ecological restoration at scale timothy l. h. treuer …...of doctor of philosophy recommended for...
TRANSCRIPT
ECOLOGICAL RESTORATION AT SCALE
Timothy L. H. Treuer
A DISSERTATION
PRESENTED TO THE FACULTY
OF PRINCETON UNIVERSITY
IN CANDIDACY FOR THE DEGREE
OF DOCTOR OF PHILOSOPHY
RECOMMENDED FOR ACCEPTANCE BY
THE DEPARTMENT OF
ECOLOGY AND EVOLUTIONARY BIOLOGY
Advisers: David S. Wilcove & Andrew P. Dobson
September 2018
© Copyright by Timothy Lucas Hiebert Treuer, 2018. All rights reserved.
iii
Abstract:
In the age of Biodiversity Hotspots, ecological restoration at a landscape scale often
provides the only avenue to successfully achieving our conservation goals. However, the
resources needed to protect endangered ecosystems are limited. Here, we add four
incremental advances in our understanding of how restoration strategies can be scaled up
to play a meaningful global role in biodiversity conservation without incurring
prohibitive costs. We explore several questions related to large-scale tropical forest
restoration, using Area de Conservacion Guanacaste (ACG), in Costa Rica as a case
study. First, we ask how successful has the largely passive restoration strategy (centered
on wildfire prevention) been for restoring ACG dry forest, and what factors underpin the
variation in forest recovery? Second, can organic waste disposal be harnessed as a
restoration strategy for tropical forests? Third, how does second-growth forest
heterogeneity translate into differential occupancy of regenerating habitat by large
terrestrial vertebrates? Finally, moving beyond tropical forests, we look at how a passive
restoration approach could be deployed in a deep-water marine context, asking where it
would be most effective.
The results of these research projects paint a picture of major opportunities to achieve
significant conservation gains without breaking the proverbial bank. When soil
conditions are favorable, fire management in ACG has been sufficient deliver canopy
closure in less than 30 years. Further, when environmental conditions are not conducive
to forest recovery, application of low-cost organic wastes seems to offer a powerful tool
for promoting passive forest recovery. We find that overall, the value of second-growth
forest for terrestrial vertebrates is high, and all threatened and endangered terrestrial
vertebrates in ACG utilize young forest. Dndangered Baird’s tapirs are more frequently
encountered in second-growth forest habitat than old growth forest. These lessons on the
high but variable value of passive forest restoration methods offers a guide of sorts for
how policymakers could prioritize the passive restoration of other habitats. Turning this
thinking to the high seas reveals that eliminating international fishing in just 25% of its
extent would likely protect more than 50% of the high seas’ total productivity.
iv
ACKNOWLEDGEMENTS
It takes a village to raise a dissertation.
First and foremost, I would like to acknowledge my advisors, David Wilcove and Andy
Dobson for taking me on as their student at Princeton, and providing me with financial,
material, and intellectual support and encouragement over the years. Their flexibility and
feedback made this dissertation possible, and opened so many other doors for me through
my years at Princeton.
I would also like to thank my other committee members, Steve Pacala and Lars Hedin,
for their advice and comments over the years—they greatly enriched the intellectual
depth of my research projects.
I would also like to acknowledge Michael Oppenheimer as my PEI-STEP policy advisor.
Working with him and being part of his research group was one of the greatest treats of
my academic experience and opened my horizons to the policy realm.
Next, I’d like to recognize the enormous contributions made by my many other co-
authors, without whom this tome would be a vastly thinner and sadder document.
Jennifer Powers, Justin Becknell, Leland Werden, Jon Choi (especially Jon Choi), Alex
Gow, Fangyuan Hua, T. Wangyal Shawa, Laura Shanks—thank you!
Dan Janzen and Winnie Hallwachs deserve a special shout out, seeing as, in many ways,
they are the intellectual architects of this entire dissertation. I will forever be in awe of
what they have accomplished. Also, I should acknowledge Rob Pringle here too, both for
effectively being an extra committee member for me and for introducing me to Dan and
Winnie so many years ago.
