 28
estimating the minimum cutting cycle for native woody species from an area covered by a 29
sustainable forest management plan Individuals (n = 240) of twelve species native to the 30
semi-arid region were measured and wood discs were taken from them to collect data relative 31
to i) volume by an industrial 3D scanner ii) density using a high-frequency densitometer 32
iii) age by a growth ring measuring device and iv) the lower heating value by an oxygen 33
bomb calorimeter and an elemental analyzer Data were fitted to mathematical models by 34
regression analysis to determine the relationship and significance between the variables 35
analyzed The estimated minimum age for the harvesting of native woody species was 47 36
years determined by the age at which there was stabilization of volume growth curves over 37
time and an increase in density and heating power (wood retraction) 38
39
Keywords growth rings firewood rotation age bioenergy 40
41
42
43
44
45
46
47
48
49
50
Page 2 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
1 Introduction 51
The mean annual increment (MAI) and the current annual increment (CAI) are parameters 52
that allow estimation of the optimal age of timber harvest under sustainable forest 53
management but their calculation is laborious (costly expensive and requires a large data set) 54
and time consuming in woody species of the Brazilian semi-arid region Although sustainable 55
forest management plans are in place for this region such difficulty results in ages for clear 56
cut cycles without a scientific basis to justify these possible cycles Unlike rapidly growing 57
species of Eucalyptus and Pinus that invest their biomass mainly in shoots native woody 58
species of the Brazilian semi-arid region have slow initial growth investing most of their 59
biomass initially in roots (high root to shoot ratio) this as a support structure to reduce water 60
loss and against herbivorous attack (from which organs find refuge in the soil) (Lima and 61
Rodal 2010 Costa et al 2014 Santana and Encinas 2016) In addition the individual 62
architecture of woody plants in the Brazilian semi-arid region is based on broad ramifications 63
of the trunk and branches which makes it difficult to estimate the independent variables 64
namely height and diameter used in volumetric calculations (Barros et al 2010) 65
In addition to shoot volume other variables are essential for estimating the energy 66
potential of wood The bulk density chemical composition (eg cellulose and lignin) 67
percentage of vessels and parenchyma moisture content and size of fibers are some 68
parameters that influence the heating value of wood However these parameters are 69
correlated and represented by the density of the wood which has a significant direct and 70
proportional relationship with the heating value of wood (Gebreegziabher et al 2013) The 71
age of the individual plant is another important factor because with advancing age and the 72
stabilization of growth in height woody individuals increase the carbon accumulation on their 73
trunks and branches thus increasing the value of the massvolume ratio of the timber Adult 74
wood is characterized by high density long tracheids thick cell walls high percentage of 75
Page 3 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
4
latewood low percentage of spiral grain low percentage of nodes high percentage of 76
cellulose low percentage of compression wood high transverse shrinkage smaller microfibril 77
angle and greater mechanical strength (Rowell 2012) 78
The optimal age for harvesting a planted forest stand for wood production is when the 79
current annual increment curve intersects the mean annual increment curve At that point the 80
rate of growth in wood volume begins to reduce over time and the economic profitability in 81
relation to forest investment also begins to fall (Prodan 1968) In semi-arid regions in 82
addition to a reduced rate of wood volume growth as trees age wood density often increases 83
over time as an ecophysiological strategy of native tree species to cope with moisture 84
limitations (Liu et al 2013 Klein et al 2014 Santana and Encinas 2016) This wood 85
retraction is evident in the aerial part of the plant and can enhance the rootshoot ratio even if 86
the leaf area index continues to develop over time (Santana and Encinas 2016) Before of the 87
wood retraction occurs a reduction of the rate of growth in wood volume and an increasing of 88
the rate of growth in wood density are observed (Paula 1993 Santana and Encinas 2016) 89
Thus in semi-arid regions the time at which wood volume growth levels off and when the 90
density and LHV curves change their inclination angle in relation to age could represent a 91
parameter defines the best time for harvesting wood destined for bioenergy uses (Liu et al 92
2013 Klein et al 2014) There is not an analytical or exact method for calculating this point 93
of levelling off or for detecting a critical change in the inclination angle of wood density and 94
LHV curves but these trends are supported by observations in the literature (Althoff et al 95
2016 Santana and Encinas 2016) 96
In 2013 91 of the Brazilian energy matrix was based on wood used as firewood and 97
charcoal which has increased the demand for this resource by approximately 5 per year 98
(Brasil 2013) In the semi-arid region of the state of Pernambuco the extracted wood is used 99
as an energy source of the Polo Gesseiro do Araripe (Araripina Microregion) which has 39 100
Page 4 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Minimum age for clear-cutting native species with energetic potential in the Brazilian 1
semi-arid region 2
3
Otacilio Antunes Santana1 4
Av Prof Moraes Rego ndeg 1235 Departamento de Biofiacutesica e Radiobiologia (DBR) 5
Mestrado Profissional em Rede Nacional para o Ensino das Ciecircncias Ambientais 6
(PROFCIAMB) Centro de Biociecircncias (CB) Universidade Federal de Pernambuco (UFPE) 7
Cidade Universitaria Recife Pernambuco Brazil 50670-901 8
otaciliosantanagmailcom 9
otaciliosantanaufpebr 10
+55 81 997371266 11
1Author for correspondence 12
13
14
15
16
17
18
19
20
21
22
23
24
25
Page 1 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Abstract 26
What is the minimum age for clear-cutting native species with energetic potential in the 27
Brazilian semi-arid region This was the central question of this study which aimed at 28
estimating the minimum cutting cycle for native woody species from an area covered by a 29
sustainable forest management plan Individuals (n = 240) of twelve species native to the 30
semi-arid region were measured and wood discs were taken from them to collect data relative 31
to i) volume by an industrial 3D scanner ii) density using a high-frequency densitometer 32
iii) age by a growth ring measuring device and iv) the lower heating value by an oxygen 33
