1

Recovery from N saturation in Flemish forests under high N deposition

Arne Verstraeten, Johan Neirynck,Nathalie Cools, Peter Roskams, Maarten Hens

3rd ICP Forests Scientific Conference - 201426 - 28 May 2014, Athens, Greece

2

• Introduction – N saturation and recovery

• Objectives of this study

• Materials and methods- Study area- Sample collection- Chemical analysis- Data handling- Statistical analysis

• Results- pH- NO3

- leaching - N retention- Critical limits - DIN:DON

• Conclusions

I will present

3

• During the past decades elevated nitrogen (N) + sulphur (S) deposition caused N saturation and acidification of temperate forestsin Europe and North America.(e.g. Aber et al., 1989; Lorenz & Granke, 2009)

• Thanks to emission reduction, depositions stabilized or decreased in many areas. In European forests (2000-2010):

- S: 6% y-1 - N: 2% y-1

(Waldner et al., in press)

• Recently, several studies reported initial chemical recovery in the soil compartment, which was indicated by:

NO3- leaching

pH mobilization of accumulated organic C and N stocks

(e.g. Vanguelova et al., 2010; Oulehle et al., 2011)

Introduction – N saturation and recovery

Temperate forest ecosystems

4

Dissolved organic carbon (DOC) concentrations and fluxes increased in soil solution (2002-2012):• DOC concentrations: 3-7% y-1 (100% of plots)• DOC fluxes: 5-6% y-1 (40% of plots)

(results for deeper mineral soil)

(Verstraeten et al., 2014)

In 1994, acidifying depositions (throughfall + stemflow) at intensive forest monitoring plots were high : 42 kg inorganic N ha-1 y-1 and 27 kg S ha-1 y-1.

Trend analysis confirmed a decrease thereafter (1994-2010):

• SO42-: 5-6% y-1 (100% of plots)

• NH4+: 3-5% y-1 (100% of plots)

• NO3-: 1-2% y-1 (60% of plots)

Simultaneously, soil solution fluxes of NO3-, SO4

2- and Al

But critical loads/levels were still exceeded in 2010.

(Verstraeten et al., 2012)

Decline of depositions is reflected in soil solution

Flanders, northern Belgium

5

The main objective of the present study was to give a more complete answer to this question:• Is the N status of Flemish forests improving?

Objectives of this study

6

Data from 5 ICP Forests intensive monitoring plots were selected:- 2 coniferous Ravels (RAV) Brasschaat (BRA)

- 3 deciduous Wijnendale (WIJ) Gontrode (GON) Hoeilaart (HOE)

Materials and MethodsStudy area

Plot Coordinates Elevation Temperature Precipitation Tree species Age Former land use Basal area Throughfall Stemflow

  N E m °C mm   years   m²/ha % %

Coniferous forests                    

RAV 51°24'07'' 05°03'15'

'

35 10.4 887 Pinus nigra subsp. laricio 80 heath 44.9 68.3 -

BRA 51°18'28'' 04°31'11'' 14 10.8 882 Pinus sylvestris 81 heath 29.2 78.6 -

Deciduous forests                    

WIJ 51°04'11'' 03°02'14'

'

31 11.0 867 Fagus sylvatica 75 arable 36.5 62.0 14.7

GON 50°58'31'' 03°48'15'

'

26 10.6 786 Quercus robur, Fagus sylvatica 92 old growth 31.9 71.8 3.2

HOE 50°44'45'' 04°24'47'

'

129 10.7 854 Fagus sylvatica 101 old growth 28.9 70.9 5.8

7

8

•1994-2013•2 times per month•4 open field precipitation collectors•10 throughfall collectors•3-5 stemflow collectors (beech)•soil solution:

- organic layer (4-6 zero-tension lysimeters)- mineral soil (3 locations, 2-4 suction cup lysimeters per depth)

- topsoil- subsoil- deeper mineral soil

Sample collection

Throughfall collectors

10 m

9

on 500 ml composite subsamples of each fraction

1994-2013:

• pH (potentiometric, WTW multi 340i)

• NH4+, NO3

- and NO2- concentrations (ion chromatography)

2005-2013:

• Total N concentrations (continuous flow method, Skalar)

Chemical analysis

10

• Dissolved organic nitrogen: DON = Total N – (NH4+-N +

NO3- -N + NO2

--N)