Thank you to my lab mates who I have not mentioned by name already. Drongos
meetings, in particular, were enormously helpful in charting my course through my
research. Thank you also to my cohort—I don’t know how we all lucked out by having
such an amazing group. Thank you also to my family, including my mom and dad, to
whom I owe everything. And thank you to Cleo Chou—you the real MVP!
I am also grateful for support provided by Roger Blanco, J. Rowlett, G. Funes, B.
Waring, M. M. Chavarria, the Costa Rican Ministry of the Environment, Geraldine
Derroire, Katie O’Malley, Lisa Sheridan, Will Atkinson, and Adrian Benigno Guadamuz
Chavarria. Jacob Socolar could not have been a more helpful sounding board for
questions of statistics and other issues.
Finally, I would like to acknowledge the financial support I received Princeton
Environmental Institute, through the Walbridge Fund, the Garden Club of America award
in Tropical Botany, the High Meadows Foundation and by the National Science
Foundation under grant no. DEB-1501175, the Princeton Department of Ecology and
Evolutionary Biology, The Office of the Dean of the College at Princeton University,
Garden Society of America and Area de Conservacion Guanacaste.
v
TABLE OF CONTENTS
Abstract ............................................................................................................................. iii
Acknowledgements .......................................................................................................... iv
Table of Contents ...............................................................................................................v
Introduction ..................................................................................................................... vii
Chapter 1: Nuclear Spring ................................................................................................1
Abstract ............................................................................................................................2
Introduction ......................................................................................................................4
Methods ............................................................................................................................7
Results ............................................................................................................................13
Discussion ......................................................................................................................23
Works Cited ....................................................................................................................31
Chapter 2: Low-Cost Ag Waste Accelerates Tropical Forest Regeneration .............36
Abstract ..........................................................................................................................37
Introduction ....................................................................................................................39
Methods ..........................................................................................................................44
Results ............................................................................................................................48
Discussion ......................................................................................................................54
Works Cited ....................................................................................................................62
Chapter 3: Organic Wastes and Tropical Forest Restoration .....................................67
Abstract ..........................................................................................................................68
Introduction ....................................................................................................................69
Mitigating arrested or delayed succession .....................................................................70
Promising directions .......................................................................................................72
Climate mitigation ..........................................................................................................74
Works Cited ....................................................................................................................77
Chapter 4: Where the Wild Things Are ........................................................................81
Abstract ..........................................................................................................................82
Introduction ....................................................................................................................84
Methods ..........................................................................................................................87
Results ............................................................................................................................94
Discussion ....................................................................................................................100
vi
Works Cited ..................................................................................................................104
Chapter 5: Efficiency and Expediency of a Partial Closure of High Seas Fisheries109
Abstract ........................................................................................................................110
Introduction ..................................................................................................................112
Methods ........................................................................................................................114
Results ..........................................................................................................................118
Discussion ....................................................................................................................120
Works Cited ..................................................................................................................130
Conclusion ......................................................................................................................134
vii
INTRODUCTION
In the field of ecology, the term ‘global change’ has become a catchall for the enormous
impacts human action have had on ecosystems around the world, from climate change to
wholesale conversion of wildlands. These impacts have become so extensive and
pervasive that geological societies debate whether the planet has entered a new epoch, the
Anthropocene.
While these pressures have generated profound conservation challenges, trends in
globalization, urbanization, and technological advancement, have also created
opportunities to shift the paradigm from ubiquitous habitat loss to restoration at a scale
without modern precedent—at least in some regions.
Further, ecologists now possess a restoration toolkit that has matured to the point that it is
possible to confidently plan and implement meaningful recovery plans for systems as
varied as tropical forests and coral reefs. From a technical standpoint, the potential for
conserving species and bolstering ecosystem services through restoration on a landscape
scale seems unprecedentedly high. However, the challenge of funding such initiatives
looms large.