bomb calorimeter and an elemental analyzer Data were fitted to mathematical models by 34
regression analysis to determine the relationship and significance between the variables 35
analyzed The estimated minimum age for the harvesting of native woody species was 47 36
years determined by the age at which there was stabilization of volume growth curves over 37
time and an increase in density and heating power (wood retraction) 38
39
Keywords growth rings firewood rotation age bioenergy 40
41
42
43
44
45
46
47
48
49
50
Page 2 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
1 Introduction 51
The mean annual increment (MAI) and the current annual increment (CAI) are parameters 52
that allow estimation of the optimal age of timber harvest under sustainable forest 53
management but their calculation is laborious (costly expensive and requires a large data set) 54
and time consuming in woody species of the Brazilian semi-arid region Although sustainable 55
forest management plans are in place for this region such difficulty results in ages for clear 56
cut cycles without a scientific basis to justify these possible cycles Unlike rapidly growing 57
species of Eucalyptus and Pinus that invest their biomass mainly in shoots native woody 58
species of the Brazilian semi-arid region have slow initial growth investing most of their 59
biomass initially in roots (high root to shoot ratio) this as a support structure to reduce water 60
loss and against herbivorous attack (from which organs find refuge in the soil) (Lima and 61
Rodal 2010 Costa et al 2014 Santana and Encinas 2016) In addition the individual 62
architecture of woody plants in the Brazilian semi-arid region is based on broad ramifications 63
of the trunk and branches which makes it difficult to estimate the independent variables 64
namely height and diameter used in volumetric calculations (Barros et al 2010) 65
In addition to shoot volume other variables are essential for estimating the energy 66
potential of wood The bulk density chemical composition (eg cellulose and lignin) 67
percentage of vessels and parenchyma moisture content and size of fibers are some 68
parameters that influence the heating value of wood However these parameters are 69
correlated and represented by the density of the wood which has a significant direct and 70
proportional relationship with the heating value of wood (Gebreegziabher et al 2013) The 71
age of the individual plant is another important factor because with advancing age and the 72
stabilization of growth in height woody individuals increase the carbon accumulation on their 73
trunks and branches thus increasing the value of the massvolume ratio of the timber Adult 74
wood is characterized by high density long tracheids thick cell walls high percentage of 75
Page 3 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
4
latewood low percentage of spiral grain low percentage of nodes high percentage of 76
cellulose low percentage of compression wood high transverse shrinkage smaller microfibril 77
angle and greater mechanical strength (Rowell 2012) 78
The optimal age for harvesting a planted forest stand for wood production is when the 79
current annual increment curve intersects the mean annual increment curve At that point the 80
rate of growth in wood volume begins to reduce over time and the economic profitability in 81
relation to forest investment also begins to fall (Prodan 1968) In semi-arid regions in 82
addition to a reduced rate of wood volume growth as trees age wood density often increases 83
over time as an ecophysiological strategy of native tree species to cope with moisture 84
limitations (Liu et al 2013 Klein et al 2014 Santana and Encinas 2016) This wood 85
retraction is evident in the aerial part of the plant and can enhance the rootshoot ratio even if 86
the leaf area index continues to develop over time (Santana and Encinas 2016) Before of the 87
wood retraction occurs a reduction of the rate of growth in wood volume and an increasing of 88
the rate of growth in wood density are observed (Paula 1993 Santana and Encinas 2016) 89
Thus in semi-arid regions the time at which wood volume growth levels off and when the 90
density and LHV curves change their inclination angle in relation to age could represent a 91
parameter defines the best time for harvesting wood destined for bioenergy uses (Liu et al 92
2013 Klein et al 2014) There is not an analytical or exact method for calculating this point 93
of levelling off or for detecting a critical change in the inclination angle of wood density and 94
LHV curves but these trends are supported by observations in the literature (Althoff et al 95
2016 Santana and Encinas 2016) 96
In 2013 91 of the Brazilian energy matrix was based on wood used as firewood and 97
charcoal which has increased the demand for this resource by approximately 5 per year 98
(Brasil 2013) In the semi-arid region of the state of Pernambuco the extracted wood is used 99
as an energy source of the Polo Gesseiro do Araripe (Araripina Microregion) which has 39 100
Page 4 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Abstract 26
What is the minimum age for clear-cutting native species with energetic potential in the 27
Brazilian semi-arid region This was the central question of this study which aimed at 28
estimating the minimum cutting cycle for native woody species from an area covered by a 29
sustainable forest management plan Individuals (n = 240) of twelve species native to the 30
semi-arid region were measured and wood discs were taken from them to collect data relative 31
to i) volume by an industrial 3D scanner ii) density using a high-frequency densitometer 32
iii) age by a growth ring measuring device and iv) the lower heating value by an oxygen 33
bomb calorimeter and an elemental analyzer Data were fitted to mathematical models by 34
regression analysis to determine the relationship and significance between the variables 35
analyzed The estimated minimum age for the harvesting of native woody species was 47 36
years determined by the age at which there was stabilization of volume growth curves over 37
time and an increase in density and heating power (wood retraction) 38
39
Keywords growth rings firewood rotation age bioenergy 40
41
42
43
44
45
46
47
48
49
50
Page 2 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
1 Introduction 51
The mean annual increment (MAI) and the current annual increment (CAI) are parameters 52
that allow estimation of the optimal age of timber harvest under sustainable forest 53
management but their calculation is laborious (costly expensive and requires a large data set) 54
and time consuming in woody species of the Brazilian semi-arid region Although sustainable 55
forest management plans are in place for this region such difficulty results in ages for clear 56