• N fluxes = water fluxes * concentrations

• Stand deposition = throughfall + stemflow

• Soil water fluxes were calculated using Na+ as a “tracer ion” for each soil depth (Bailey et al., 2003)

• DIN:DON = (NH4+-N + NO3

- -N + NO2--N) * DON-1

Was proposed as an indicator to detect changes in ecosystem N statusAssumption: DON export is not related to N input(Williams et al., 2004)

Williams, M.W., Clow, D., Blett, T., 2004. A Novel Indicator of Ecosystem N Status: DIN to DON Ratio in Riverine Waters. American Geophysical Union, Fall Meeting 2004, abstract #H53F-05.

Data handling

N status DIN:DON0 <0.51 0.5-2.02 >2.0

11

• Trend analysis - Mann-Kendall test (MK) (Mann, 1945; Helsel & Hirsch, 2002)

- Seasonal Mann-Kendall test (SMK) (Hirsch et al., 1982; Hirsch & Slack, 1984)

estimated following Sen (Sen, 1968)

• Linear regression

Statistical analysis

12

pH in the deeper mineral soil (1994-2013)

1994 1996 1998 2000 2002 2004 2006 2008 2010 20123.0

3.5

4.0

4.5

5.0

pH

RAVBRAWIJGONHOEmean

Overall , on average 0.015 pH units y-1.

Plot Mean Slope RAV 4.14 0.016*** BRA 3.79 0.020*** WIJ 4.14 0.016*** GON 4.07 0.008*** HOE 4.54 0.007*

Trend (SMK)

Results

13

Mean annual N fluxes(2005-2013)

DIN

• depositionconiferous > deciduous(pollution climate)

• soil solutionconsiderable differences between plots

DON

• deposition3-4 kg ha-1 y-1 DON in open field, elevated:<1-2 kg ha-1 y-1 in unpolluted areas (e.g. Mustajärvi et al., 2008)

• soil solutionhighest in organic layer (litter and SOM decomposition)declines with depth (adsorption and uptake)1-4 kg ha-1 y-1 DON leaching

14

NH4+-N dominates deposition; nearly absent in mineral soil

NO3--N dominates in mineral soil

NO2--N negligible

%DON ±constant in deposition; considerable variation in soil solution

Mean flux of NH4+-N (grey bars), NO3

--N (white dashed bars), NO2--N

(grey dashed bars) and DON (black bars) shown as % of Total N flux.

Composition of N fluxes (%)(2005-2013)

0 20 40 60 80 100

%

Open field precipitation

Stand precipitation

Organic layer

Topsoil

Subsoil

RAVBRAWIJGONHOE

20-24%

22-31%

28-47%

15-58%

15

1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

10

20

30

40

50

NO

3- -N le

achi

ng f

lux

(kg

ha-1

y-1

)

RAVBRAWIJGONHOEmean

NO3-- leaching (1994-2013)

Flux of NO3-N (kg N ha-1 y-1) in the deeper mineral soil

Overall , on average 0.75 kg N ha-1 y-1.NO3

- leaching nearly halted quite suddenly in WIJ (2004) and HOE (2006).A sudden halt of NO3

- leaching was also observed by Oulehle et al. (2011).

Plot Mean Slope RAV 16.45 -0.70*** BRA 17.95 -1.27*** WIJ 12.37 -1.06*** GON 17.16 -0.44*** HOE 6.01 -0.25***

Trend (SMK)

16

N Retention (1994-2013)

1994 1996 1998 2000 2002 2004 2006 2008 2010 2012-100

-80

-60

-40

-20

0

20

40

60

80

100

N r

eten

tion

in t

he d

eepe

r m

iner

al s

oil (

% o

f st

and

depo

sitio

n)

RAVBRAWIJGONHOEmean

100 – (NO3--N flux deeper mineral soil) / (NO3

- -N + NH4+ -N stand deposition) * 100

Plot Mean Slope RAV 53.81 1.48*** BRA 41.17 3.25*** WIJ 57.78 3.63*** GON 16.59 HOE 63.16 1.33***

Trend (SMK)

Overall except in GON (high NO3- exports particularly in years

with spring drought and autumn wetness).