This dissertation seeks to add several pieces to the puzzle of how human societies should
best utilize their limited resources to resuscitate globally significant expanses of habitat,
both on land and at sea. Like colorful threads in a tapestry, these chapters are diverse in
approach and topic, often running orthogonal in perspective. Yet, they also intersect in a
way that helps weave together a broader picture—in this case a depiction of the ambitious
and actionable frontier of scaled-up ecological restoration in the 21st century across at
least two very different biomes.
While these studies are motived in part by their conservation and policy implications,
Chapters 1, 2, and 4 also represent windows into the processes shaping the assembly of
biological communities. Chapters 1 and 2, in particular, reinforce a decidedly
deterministic view of patterns of ecological succession driven by variations in initial
conditions. Succession of a complex ecosystem like a tropical forest is an experiment in
community assembly, often conducted at a scale that is difficult or unethical to achieve in
a natural setting.
Chapter 1 takes the first of three looks at Área de Conservación Guanacaste (ACG) in
Costa Rica, a protected area that has been called the world’s largest tropical forest
restoration project. The study zeroes in on ACG’s young tropical dry forest, using
established methods within the field of community ecology to begin assessing when and
where passive restoration approaches can be relied on to quickly deliver the return of
viii
diverse pockets of Mesoamerican dry forest, a critically endangered habitat type. Using a
pseudo-experimental approach, the study examines how the biological and functional
diversity of tree assemblages in young dry forests are shaped by a key characteristic of
the former ranchland they occupy, namely the presence of so-called ‘nuclear trees’
dotting the erstwhile pastures. This chapter also underscores the succession-arresting
threat of fire for tropical forest restoration efforts.
Chapter 1 sidesteps an important source of variation in the pace of forest recovery,
however, by only examining the regenerating habitat on one soil formation. If anecdotal
accounts are to be believed, challenging edaphic conditions have greatly stalled forest
recovery elsewhere in ACG, in some cases in conjunction with invasive jaragua grass
(Hyparrhennia rufa). In extreme cases in ACG and around the humid tropics, the
combination of invasive vegetation, soil conditions, and fire can entirely preclude the
success of passive approaches forest recovery, indefinitely arresting succession and
creating what amounts to an alternative stable state for these systems as a tropical
savanna (at least if human triggered fires are treated as an endogenous occurrence).
The second chapter of this dissertation examines a case study of a possible approach for
dealing with such challenging conditions, without resorting to expensive active
restoration approaches like plantation-style reforestation techniques. Indeed, the strategy
examined in Chapter 2 was effectively negative in cost from the perspective of the
conservationists involved. It was not originally implemented as an experimental
restoration treatment at all, but rather as a waste management partnership between an
orange juice company and ACG.
The case study involves the application of 12,000 tonnes of orange waste to a 3 ha corner
of a heavily degraded former pasture just inside the northern edge of ACG. The waste
was the leftover organic material from juice and essential oil extraction processes at a
nearby facility. The study surveys the soil and vegetation characteristics of the site and
adjacent control area sixteen years after the material was applied. The findings of Chapter
2 were published in the journal Restoration Ecology in 2017, and documented the
dramatic impact the agricultural waste had on forest recovery sixteen years later.
Chapter 3 builds on Chapter 2, zooming out to examine the conceptual model of using
agricultural and other organic wastes to accelerate tropical forest restoration. Chapter 3
points out that there are several first principle reasons to expect that these strategies could
be deployed with similar results in other moist tropical settings, despite the Restoration
Ecology study being the only published example of an unprocessed organic waste
achieving a restoration end in a tropical forest setting.
Chapter 4 refocuses on the regenerating dry forests of ACG, this time assessing how
variations in forest structure and composition affect usage of the habitat by terrestrial
vertebrates, including the endangered Baird’s tapir (Tapirus bairdii), the largest extant
native animal in Mesoamerica and an important seed disperser. The study is one of the
first to take into account how multiple axes of second-growth forest heterogeneity might
ix
modulate the value of these regenerating ecosystems for wildlife, while also accounting
for possible differences in the detection probability of species along those axes of floristic
variation.