cut cycles without a scientific basis to justify these possible cycles Unlike rapidly growing 57
species of Eucalyptus and Pinus that invest their biomass mainly in shoots native woody 58
species of the Brazilian semi-arid region have slow initial growth investing most of their 59
biomass initially in roots (high root to shoot ratio) this as a support structure to reduce water 60
loss and against herbivorous attack (from which organs find refuge in the soil) (Lima and 61
Rodal 2010 Costa et al 2014 Santana and Encinas 2016) In addition the individual 62
architecture of woody plants in the Brazilian semi-arid region is based on broad ramifications 63
of the trunk and branches which makes it difficult to estimate the independent variables 64
namely height and diameter used in volumetric calculations (Barros et al 2010) 65
In addition to shoot volume other variables are essential for estimating the energy 66
potential of wood The bulk density chemical composition (eg cellulose and lignin) 67
percentage of vessels and parenchyma moisture content and size of fibers are some 68
parameters that influence the heating value of wood However these parameters are 69
correlated and represented by the density of the wood which has a significant direct and 70
proportional relationship with the heating value of wood (Gebreegziabher et al 2013) The 71
age of the individual plant is another important factor because with advancing age and the 72
stabilization of growth in height woody individuals increase the carbon accumulation on their 73
trunks and branches thus increasing the value of the massvolume ratio of the timber Adult 74
wood is characterized by high density long tracheids thick cell walls high percentage of 75
Page 3 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
4
latewood low percentage of spiral grain low percentage of nodes high percentage of 76
cellulose low percentage of compression wood high transverse shrinkage smaller microfibril 77
angle and greater mechanical strength (Rowell 2012) 78
The optimal age for harvesting a planted forest stand for wood production is when the 79
current annual increment curve intersects the mean annual increment curve At that point the 80
rate of growth in wood volume begins to reduce over time and the economic profitability in 81
relation to forest investment also begins to fall (Prodan 1968) In semi-arid regions in 82
addition to a reduced rate of wood volume growth as trees age wood density often increases 83
over time as an ecophysiological strategy of native tree species to cope with moisture 84
limitations (Liu et al 2013 Klein et al 2014 Santana and Encinas 2016) This wood 85
retraction is evident in the aerial part of the plant and can enhance the rootshoot ratio even if 86
the leaf area index continues to develop over time (Santana and Encinas 2016) Before of the 87
wood retraction occurs a reduction of the rate of growth in wood volume and an increasing of 88
the rate of growth in wood density are observed (Paula 1993 Santana and Encinas 2016) 89
Thus in semi-arid regions the time at which wood volume growth levels off and when the 90
density and LHV curves change their inclination angle in relation to age could represent a 91
parameter defines the best time for harvesting wood destined for bioenergy uses (Liu et al 92
2013 Klein et al 2014) There is not an analytical or exact method for calculating this point 93
of levelling off or for detecting a critical change in the inclination angle of wood density and 94
LHV curves but these trends are supported by observations in the literature (Althoff et al 95
2016 Santana and Encinas 2016) 96
In 2013 91 of the Brazilian energy matrix was based on wood used as firewood and 97
charcoal which has increased the demand for this resource by approximately 5 per year 98
(Brasil 2013) In the semi-arid region of the state of Pernambuco the extracted wood is used 99
as an energy source of the Polo Gesseiro do Araripe (Araripina Microregion) which has 39 100
Page 4 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
1 Introduction 51
The mean annual increment (MAI) and the current annual increment (CAI) are parameters 52
that allow estimation of the optimal age of timber harvest under sustainable forest 53
management but their calculation is laborious (costly expensive and requires a large data set) 54
and time consuming in woody species of the Brazilian semi-arid region Although sustainable 55
forest management plans are in place for this region such difficulty results in ages for clear 56
cut cycles without a scientific basis to justify these possible cycles Unlike rapidly growing 57
species of Eucalyptus and Pinus that invest their biomass mainly in shoots native woody 58
species of the Brazilian semi-arid region have slow initial growth investing most of their 59
biomass initially in roots (high root to shoot ratio) this as a support structure to reduce water 60
loss and against herbivorous attack (from which organs find refuge in the soil) (Lima and 61
Rodal 2010 Costa et al 2014 Santana and Encinas 2016) In addition the individual 62
architecture of woody plants in the Brazilian semi-arid region is based on broad ramifications 63
of the trunk and branches which makes it difficult to estimate the independent variables 64
namely height and diameter used in volumetric calculations (Barros et al 2010) 65
In addition to shoot volume other variables are essential for estimating the energy 66
potential of wood The bulk density chemical composition (eg cellulose and lignin) 67
percentage of vessels and parenchyma moisture content and size of fibers are some 68
parameters that influence the heating value of wood However these parameters are 69
correlated and represented by the density of the wood which has a significant direct and 70
proportional relationship with the heating value of wood (Gebreegziabher et al 2013) The 71
age of the individual plant is another important factor because with advancing age and the 72
stabilization of growth in height woody individuals increase the carbon accumulation on their 73
trunks and branches thus increasing the value of the massvolume ratio of the timber Adult 74
wood is characterized by high density long tracheids thick cell walls high percentage of 75
Page 3 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
4
latewood low percentage of spiral grain low percentage of nodes high percentage of 76
cellulose low percentage of compression wood high transverse shrinkage smaller microfibril 77
angle and greater mechanical strength (Rowell 2012) 78
The optimal age for harvesting a planted forest stand for wood production is when the 79
current annual increment curve intersects the mean annual increment curve At that point the 80
rate of growth in wood volume begins to reduce over time and the economic profitability in 81