171994 1996 1998 2000 2002 2004 2006 2008 2010 2012

0

2

4

6

8

10

12M

on

ths

with

NO

3- co

nce

ntr

atio

n >

25

mg

l-1

1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

2

4

6

8

10

12

Mo

nth

s w

ith N

co

nce

ntr

atio

n >

5 m

g l-1

1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

2

4

6

8

10

12

Mo

nth

s w

ith N

co

nce

ntr

atio

n >

1 m

g l-1

RAVBRAWIJGONHOEmean

Critical limits: number of months with exceedance in the deeper mineral soil (1994-2013)(EEC, 1991; Sverdrup and Warfvinge ,1993; UNECE, 2007)

Enhanced sensitivity to frost and fungi (> 5 mg N l-1)

Reduced fine root biomass/root length (>3 mg N l-1)Elevated N leaching/N saturation (>1 mg N l-1)

1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

2

4

6

8

10

12

Mo

nth

s w

ith N

co

nce

ntr

atio

n >

3 m

g l-1

Guide value drinking water directive (>25 mg NO3- l-1)

Plot Mean Slope RAV 10.3 BRA 11.3 WIJ 7.9 -0.65*** GON 11.8 HOE 5.5 -0.54**

Plot Mean Slope RAV 8.5 -0.41** BRA 8.4 -0.41* WIJ 5.9 -0.71*** GON 10.2 HOE 2.8

Plot Mean Slope RAV 7.0 -0.67*** BRA 5.3 -0.57*** WIJ 5.8 -0.72*** GON 8.1 HOE 1.1

Plot Mean Slope RAV 6.8 -0.67*** BRA 4.5 -0.61*** WIJ 5.8 -0.79*** GON 7.5 HOE 1.0

Trend (MK) Trend (MK)

Trend (MK) Trend (MK)

18

DON – trend analysis (2005-2013)

( ). Longer time series needed (Waldner et al., in press)

2005 2006 2007 2008 2009 2010 2011 2012 20130

5

10

15

DO

N f

lux

deep

er m

iner

al s

oil (

kg h

a-1

y-1

)

2005 2006 2007 2008 2009 2010 2011 2012 20130

5

10

15

DO

N s

tand

dep

ositi

on (

kg h

a-1 y

-1)

RAVBRAWIJGONHOEmean

Plot Mean Slope RAV 8.6 BRA 7.3 WIJ 5.9 GON 6.5 0.03** HOE 6.1 0.04***

Plot Mean Slope RAV 3.8 BRA 2.7 WIJ 1.2 GON 2.5 HOE 2.5 0.02***

Trend (SMK)Trend (SMK)

Stand deposition Deeper mineral soil

19

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

ln(D

ON

flux

)

y = -0.38 - 0.04yR²A=-0.001

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -1.13 + 0.06yR²A=-0.001

Relation between (monthly) DON leaching and DIN deposition (2005-2013)

DON export not related to DIN input ( assumption for use of DIN:DON not violated)

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -1.72+ 0.08yR²A=0.0003

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -2.11 + 0.11yR²A=0.003

Organic layer Topsoil

Subsoil Deeper mineral soil

20

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -0.60 + 0.56yR²A=0.2812***

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

ln(D

ON

flux

)

y = 0.16 + 0.61yR²A=0.3742***

Relation between (monthly) DON leaching and DON deposition (2005-2013)

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -1.20 + 0.53yR²A=0.1961***

-6 -5 -4 -3 -2 -1 0 1 2-7

-6

-5

-4

-3

-2

-1

0

1

2

3

y = -1.53 + 0.52yR²A=0.1965***

On the contrast, DON export is significantly related to DON input at all depths

Organic layer Topsoil

Subsoil Deeper mineral soil

21

Relation between DON leaching and DON deposition (2005-2013)

BUT: Seasonal peak of DON in stand deposition (summer) not simultaneous with organic layer (winter) impact of DON deposition on soil solution DON is limited.