Chapters 1 to 4 paint a picture of the possibility for hugely successful applications of
cheap and largely passive approaches for ecological restoration of at least one tropical
forest system. Chapter 5 examines how similarly paradigm-bending thinking could result
in the restoration of an entirely different ecosystem. The study assesses a proposal for
ecological restoration for enormous tracts of the high seas, via a cessation of international
fishing efforts, as negotiated through a framework that is game-theoretically tractable.
The study presents a novel analysis of the distribution of productivity through the high
seas to determine which areas should be closed to fishing to achieve optimal results.
This dissertation is neither the beginning nor the end of the ecology and policy research
needed to understand both the theoretical and practical dimensions of scaling up
ecological restoration efforts. However, the hope is that they present a coherent picture of
the potential for harnessing and augmenting the resilience of disturbed natural systems to
leverage limited financial resources for ecological restoration.
1
CHAPTER 1
Nuclear spring: legacy trees, fire history, and landscape composition shape forest
structure, diversity and functional composition in a second-growth dry forest system
Author names and affiliations: Timothy L. H. Treuera, Andrew P. Dobsona, Alexander
Gowa, David S. Wilcovea,b
aPrinceton University, Department of Ecology and Evolutionary Biology, 106a Guyot
Hall, Princeton University, Princeton, NJ, USA 08544
bPrinceton University, Woodrow Wilson School of Public and International Affairs,
Princeton, NJ 08540
2
Abstract
1. Understanding the drivers of secondary tropical forest heterogeneity offers both a
window into the process of community assembly and helps to resolve pressing
conservation questions in a world where second-growth tropical forest has surpassed
primary tropical forest in extent. It has been hypothesized that a subtle difference in
initial conditions, including the presence of a single tree in a field undergoing succession,
can shape functional and species diversity, but such effects have been largely
unexamined empirically.
2. We surveyed trees and saplings in twenty-eight second-growth dry forest sites that
vary in fire history, distance from old growth forest, and proximity to ‘nuclear trees’
(former pasture shade trees) within one of the largest tropical forest restoration initiatives
in the world, Área de Conservación Guanacaste, Costa Rica. Our principal goal was to
assess how these factors shape diversity, functional composition, and structure of the
emerging large tree and sapling assemblages.
3. We found that proximity to nuclear trees had a significant effect on both tree and
sapling diversity, as well as the proportion of trees and saplings that were animal
dispersed species. Over 90% of tree stems >50 m from a nuclear tree were wind-
dispersed species, despite these species making up less than a quarter of the native tree
flora of the region, while just over half of the trees within 10 m of a nuclear tree were
non-anemochorous species. Fire was the only significant predictor of aboveground tree
biomass in plots, while proximity to old growth forest was only a significant predictor of
sapling number and diversity.
3
4. Synthesis: Our findings reveal factors that shape dimensions of second-growth tropical
forest variation, suggesting that the common treatment of second-growth forests as either
uniform or randomly assembled entities distinguished solely by age is problematic.
Further, our findings can inform conservation and restoration initiatives, first
underscoring the power of passive regeneration to deliver landscape-scale forest
restoration, but also highlighting when and where actions such as enrichment planting are
needed to ensure a diverse forest capable of supporting a robust and diverse faunal
community.
Keywords: dry forest; fire; passive restoration; nuclear trees; remnant trees; saplings;
second-growth; seed dispersal
4
INTRODUCTION
Second-growth forests now constitute more than half of all tropical forests, and that
proportion is increasing, yet they are vastly understudied relative to primary forests
(Chazdon 2014). While the study of tropical forest succession has matured greatly, much
of the research on second-growth forest structure, function, and composition entirely
ignores the heterogeneity of these systems, or only considering stand age as a potential
driver of these ecologically fascinating and critically important dimensions of these
systems. While exceptions do exist, the relative rarity of such studies is highlighted in
several reviews related to the conservation value or biodiversity of secondary tropical
forests, which cite dozens of studies comparing secondary and primary forest sites either
implicitly or explicitly assuming the former to be homogenous (Dunn 2004; Chazdon et
al. 2009; Dent & Wright 2009), often without including adequate spatial replicates
(Gardner et al. 2009). Greater attention to potential sources of variation in second-growth
forest structure and functional composition would provide a deeper insight into basic
ecological questions regarding patterns and processes of what the International Union for
the Conservation of Nature have branded the ‘Forests for the 21st Century.’