relation to forest investment also begins to fall (Prodan 1968) In semi-arid regions in 82
addition to a reduced rate of wood volume growth as trees age wood density often increases 83
over time as an ecophysiological strategy of native tree species to cope with moisture 84
limitations (Liu et al 2013 Klein et al 2014 Santana and Encinas 2016) This wood 85
retraction is evident in the aerial part of the plant and can enhance the rootshoot ratio even if 86
the leaf area index continues to develop over time (Santana and Encinas 2016) Before of the 87
wood retraction occurs a reduction of the rate of growth in wood volume and an increasing of 88
the rate of growth in wood density are observed (Paula 1993 Santana and Encinas 2016) 89
Thus in semi-arid regions the time at which wood volume growth levels off and when the 90
density and LHV curves change their inclination angle in relation to age could represent a 91
parameter defines the best time for harvesting wood destined for bioenergy uses (Liu et al 92
2013 Klein et al 2014) There is not an analytical or exact method for calculating this point 93
of levelling off or for detecting a critical change in the inclination angle of wood density and 94
LHV curves but these trends are supported by observations in the literature (Althoff et al 95
2016 Santana and Encinas 2016) 96
In 2013 91 of the Brazilian energy matrix was based on wood used as firewood and 97
charcoal which has increased the demand for this resource by approximately 5 per year 98
(Brasil 2013) In the semi-arid region of the state of Pernambuco the extracted wood is used 99
as an energy source of the Polo Gesseiro do Araripe (Araripina Microregion) which has 39 100
Page 4 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
4
latewood low percentage of spiral grain low percentage of nodes high percentage of 76
cellulose low percentage of compression wood high transverse shrinkage smaller microfibril 77
angle and greater mechanical strength (Rowell 2012) 78
The optimal age for harvesting a planted forest stand for wood production is when the 79
current annual increment curve intersects the mean annual increment curve At that point the 80
rate of growth in wood volume begins to reduce over time and the economic profitability in 81
relation to forest investment also begins to fall (Prodan 1968) In semi-arid regions in 82
addition to a reduced rate of wood volume growth as trees age wood density often increases 83
over time as an ecophysiological strategy of native tree species to cope with moisture 84
limitations (Liu et al 2013 Klein et al 2014 Santana and Encinas 2016) This wood 85
retraction is evident in the aerial part of the plant and can enhance the rootshoot ratio even if 86
the leaf area index continues to develop over time (Santana and Encinas 2016) Before of the 87
wood retraction occurs a reduction of the rate of growth in wood volume and an increasing of 88
the rate of growth in wood density are observed (Paula 1993 Santana and Encinas 2016) 89
Thus in semi-arid regions the time at which wood volume growth levels off and when the 90
density and LHV curves change their inclination angle in relation to age could represent a 91
parameter defines the best time for harvesting wood destined for bioenergy uses (Liu et al 92
2013 Klein et al 2014) There is not an analytical or exact method for calculating this point 93
of levelling off or for detecting a critical change in the inclination angle of wood density and 94
LHV curves but these trends are supported by observations in the literature (Althoff et al 95
2016 Santana and Encinas 2016) 96
In 2013 91 of the Brazilian energy matrix was based on wood used as firewood and 97
charcoal which has increased the demand for this resource by approximately 5 per year 98
(Brasil 2013) In the semi-arid region of the state of Pernambuco the extracted wood is used 99
as an energy source of the Polo Gesseiro do Araripe (Araripina Microregion) which has 39 100
Page 4 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
5
gypsum mines and 869 industries to cement manufacture which fill 73 of their energy 101
requirements with wood but the small manufacturers use 100 wood to meet their energy 102
needs The production of a 10000 kg of plaster requires 05 st native wood which has an 103
annual demand of 55middot104 million kgyear (Silva 20082009) The sustainable production of 104
wood fuel would require a total area of 155000 hectares specifically for the plaster industry 105
with a cutting cycle ranging from 10 to 15 years But the Northeastern Plants Association 106
(APNE 2014) has the registration of 94 sustainable forest management plans in the state of 107
Pernambuco covering an area of 39748 hectares insufficient to meet the energy demands of 108
the Polo which grows 10 (Sindugesso 2011) to 25 per year in addition to a growing 109
ceramics industry (CPRH 2014) 110
The goal of this study was to estimate the minimum age for a clear cutting cycle of 111
native woody species from an area governed by a sustainable forest management plan in the 112
semi-arid region according to requirements for implementation and maintenance of Annual 113
Production Unit described by the Normative Instruction of the Brazil Ministry of the 114
Environment (number 1 of 06252009 ndash MMA 2009) 115
116
2 Materials and Methods 117
The wood analyzed came from a forest management farm (Figure 1A) in the state of 118
Pernambuco (8deg35rsquoS and 37deg59rsquoW) having a semi-arid climate with annual rainfall below 119
1000 mm (550 mm in 2015 APAC 2016) and mean annual temperature greater than 27degC 120
or a BSh climate according to the Koumlppen classification (Peel et al 2007) This farm is 121
located on the lsquoSertaneja Meridional e Raso da Catarinarsquo geomorphological depression The 122
soils have been classified into four classes Haplic Cambisol Salic Gleysol Yellow Oxisol 123
Haplic Planosol and the vegetation has been classified into Closed Arboreal Open Arboreal 124
and Shrub structural classes (Aguiar et al 2013) 125
Page 5 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
6
The vegetation analyzed came from the clearcutting of experimental plots in a 126
sustainably managed forest of the Semi-Arid Forest Management Network (MMA 2009) The 127
twelve predominant tree species have a cumulative importance value index of 96 measured 128
in 20 plots of 50 x 100 m (10 ha) with a 1271 stemsha density (diameter at breast height ge 1 129
cm) (Aguiar et al 2013) evaluated according to the Permanent Plot Measurements Protocol 130
of the Semi-Arid Forest Management Network and Rodal et al (1992) 131
The species and number of evaluated individuals of each species were Acacia 132
kallunkiae JWGrimes amp Barneby Fabaceae (n = 13) Acacia piauhiensis Benth Fabaceae 133
(n = 15) Anadenanthera colubrina (Vell) Brenan Var cebil (Griseb) Reis Fabaceae (n = 134