Similar result as for DOC (Verstraeten et al., 2014)

1 2 3 4 5 6 7 8 9 10 11 120

1

2

3

DO

N s

tand

dep

ositi

on (

kg h

a-1 y

-1)

1 2 3 4 5 6 7 8 9 10 11 120

2

4

6

8

DO

N fl

ux o

rgan

ic la

yer

(kg

ha-1

y-1

)

22

N-status - Relation between DIN:DON and DIN (2005-2013)

0 5 10 15 20 25 30 35 40 45

DIN deposition (kg ha-1 y-1)

0

1

2

3

4

5

6

DIN

:DO

N r

atio

open field precipitationstand precipitation

y = 2.09 + 0.11x

R²=0.2255

y = 1.43 + 0.08xR²=0.7325(*)

0 5 10 15 20 25 30 35 40 450

1

2

3

4

5

6

DIN

:DO

N r

atio

y = 0.05 + 0.08x

R²=0.896*

y = 0.79 + 0.17x

R²=0.9141*

DIN flux (kg ha-1 y-1)

organic layertopsoilsubsoildeeper mineral soilstage: 0

y = 0.39 + 0.12xR²=0.9839***

y = 0.37 + 0.32xR²=0.8664*

stage: 1

stage: 2

• deposition not significant (but only 5 plots) elevated DIN deposition (although DON is also elevated)

• soil solution significant at all depths (but only 5 plots) DIN:DON illustrates the differences in N saturation between plots Allows to evaluate the N status of each individual soil horizon

23

N status - DIN:DON (2005-2013)

• depositionaverage DIN:DON = 2.2-4.0 DIN:DON (60% of plots )

• soil solutionaverage DIN:DON = 0.7-5.4 DIN:DON (100% of plots )GON highly N-saturated (2)RAV and BRA (2 1)HOE and WIJ (1 0)

2007 and 2013: dry summer followedby wet autumn ‘flushing’longer time series needed

N status 0 1 2

2005 2006 2007 2008 2009 2010 2011 2012 20130

2

4

6

8

10

12

DIN

:DO

N r

atio

for

sta

nd d

epos

ition

RAVBRAWIJGONHOEmean

Trend (SMK)

Trend (SMK)Plot Mean Slope

RAV 3.7 -0.24* BRA 3.9 WIJ 3.1 -0.24** GON 3.7 -0.36** HOE 3.3

Plot Mean Slope RAV 3.3 -0.36*** BRA 3.6 -0.30*** WIJ 3.3 -0.35*** GON 2.9 -0.39*** HOE 2.2 -0.38***

2005 2006 2007 2008 2009 2010 2011 2012 20130

2

4

6

8

10

12

DIN

:DO

N r

atio

for

the

dee

per

min

eral

soi

l

24

Conclusions• In spite of high N depositions, N status of Flemish forests

improved, but not at equal rate.

• The improving N status was indicated by:- pH of the deeper mineral soil 100% of plots- NO3

- leaching 100% of plots- N retention 80% of plots- months with exceedance of critical limits - DIN:DON ratio in soil solution 100% of plots

• 1 deciduous plot remained highly N saturated (N status = 2)2 coniferous plots are gradually recovering (N status = 2 1)2 deciduous plots showed fast recovery (N status = 1 0)

25

Questions?

26

Relationship between DON and Soil C:N (2005-2013)

27

1994 1996 1998 2000 2002 2004 2006 2008 2010 20120

2

4

6

8

10

12

RAVBRAWIJGONHOEmean

Mon

ths

with

N c

once

ntra

tion

>0.

2 (c

onife

rous

tree

s)

o

r >

0.4

(dec

iduo

us tr

ees)

mg

l-1

CL Nutrient deficiency for Pinus (>0.2 mg N l-1) and Quercus and Fagus (>0.4 mg N l-1)

Plot Mean Slope RAV 11.3 BRA 11.8 WIJ 9.6 -0.19** GON 11.9 HOE 7.8 -0.50**

Trend (MK)

28

Relation between DON and DIN (2005-2013)

Not significant

0 5 10 15 20 25 30 35 40 45

DIN deposition (kg ha-1 y-1)

0

2

4

6

8

10

12

14

DO

N d

epos

ition

(kg

ha-1

y-1

)

open field precipitationstand precipitation

y = 3.58 + 0.15x

R²=0.7567(*)

y = 1.64 + 0.15xR²=0.3435

0 5 10 15 20 25 30 35 40 450

2

4

6

8

10

12

14

DO

N fl

ux (

kg h

a-1 y

-1)

organic layertopsoilsubsoildeeper mineral soil

y = 1.79 + 0.11x

R²=0.3847

y = 4.69 + 0.09x

R²=0.5421

DIN flux (kg ha-1 y-1)

y = 2.58 + 0.12xR²=0.4392

y = 11.59 - 0.0005xR²=0.00001