The lack of information on the causes and consequences of second-growth forest
heterogeneity is also unfortunate because these systems can potentially provide unique
windows into the dynamics of primary forests. Large-scale manipulative experiments in
tropical forests are difficult to carry out, and carefully selected regenerating forests could
serve as pseudo-experimental stand-ins. Understanding community dynamics in
hyperdiverse tropical forests through direct experimental modifications presents high
5
costs and potentially even moral dilemmas. However, large second-growth systems, such
as Área de Conservación Guanacaste (ACG) in northwest Costa Rica, provide a means of
examining many basic ecological processes: plant-soil feedbacks (e.g. Batterman et al.
2013), ecological filtering (e.g. Lebrija-Trejos et al. 2010), facilitation (e.g. Guevara &
Laborde 1993), and the relative roles of stochastic and deterministic forces in shaping
community structure and composition. Utilizing second-growth forests to better
understand primary forests demands closer examination of the causes and consequences
of second-growth forest heterogeneity. Several potential deterministic drivers of tropical
forest structure and composition, including edaphic conditions, site history, and
landscape composition, have been a subject of considerable debate in the ecological
literature. Examining whether or how they shape ecological function and community
dynamics in second-growth forests offers greater clarity into their role as potential
deterministic drivers of mature forest communities, in addition to being important
questions in their own right.
Moreover, unpacking the causes of variation in second-growth forest structure
and functional composition could help to answer pressing applied questions, for example,
when and where should passive regeneration be deployed as a tool within broader
conservation strategies, particularly relative to investments in active restoration or intact
habitat acquisition (Possingham, Bode & Klein 2015)? This is particularly topical in light
of recent claims that passive forest regeneration can deliver at least as many positive
outcomes as active forest restoration initiatives (Crouzeilles et al. 2017). Additionally,
variations in forest structure are directly tied to the ability of these forests to deliver a
critically important ecosystem service: carbon storage. Developing a better understanding
6
of what shapes the carbon sequestration capacity of young tropical forests is key to
understanding both the carbon cycle in the Anthropocene and whether actions beyond
passive restoration are needed for forest regeneration to serve as a negative-emissions
technology (Treuer et al. 2017). Large-scale implementation of active and passive forest
restoration will be essential for holding global climate change under 2 degrees Celsius
(Gasser et al. 2015; Boysen et al. 2017).
Studies of secondary tropical dry forests are particularly needed, as most studies
of second-growth tropical forest heterogeneity have focused on mesic systems (but see
Powers et al. 2009; Becknell & Powers 2014). Moreover, tropical dry forests have
suffered a much greater conversion rate to other land-uses as well as a greater percentage
of remaining habitat existing in a non-primary state compared to wetter tropical forests
(Portillo-Quintero & Sánchez-Azofeifa 2010; Sloan et al. 2014).
The role of remnant vegetation in shaping successional trajectories and driving
community composition offers an especially interesting model of the effect of
environmental variation on community heterogeneity. The presence of even just a single
tree in a landscape undergoing secondary succession after prolonged agricultural or
pastoral use might seem to be trivial, but it has been hypothesized to have the capacity to
radically alter community structure and composition by drawing in seed-dispersing
animals (Guevara, Purata & Van der Maarel 1986; Janzen 1988a). The effect of remnant
or pasture-grown trees on seed rain and sapling composition in actively managed
agricultural land and old fields has been confirmed (Guevara et al. 1992; Guevara &
Laborde 1993; Toh, Gillespie & Lamb 1999; Slocum & Horvitz 2000; Derroire et al.