27) Aspidosperma pyrifolium Mart Apocynaceae (n = 35) Poincianella pyramidalis (Tul) 135
L P Queiroz Fabaceae (n = 12) Croton blanchetianus Baill Euphorbiaceae (n = 9) 136
Erythrina velutina Willd Papilionoideae (n = 17) Jatropha elliptica (Pohl) MullArg 137
Euphorbiaceae (n = 33) Maytenus rigida (Mart) Benth Celastraceae (n = 16) Mimosa 138
caesalpiniifolia Benth Fabaceae (n = 8) Myracrodruon urundeuva All Anacardiaceae (n = 139
29) and Peltophorum dubium (Spreng) Taub Fabaceae (n = 26) 140
Harvested trees were analyzed for volumetry (Figure 1B) densitometry (Figure 1C) 141
dating (Figure 1C) and calorimetry (Figure 1D) The volume of the wood (m3) was measured 142
by an industrial 3D scanner (3D scanner - SK-DK-FX - four Lens Foshan Shangke 143
Machinery Co Ltd Guangdong Province China) in which all parts (trunk + branches) were 144
reduced to 1 m length (Figure 1B) in a mobile studio in locus The density of the wood and 145
the age (growth rings) were measured from sectioned discs 2 cm thick by means of high-146
frequency densitometry (Schinker et al 2003) in LignoStationTM using LignoScop (Rinntech 147
Heidelberg Germany) (Figure 1C) to scan surfaces using cameras attached to the microscope 148
and LignoScan to scan the wood surface using high-frequency (1600 dpi) with 0001 mm 149
precision (Shchupakivskyy et al 2014) Standard procedures (Fritts 1976 Lisi et al 2008) 150
Page 6 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
7
and COFECHA software (The University of Tennessee Knoxville Grissino-Mayer 2001) 151
were used for cross-dating and standardization The irregularity of inter-annual rain or of 152
another climate event could be revised by the crossdating analysis This analysis could 153
identify growth zone and false growth rings in an inter-annual period (Pagotto et al 2015) 154
The equipment performed 1200 factorial (1200) crossdating analyzes (240 studied 155
individuals with five replicates each) 156
The Higher Heating Value (HHV) of the wood was calculated by collecting data 157
measured with a calorimeter (C6000 global standards 210 IKAreg Staufen Germany) 158
(Guumlnther et al 2012) Samples were ground to a maximum size of 25 microm Lower Heating 159
Value (LHV) was determined by subtracting the heat of vaporization of the water vapor from 160
the higher heating value The water vapor was determined by variance of the hydrogen 161
content of the sample measured through the Element-Analyzer (Vario EL Elementar 162
Analysesystem GmbH Langenselbold Germany) (Figure 1D) These procedures followed the 163
standards DIN EN ISO 1716-2009 DIN 51900-1 2000 DIN 51900-3 2005 (Guumlnther et al 164
2012) and ABNTNBR 863384 165
The relationships between variables (Y = volume and X = density Y= volume and X 166
= age Y = lower heating value and X = density Y = density and X = age Y = lower heating 167
value and X = volume and Y = lower heating value and X = age) were performed by fitting 168
the data to a wide array of mathematical growth models (Table 1) using regression analysis to 169
calculate a coefficient of determination (R2) root-mean-square error (RMSE) significance 170
level (p) and the fit curve and the selection of the best fit model (based on maximizing R2 171
minimizing RMSE and minimizing p) for each case (Zar 1999) The regression analysis was 172
preceded by the DrsquoAgostino Normality test (DAgostino et al 1990) for each variable to 173
validate statistical premises The models were chosen as indicated by Kleinbaum et al (2013) 174
The possible multicollinearity among variables was calculated by the Farrar-Glauber test 175
Page 7 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
8
(Farrar amp Glauber 1976) Adjustments graphics statistical deviation and coefficient of 176
variation were calculated using Statistica 12 (Statsoft Dell Tulsa USA) 177
178
3 Results 179
The mean age of the study population was 46 years (CV = 152 Figure 2A) ranging from 180
98 years (Peltophorum dubium) to 5 years (Croton blanchetianus) The mean volume was 181
012 m3 (CV = 119 Figure 2B) per individual tree with individuals ranging in size from 182
a minimum volume of 0003 m3 (Croton blanchetianus) to a maximum volume of 040 m3 183
(Acacia kallunkiae) The mean wood density was 08 g cm-3 (CV = 71 Figure 2C) 184
varying between 04 g cm-3 (Jatropha elliptica) and 105 g cm-3 (Acacia kallunkiae) The 185
mean lower heating value was 4863 kcal kg-1 (CV = 62 Figure 2D) ranging from 186
3783 kcal kg-1 (Acacia piauhiensis) to 5697 kcal kg-1 (Erythrina velutina) 187
Data reflecting the relationship between volume and density and between volume and 188
age were well fit to sigmoidal models (5 parameters) significantly (R2 gt 086 RMSE lt 005 189
p lt 0001 Figure 3A and 3B Table 2) In both relationships volume began to rise sharply 190
(curve growth gt 45deg slope relative to the dependent variable) to the point where it became 191
constant (curve growth lt 5deg slope in relation to the dependent variable) and then slowed 192
relative to growth of the independent variable (density or age) The relationship between the 193
lower heating value (LHV) and density was the most significant (R2 gt 097 RMSE lt 003 p 194
lt 0001) directly and significantly increasing with increasing density (Figure 3C Table 2) 195
Multicollinearity was not found in the models tested (p gt 0800) Considering the relationship 196
between density and LHV with the age of individuals data fitted to the exponential growth 197
model (double and with five parameters) in which from a given year 47 years for density 198
and 495 for LHV there is a levelling off of the curve (Figure 3D and 3F Table 2) with a 199
more marked increase in the values of these variables (curve growthgt 45 degslope relative to the 200
Page 8 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
9
dependent variable) than previously (growth curve asymp 30 deg slope relative to the dependent 201
variable) As for the relationship between LHV and volume data fitted to the Chapman model 202
(four parameters) (Figure 3E Table 2) indicating that in the higher volume growth phase the 203
increase in LHV value is less pronounced (curve growth lt 15deg slope in relation to the 204
dependent variable) than when the volume growth begins to stabilize (growth curve gt 45deg 205
slope in relation to the dependent variable) 206
The minimum age for clear cutting cycle in an area of Sustainable Forest Management 207
Plan in the semi-arid region with the destination of wood with energy purposes suggested by 208
the variables analyzed was from 468 (47) years (Figure 3B) This is the average tree age at 209
which there is a stabilization of the volume growth (curve growth lt 5deg slope in relation to the 210
dependent variable) and pronounced growth of the LHV (from 495 years Figure 3F) and 211
density (from 47 years Figure 3D) 212
213
4 Discussion 214
One limitation of this study lies in the impracticality of working with a larger sample size 215
given the time required for cutting and loading for transportation (about 24 hours) making it 216
difficult to scan other woody individuals in the semi-arid region However 240 individuals 217
were measured with five replicates of each measurement or analysis improving the reliability 218
and accuracy of data collected 219
Our findings corroborate what has already been reported for the area (Aguiar et al 220
2013) which before the Sustainable Forest Management Plan had reduced logging (gt 20 221
individuals ha year) The values of volume density and LHV varied within the 95 222
confidence intervals reported in the literature for the mean values of the species analyzed 223
volume (from biomass) from 0001 m3 (Santana and Souto 2006) to 050 m3 (Silva and 224
Sampaio 2008) density from 03 g cm-3 (Lima and Rodal 2010) to 12 g cm-3 (Paula 1993) 225
Page 9 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
10
and LHV from 3800 kcal kg-1 (Medeiros Neto et al 2012) to 5800 kcal kg-1 (Quirino et al 226
2004) 227
The rapid growth in the volume of tree species over the years and a subsequent 228
stabilization of this growth is commonly observed in the literature not only for fast growing 229
plantation species (Encinas et al 2011) but also for native species (Felker 1986 Lima 1986 230
Vico et al 2015 Santana and Encinas 2016) This stabilization is considered as the saturation 231
period of the plant individual both by limited environmental resources by competition with 232
others in their environment but also by individual senescence marked by the genetics of the 233
species (Hunt 1982 Felker 1986 Soares et al 2011) The onset of the saturation senescence 234
or slow growth process as shown in Figure 3B from 47 years is marked in forestry science 235
by the beginning of the cutting cycle for vegetation intended for production of firewood and 236
charcoal (Rode et al 2014 Althoff et al 2016) When the volume tends to stabilize other 237
variables such as density (Figure 3 D) and heating value (Figure 3F) which have a high 238
correlation (R2 gt 097) continue to grow in value While a saturation of those values was not 239
observed in this work it is however emphasized in the literature (Hunt 1982 Felker 1986) 240
This reduction of the rate of growth of the wood volume and an increasing of the rate of 241
growth of the wood density and lower heating value (LHV) infer in the begin of wood 242
retraction Thus from the age of 47 the harvesting is strongly indicated 243
In other dry forests on Earth the cutting cycle of wood is ranges from 63 to 97 years 244
much longer than the 47 years derived here The causes for these longer rotation periods are 245
i) in India to avoid urban occupation (Agarwal et al 2016) ii) in Australia and Africa to 246
conserve the habitat for large mammals (Bhadouria et al 2016) iii) in West Africa to reduce 247
wildfire susceptibility and microclimate changes (Scheitera and Savadogo 2016) iv) in 248
Cameroon and Panama to preserve forest stands that are used in popular religious and sacred 249
rituals (Kemeuze et al 2016 Seijo et al 2016) v) in Costa Rica India Papua New Guinea 250
Page 10 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
11
and Southern Africa to produce medicinal bioproducts (Schmiedel et al 2016) vi) in 251
Morocco and Argentina to perpetuate ethnobotanical and other cultural uses of natural 252
resources (Martiacutenez 2015 Blanco and Carriegravere 2016) and vii) in China to maintain the 253
groundwater flow near to surface by hydraulic lift from trees (Xiao and Huang 2016) In all 254
cases like in this work biodiversity preservation and firewood sustainability also were 255
objectives of forest management (Althoff et al 2016 Santana 2016) 256
257
5 Conclusion 258
The estimate of the minimum age for cutting cycle of native woody species in the evaluated 259
area in Brazilrsquos semi-arid region was 47 years determined by the age at which there was 260
stabilization of volume growth over time and an increase in density and heating power (wood 261
retraction) 262
263
Acknowledgements 264
The author is grateful to Pro-Reitoria de Pesquisa e Pos-Graduacao of the Universidade 265
Federal de Pernambuco (PROPESQUFPE) for the financial and logistical support and to 266
Research Group lsquoEducometriarsquo (UFPECNPq) by discussion and survey support The author 267
is very grateful to the reviewers for their careful and meticulous reading of the paper 268
269
References 270
Agarwal S Nagendra H and Ghate R 2016 The Influence of Forest Management 271
Regimes on Deforestation in a Central Indian Dry Deciduous Forest Landscape Land 5 27-272
43 doi 103390land5030027 273
Aguiar M M B Santana O A Inaacutecio E dos S B Amorim L B de and Almeida-274
Cortez J S 2013 Tree resilience after clear-cutting in sustainable forest management of 275
Page 11 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
12
semi-arids areas In 2nd United Nations Convention to Combat Desertification Scientific 276
Conference Edited by Walter J Ammann UN Bonn p 157-158 Available from 277
lthttpsgooglmf9uYUgt [accessed 12 January 2016] 278
Althoff T D Menezes R S C Carvalho A L Pinto A S Santiago G A C F Ometto 279
J P H B Randow C and Sampaio E V D S B 2016 Climate change impacts on the 280
sustainability of the firewood harvest and vegetation and soil carbon stocks in a tropical dry 281
forest in Santa Teresinha Municipality Northeast Brazil Forest Ecology and Management 282
360 367-375 doi 101016jforeco201510001 283
APAC ndash Agecircncia Pernambucana de Aacuteguas e Clima 2016 Monitoramento Pluviometrico 284
[online] Available from lt httpwwwapacpegovbrgt [accessed 11 November 2016] 285
APNE ndash Associaccedilatildeo de Plantas do Nordeste 2014 Centro Nordestino de Informaccedilotildees sobre 286
Plantas Planos de Manejo Sustentaacuteveis da Caatinga [online] Available from 287
lthttpwwwcniporgbrgt [accessed 12 January 2014] 288
Associaccedilatildeo Brasileira de Normas Teacutecnicas ndash ABNT 1984 NBR 8633 Carvatildeo vegetal ndash 289
determinaccedilatildeo do poder caloriacutefico ABNT Rio de Janeiro 12p 290
Barros B C Silva J A A Ferreira R L C and Rebouccedilas A C M N 2010 Volumetria 291
e sobrevivecircncia de espeacutecies nativas e exoacuteticas no Poacutelo Gesseiro do Araripe-PE Ciecircncia 292
Florestal 20 641-647 doi 105902198050982422 293
Bhadouria R Singh R Srivastava P and Raghubanshi A S 2016 Understanding the 294
ecology of tree-seedling growth in dry tropical environment a management perspective 295
Energy Ecology and Environment 1 296ndash309 doi 101007s40974-016-0038-3 296
Blanco J and Carriegravere S M 2016 Sharing local ecological knowledge as a human 297
adaptation strategy to arid environments Evidence from an ethnobotany survey in Morocco 298