2016), and the evidence suggests these ‘nuclear trees’ play a strong role in fostering
7
pockets of second-growth forest that differ in structure or composition from surrounding
regenerating forest. To date, however, studies of this effect in advanced regeneration
have been typically been either indirect (Castillo et al. 2012) or anecdotal (Janzen
1988b). Just two studies have empirically examined the downstream effect of remnant
trees on forest diversity and community composition in closed-canopy second-growth
forests, and they both come from moist forest systems in Costa Rica (Schlawin & Zahawi
2008; Sandor & Chazdon 2014). The role of nuclear trees in shaping dimensions of
functional diversity in dry forests seems particularly important to conservation; in their
absence, these forests are thought to be dominated by the small number of species with
wind-mediated dispersal strategies (Janzen 1988b) that offer negligible resources to
frugivores (Sorensen & Fedigan 2000).
In this study, we explore how three different environmental factors operate
together to shape forest structure, species and functional diversity in ways relevant to
faunal assemblages in a regenerating tropical dry forest system in northwestern Costa
Rica. Specifically, we explore three questions regarding ACG’s extensive 30-year old
regenerating forests: (1) how do species richness, diversity, and functional diversity (with
respect to dispersal syndromes) of trees and saplings relate to the presence of nuclear
trees, fire history, and proximity to old growth forest? (2) Does proximity to nuclear trees
increase the odds that nearby trees and saplings will have animal-linked dispersal
syndromes in a way that could explain the richness and diversity disparities in Hypothesis
1? (3) How do aboveground biomass and density of saplings relate to nuclear trees, fire
history, and landscape composition?
8
MATERIALS AND METHODS
This study was conducted in Parque Nacional Santa Rosa (SRNP) (10°48′53″N,
85°36′54″W), a constituent Sector of Área de Conservación Guanacaste (ACG) in
northwest Costa Rica. SRNP has an unusual history for a protected area. When it was
expanded from a 1,000 ha national monument in 1971, it is thought to have been actively
managed as a cattle-producing hacienda since the 1500s (Allen 2001). The mosaic
pattern of ranches in the region, in particular the persistence of pockets of old-growth
forest, maintained a high diversity of flora and fauna on a landscape scale (Janzen 2000).
Cattle were fully removed from SRNP by 1978 (Janzen 1988b), though a small number
of free-ranging horses were present until the early 1990s (Janzen 2000). Fire suppression
in SRNP began in earnest in the mid-1980s, but occasional fires spilling over from
neighboring ranches combined with human-ignited fires within the park have resulted in
variable fire history for many areas of regenerating forest in SRNP. Time since most
recent fire for each site in our study was determined by consulting meticulous Santa Rosa
park records (W. Medina Sandoval pers. Comm.). We focused our study on an area that
has previously been classified as ‘Santa Rosa Tropical Dry Forest,’ characterized by
mesic soil conditions (Becknell & Powers 2014) that were covered in pasture when fire
management and expansion of ACG began in the mid-1980s (Janzen 1988c). The climate
of this system is strongly seasonal, with a windy dry season lasting from approximately
December to the middle of May during which time little if any of the 900 to 2600 mm of
annual rainfall (30 year average of 1765 mm) falls (Arroyo-Mora et al. 2005; Becknell &
Powers 2014). Winds during the dry season blow predominantly out of the northeast
(Castillo-Núñez et al. 2011).
9
Study Design
We established a network of twenty-eight plots measuring 50 m by 20 m (0.1 ha)
in size. The plot dimensions were chosen to correspond to other vegetation survey plots
within ACG and the region (Powers et al. 2009; Becknell & Powers 2014). Within each
plot we measured the diameter at breast height (DBH) of all large trees (DBH > 10 cm)
and height using a laser range-finder aimed at the highest foliage of each tree (Scout 1000
ARC from Bushnell Outdoor Products). The DBH size threshold for trees was chosen
both because it corresponded to approaches taken in the literature (Powers et al. 2009)
and because it was the approximate boundary between understory and canopy forming
trees in this second-growth system. Within a 6 m by 50 m strip through the center of each
plot we also measured the DBH and heights of all trees between 1 cm DBH and 10 cm
DBH with a height of at least 1.3 m, again following the approach of Powers et al.