Journal of Arid Environments 127 30-43 doi 101016jjaridenv201510021 299
Page 12 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
13
Brasil Empresa de Pesquisa Energeacutetica 2013 Balanccedilo Energeacutetico Nacional 2013 ndash Ano base 300
2012 Relatoacuterio Siacutentese EPE Rio de Janeiro 55p 301
CPRH ndash Agecircncia Ambiental do Meio Ambiente de Pernambuco 2014 Madeira Ilegal ndash 302
Buscas [online] Available from lthttpwwwcprhpegovbrResultadosasppage=2amptexto= 303
MADEIRA20ILEGALgt [accessed 01 June 2014] 304
Costa T L Sampaio E V S B Sales M F Accioly L J O Althoff T D Pareyn F 305
G C Albuquerque E R G M and Menezes R S C 2014 Root and shoot biomasses in 306
the tropical dry forest of semi-arid Northeast Brazil Plant and Soil 378 113-123 doi 307
101007s11104-013-2009-1 308
DAgostino R B Belanger A and Dagostino Jr R B 1990 A suggestion for using 309
powerful and informative tests of normality American Statistician 44 316-321 doi 310
1023072684359 311
Encinas J I Santana O A and Imantildea C R 2011 Volumetric and Economic optimal 312
rotations for firewood production of Eucalyptus urophylla in Ipamery State of Goias 313
Floresta 41 905-912 doi 105380rfv41i425353 314
Farrar D and Glauber R 1976 Multicollinearity in Regression Analysis the Problem 315
Revisited Review of Economics and Statistics 49 92-107 doi 1023071937887 316
Felker P 1986 Establishment and productivity of tree plantings in semiarid regions Elsevier 317
Amsterdam 444 p 318
Fritts H C 1976 Tree Rings and Climate Academic Press London 566p 319
Grissino-Mayer H D 2001 Evaluating crossdating accuracy a manual and tutorial for the 320
computer program COFECHA Tree-Ring Research 57 205ndash221 321
Gebreegziabher T Oyedun A O and Hui C W 2013 Optimum biomass drying for 322
combustion ndash A modeling approach Energy 53 67-73 doi 101016jenergy201303004 323
Page 13 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
14
Guumlnther B Gebauer K Barkowski R Rosenthal M and Bues C T 2012 Calorific value 324
of selected wood species and wood products European Journal of Wood and Wood Products 325
70 755ndash757 doi 101007s00107-012-0613-z 326
Hunt R 1982 Plant growth curves The functional approach to plant growth analysis 327
Edward Arnold Ltd London 248 p 328
Kemeuze V A Sonwa D J Nkongmeneck B A and Mapongmetsem P M 2016 329
Sacred groves and biodiversity conservation in semi-arid area of Cameroon Case study of 330
Diamare plain In Quels botanistes pour le 21e siegravecle Meacutetiers enjeux et opportuniteacutes Edited 331
by N R Rakotoarisoa S Blackmore and B Riera UNESCO Paris pp 171-183 Available 332
from lthttpsgooglf0cxCAgt [accessed 12 March 2016] 333
Klein T Hoch G Yakir D and Koumlrner C 2014 Drought stress growth and nonstructural 334
carbohydrate dynamics of pine trees in a semi-arid forest Tree Physiology 71 1-12 doi 335
101093treephystpu071 336
Kleinbaum D Kupper L Nizam A and Rosenberg E 2013 Applied regression analysis 337
and other multivariable methods Cengage Learning Michigan 928p 338
Lima P C F 1986 Tree productivity in the semiarid zone of Brazil Forest ecology and 339
Management 16 5-13 doi 1010160378-1127(86)90003-4 340
Lima A L A and Rodal M J N 2010 Phenology and wood density of plants growing in 341
the semi-arid region of northeastern Brazil Journal of Arid Environments 74 1363-1373 342
doi 101016jjaridenv201005009 343
Lisi C S Tomazello Filho M Botosso P C Roig F A Maria V B R Ferreira-Fedele 344
L and Voigt A R A 2008 Tree-ring formation radial increment periodicity and 345
phenology of tree species from a seazonal semi-deciduous forest in southeast Brazil IAWA 346
Journal 29 189-207 doi 10116322941932-90000179 347
Page 14 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
15
Liu H Park W A Allen C D Guo D Wu X Anenkhonov O A Liang E 348
Sandanov DV Yin Y Qi Z and Badmaeva N K 2013 Rapid warming accelerates tree 349
growth decline in semi-arid forests of Inner Asia Global Change Biology 19 2500-2510 350
doi 101111gcb12217 351
Martiacutenez G J 2015 Cultural patterns of firewood use as a tool for conservation A study of 352
multiple perceptions in a semiarid region of Cordoba Central Argentina Journal of Arid 353
Environments 121 84-99 doi 101016jjaridenv201505004 354
Medeiros Neto P N Oliveira E Calegari L Almeida A M C Pimenta A S and 355
Carneiro A C O 2012 Caracteriacutesticas Fiacutesio-Quiacutemicas e energeacuteticas de duas espeacutecies de 356
ocorrecircncia no Semiaacuterido Brasileiro Ciecircncia Florestal 22 579-588 doi 357
105902198050986624 358
Ministeacuterio do Meio Ambiente ndash MMA 2009 Dispotildee sobre procedimentos teacutecnicos para 359
elaboraccedilatildeo apresentaccedilatildeo execuccedilatildeo e avaliaccedilatildeo teacutecnica de Planos de Manejo Florestal 360
Sustentaacutevel-PMFS da Caatinga e suas formaccedilotildees sucessoras e daacute outras providecircncias 361
Instruccedilatildeo Normativa Ndeg 1 de 25 de junho de 2009 Diaacuterio Oficial da Uniatildeo ndash Sec 1 120 93 362
Pagotto M A Roig F A Ribeiro A de S and Lisi C S 2015 Influence of regional 363
rainfall and Atlantic sea surface temperature on tree-ring growth of Poincianella pyramidalis 364
semiarid forest from Brazil Dendrochronologia 35 14-23 doi 365
101016jdendro201505007 366
Paula J E 1993 Madeiras da Caatinga uacuteteis para produccedilatildeo de energia Pesquisa 367
Agropecuaacuteria Brasileira 28 153-165 368
Peel M C Finlayson B L and McMahon T A 2007 Updated world map of the Koumlppen-369
Geiger climate classification Hydrology and Earth System Sciences 11 1633ndash1644 doi 370
105194hess-11-1633-2007 371
Page 15 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
16
Quirino W F Vale A T Andrade A P A Abreu V L S and Azevedo A C S 2004 372
Poder Caloriacutefico da Madeira e de Resiacuteduos Lignolceluloacutesicos Biomassa amp Energia 1 173-373
182 374
Prodan M 1968 Forest Biometrics Pergamon Oxford doi 101016B978-0-08-012441-375
450001-8 376
Rede de Manejo Florestal da Caatinga 2005 Protocolo de mediccedilotildees de parcelas permanentes 377
Comitecirc Teacutecnico Cientiacutefico Associaccedilatildeo Plantas do Nordeste Recife 21 p 378
Rodal M J N Sampaio E V S B and Figueiredo M A 1992 Manual sobre meacutetodos de 379
estudo floriacutestico e fitossocioloacutegico - ecossistema Caatinga Sociedade Botacircnica do Brasil 380
Brasiacutelia 24 p 381
Rode R Leite G L Silva M L Ribeiro C A A S and Binoti D H B 2014 The 382
economics and optimal management regimes of eucalyptus plantations A case study of 383
forestry outgrower schemes in Brazil Forest Policy and Economics 44 26-33 doi 384
101016jforpol201405001 385
Rowell R M 2012 Handbook of wood chemistry and wood composites CRC press Boca 386
Raton 703 p 387
Santana J A S and Souto J S 2006 Diversidade e estrutura fitossocioloacutegica da Caatinga 388