(2009). Smaller woody species that veteran ‘parataxonomists’ (Janzen & Hallwachs
2011) have never previously encountered with DBH above 10 cm within our study region
were excluded from the sapling dataset on the assumption that these species would not
mature into canopy trees. We recorded a GPS point for all trees and saplings in our
dataset as well as plot centroids using a Garmin GPSMAP 62 device. The device’s live
position accuracy estimation during use, rediscovery of marked trees, and post hoc
examination of trees known to form the borders of plots indicate that true positions rarely
were more than 5m from GPS points.
The plots were sited with the aid of aerial imagery from 1988 (Fig. 1.1a). Images
covering the study area were georeferenced by pinning the images to a modern, satellite
10
imagery-derived basemap in ArcGIS (ESRI 2011). We then identified individual large
pasture trees that were sufficiently isolated from old-growth forest and other pasture trees
to permit establishment of a plot in the adjacent terrain in such a way that the plot
centroid was at least 50 m from forest fragments. Between June and August 2014 the
locations of these trees were visited in the forests of ACG, and the trees verified based on
their size, distinctive growth form (Janzen 1988a), and position on the landscape relative
to landmarks such as roads and former stone corrals that were visible in the aerial images.
When a nuclear tree could not be identified within 20 m of the GPS point derived from
the georeferenced imagery, or it appeared that the nuclear tree had died and more than
half of its limbs had fallen to the ground, the location was rejected for the siting of a plot.
Potential nuclear tree plot sites were prioritized for investigation in a way that maximized
their geographic distribution throughout areas that had been characterized as Santa Rosa
Tropical Dry Forest (Becknell & Powers 2014), while minimizing their distance to road
and trail networks. This method was pursued to negotiate the trade off between capturing
landscape scale variation and time and resource constraints.
A plot was established next to the first fourteen nuclear trees found using the
above methods. They were established next to the nuclear tree such that the trunk of the
nuclear tree was just touching the midpoint of one of the 20 m sides of the plot. These
fourteen nuclear tree plots were then matched with fourteen additional plots that were
sited by selecting areas in the aerial images as close as possible to each nuclear tree plot
while maintaining a minimum distance of 50m from nuclear trees and other plots. All tree
species identities were determined with the help of an expert parataxonomist (Janzen &
Hallwachs 2011).
11
Biomass Calculations
Estimated aboveground biomass (ABGest) was calculated using measured
diameters and heights, and literature-derived values for wood density (Zanne et al. 2009;
Chave et al. 2009), and were calculated according to the following allometric equation
for tropical dry forests (Chave et al. 2005):
⟨AGB⟩est= 0.112×(ρD2H)0.916
Where ρ is wood density, D is diameter, and H is the height of each tree.
Landscape Composition Metrics
We ascertained the distances between individual trees and nuclear trees using
GPS points for each tree in ArcGIS (ESRI 2011). Three metrics for proximity of old-
growth forest were collected: distance to nearest old-growth, distance to the west of old-
growth forest, and percentage of old-growth forest within a 500 m radius of the plot
centroid. These three landscape composition metrics were selected because they have
been hypothesized in the literature to predict either seed rain, functional composition, or
biomass increment of second-growth tropical dry forest (Janzen 1988a). Distance to the
west of old-growth forest was hypothesized to be important because all wind-dispersed
tree species in the region release their seeds during the dry season when typically blow
from east to west. Prior to analysis we tested the three landscape variables for
collinearity, and found no problematic collinearity, which we considered to be r > 0.7.
The correlation between each pair of variables was r < 0.5. The distance metrics were
calculated by creating a vector layer of polygons within ArcGIS that covered the visible
12
forest in the 1988 aerial images (Fig. 1.1b) and calculating the metrics with reference to
plot centroids.
(a)
(b)
13
Fig. 1.1 (a) Georeferenced aerial imagery of the study system from 1988 with observation
plots indicated by clusters of purple dots; (b) a close-up of a georeferenced 1988 aerial
image with green GPS points for trees surveyed in 2015, red boxes indicated boundaries
of 20 m by 50 m plots, and a pale green-blue polygon layer obscuring old growth forest
fragments.