na Estaccedilatildeo Ecoloacutegica do Seridoacute-RN Revista de Biologia e Ciecircncias da Terra 6 232-242 389
Santana O A 2016 Resistecircncia social na Caatinga aacuterida a narrativa de quem ficou no 390
colapso ambiental Desenvolvimento e Meio Ambiente 38 419-438 doi 391
httpdxdoiorg105380dmav38i043574 392
Santana O A and Encinas J I 2016 Dendrophysiological plant strategies of Poincianella 393
pyramidalis (Tul) LP Queiroz after wood herbivory in semiarid region of Paraiacuteba - Brazil 394
Acta Scientiarum Biological Sciences 38 179-186 doi 104025actascibiolsciv38i229089 395
Page 16 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
17
Schneider P R 1998 Anaacutelise de regressatildeo aplicada agrave engenharia florestal UFSMCEPEF 396
Santa Maria 236 p 397
Schinker M G Hansen N Spiecker H 2003 High-frequency densitometry mdash a new 398
method for the rapid evaluation of wood density variations IAWA Journal 24 231ndash239 doi 399
10116322941932-90001592 400
Shchupakivskyy R Clauder L Linke N and Pfriem A 2014 Application of high-401
frequency densitometry to detect changes in early- and latewood density of oak (Quercus 402
robur L) due to thermal modification European Journal of Wood and Wood Products 72 5-403
10 doi 101007s00107-013-0744-x 404
Scheitera S and Savadogo P 2016 Ecosystem management can mitigate vegetation shifts 405
induced by climate change in West Africa Ecological Modelling 332 19ndash27 doi 406
101016jecolmodel201603022 407
Schmiedel U Araya Y Bortolotto M I Boeckenhoff L Hallwachs W Janzen D 408
Kolipaka S S Novotny V Palm M Parfondry M Smanis A and Toko P 2016 409
Contributions of paraecologists and parataxonomists to research conservation and social 410
development Conservation Biology 30 506ndash519 doi 101111cobi12661 411
Seijo M M Huerta R P Torneacute J M Torneacute C M and Vidal E A 2016 Madera 412
Carbonizada en contextos funeraacuterios de la jefatura de Riacuteo Grande Panamaacute Antracologiacutea en el 413
sitio de el Cantildeo Chungaraacute 48 277-294 doi 104067S0717-73562016005000013 414
Silva G C and Sampaio E V S B 2008 Biomassas de partes aeacutereas em plantas de 415
caatinga Revista Aacutervore 32 567-575 doi 101590S0100-67622008000300017 416
Sindusgesso 2001 Newsletter [online] Available from lthttpwwwsindusgessoorgbrgt 417
[accessed 13 June 2014] 418
Page 17 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
18
Silva J A A 20082009 Potencialidades de florestas energeacuteticas de Eucalyptus no Poacutelo 419
Gesseiro do Araripe-Pernambuco Anais da Academia Pernambucana de Ciecircncia 420
Agronocircmica 5-6 301-319 421
Soares C P B Paula Neto F and Souza A L 2011 Dendrometria e inventaacuterio florestal 422
Editora UFV Viccedilosa 272p 423
Vico G Thompson S E Manzoni S Molini A Albertson J D Almeida‐Cortez J S 424
Fay P A Feng X Guswa A J Liu H Wilson T G and Porporato A 2015 Climatic 425
ecophysiological and phenological controls on plant ecohydrological strategies in seasonally 426
dry ecosystems Ecohydrology 8 660-681 doi 101002eco1533 427
Xiao Q and Huang M 2016 Fine root distributions of shelterbelt trees and their water 428
sources in an oasis of arid northwestern China Journal of Arid Environments 130 30-39 429
doi 101016jjaridenv201603004 430
Zar J 1999 Biostatistical analysis Prentice Hall New Jersey 431
Page 18 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 1 Map showing the origin of the tree individuals studied (A) and methodological steps volumetry (B) densitometry and dating (C) and calorimetry (D)
714x851mm (72 x 72 DPI)
Page 19 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 2 Age (A) volume (B) density (C) and lower heating value (LHV) (D) of twelve native species analyzed in the semi-arid region
1103x1446mm (72 x 72 DPI)
Page 20 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
Figure 3 Relationship between the analyzed variables volume lower heating value (LHV) density and age of the 240 woody individuals measured
1431x1577mm (72 x 72 DPI)
Page 21 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
1
Table 1 Mathematical models used for data fit
Model Equation
Polynomial (linear) i 0 1 i iY Xβ β ε= + sdot +
Polynomial (Quadratic) 2
i 0 1 i 2 i iY X Xβ β β ε= + sdot + sdot +
Polynomial (Cubic) 2 3
i 0 1 i 2 i 3 i iY X X Xβ β β β ε= + sdot + sdot + sdot +
Sigmoidal (3 parameters) 1 2
3
1i iX
Y
1 e
ββ
βε
minusminus
= +
+
Sigmoidal (4 parameters) i 2
3
1i 0 iX
Y
1 e
ββ
ββ ε
minusminus
= + +
+
Sigmoidal (5 parameters) 4
i 2
3
1i 0 i
X
Y
1 e
ββ
β
ββ ε
minusminus
= + + +
Logistic (3 parameters) 3
1i i
i
2
YX
1
β
βε
β
= +
+
Logistic (4 parameters) 3
1i 0 i
i
2
YX
1
β
ββ ε
β
= + +
+
Weibull (4 parameters)
41
4i 2 3
5
X ln2
i 1 iY 1 e
β
ββ ββ
β ε
minus + minus
= minus +
Weibull (5 parameters)
41
4i 2 3
5
X ln2
i 0 1 iY 1 e
β
ββ ββ
β β ε
minus + minus
= + minus +
Gompertz (3 parameters) Xi 2
3e
i 1 iY e
ββ
β ε minus minus= +
Gompertz (4 parameters) Xi 2
3e
i 0 1 iY e
ββ
β β ε minus minus= + +
Hill (3 parameters) 2
2 2
1 ii i
3 i
XY
X
β
β β
βε
β= +
+
Page 22 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
2
Hill (4 parameters) 2
2 2
1 ii 0 i
3 i
XY
X
β
β β
ββ ε
β= + +
+
Chapman (3 parameters) ( ) 32 iX
i 1 iY 1 eβββ εminus= minus +
Chapman (4 parameters) ( ) 32 iX
i 0 1 iY 1 eβββ β εminus= + minus +
Exponential Growth (Simple 1 parameter) 1 iX
i iY eβ εsdot= +
Exponential Growth (Simple 2 parameters) 2 iX
i 1 iY eββ εsdot= sdot +
Exponential Growth (Simple 3 parameters) 2 iX
i 0 1 iY eββ β εsdot= + sdot +
Exponential Growth (Double 1 parameter) 2 i 4 iX X
i 1 3 iY e eβ ββ β εsdot sdot= sdot + sdot +
Exponential Growth (Double 2 parameter) 2 i 4 iX X
i 0 1 3 iY e eβ ββ β β εsdot sdot= + sdot + sdot +
Yi = observed value of the dependent variable Xi = observed value of the independent variable βi = regression equation
parameters εi = fit error
Page 23 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research
Draft
3
Table 2 Significant fit of data to models for six analyzed relationships volume vs density (A) volume vs age (B) lower
heating value vs density (C) density vs age (D) lower heating value vs volume (E) and lower heating value vs age (F)
Model Function R2 RMSE p
A Sigmoidal (5
parameters)
i
i 0005X 0423
0001
0509Y 012
1 e
minus minus
= minus + +
086 0044 lt 0001
B
Sigmoidal (5
parameters)
i
i 3799X 23861
4287
0031Y 0007
1 e
minus minus
= + +
089 0037 lt 0001
C
Polynomial
(linear) i iY 3736 1738 X= + sdot 097 0026 lt 0001
D
Exponential
Growth (Double 5
parameters)
i i0001 X 0019 X
iY 0193 0248 e 0210 esdot sdot= minus + sdot + sdot 091 0034 lt 0001
E
Chapman (4
parameters) ( )i 6438
6297 X
iY 0004 0001 1 eminus sdot= + minus 090 0048 lt 0001
F
Exponential
Growth (Double 5
parameters)
i i0001 X 0044 X
iY 141594 514298 e 76118 esdot sdot= minus + sdot + sdot 094 0050 lt 0012
Source Author
Page 24 of 24
httpsmc06manuscriptcentralcomcjfr-pubs
Canadian Journal of Forest Research