Life History Characteristics
We compiled dispersal syndrome information for each plant species from a collection of
online sources and personal communication with regional botanists.
14
RESULTS
Tree and Sapling Richness and Diversity
We measured a total of 1051 trees and 1012 saplings, representing 57 and 69
different species respectively. 55 of the 57 species (96.5%) found as large trees were
observed in plots adjacent to nuclear trees while only 30 (52.6%) were found in plots not
adjacent to nuclear trees. The mean number of tree species for plots adjacent to nuclear
trees was 10.5 and for plots not adjacent to nuclear trees was 7.5, indicating β-diversities
of 5.2 and 4 respectively under Whittaker’s original definition (β = γ/α). Non-overlapping
95% confidence intervals in rarefaction analysis indicating the total number of species in
plots adjacent to nuclear trees was significantly greater than in plots that were not
adjacent to nuclear trees (Fig. 1.2a).
For saplings, we found 62 of the 69 (89.9%) species in nuclear tree-adjacent plots,
and 46 of 69 species (66.7%) in non-adjacent plots. The mean number of tree species
present as saplings within each plot were 13.1 and 9.3 respectively, indicating β-
diversities of 4.7 and 4.9 respectively. Rarefaction analysis revealed non-overlapping
95% confidence intervals, which we interpreted as a statistically significant difference in
the number of sapling species in plots adjacent to nuclear trees and those not adjacent to
nuclear trees (Fig. 1.3a).
We used Poisson-family generalized mixed-effects models run in R over all
combinations of explanatory variables (excluding those with multiple old-growth forest
proximity variables), and used model averaging to determine the relative importance of
different environmental variables in shaping plot-level heterogeneity. Time since fire was
strongly correlated with plot-level species richness for both trees (Fig. 1.2c) and saplings
15
(Fig. 1.3c). In the conditional-averaged model, time since fire was the only statistically
significant in explaining variation in plot-level species richness of trees (z = 3.836, p <
0.01), and it appeared in all component models given weight. For plot-level sapling
richness, time since fire had a similar relationship in the conditional-averaged model (z =
3.183, p < 0.01) and occurred in all models given weight. Unlike with plot-level tree
richness, however, nuclear tree-adjacency also was statistically significant (z = 2.261, p <
0.05) in the conditional-average model in explaining plot-level sapling richness (Fig.
1.3b) and occurred in the top three candidate models, which accounted for 65% of total
model weight.
(a)
16
(b)
17
(c)
Fig. 1.2 Species richness of saplings (tree species <10 cm DBH) in regenerating forest
visualized by (a) plotting individual-based rarefaction and extrapolation curves for trees
found in plots adjacent to nuclear trees and those not adjacent to nuclear trees (‘Control
Plots’), (b) in a violin plots (with embedded box plots) mapping plot-level species
richness by type of plot, and (c) in a scatterplot of time since most recent fire against
species richness
18
(a)
19
(b)
20
(c)
Fig. 1.3 Species richness of saplings (tree species <10 cm DBH) in regenerating forest
visualized by (a) plotting individual-based rarefaction and extrapolation curves for
saplings found in plots adjacent to nuclear trees and those not adjacent to nuclear trees
(‘Control Plots’), (b) in a violin plots (with embedded box plots) mapping plot-level
species richness by type of plot, and (c) in a scatterplot of time since most recent fire
against species richness.
Functional Diversity
In plots adjacent to nuclear trees, 32% of large trees and 47% of saplings were
non-anemochorous species (not dispersed by wind), while for plots not adjacent to
nuclear trees those values were 8% and 38% respectively.
We used binomial generalized linear models with a logit link function to examine
whether and how quickly the effect of nuclear trees of on functional composition (in
21
terms of dispersal strategy) of the surrounding forest attenuated. We regressed distance
from nearest nuclear tree against the probability of an individual tree or sapling having a
non-anemochorous dispersal strategy for data from plots adjacent to nuclear trees. We
found that there was a significant effect for both trees (p < 0.01; Fig. 1.4a) as well as for
saplings (p < 0.05; Fig. 1.4b). The proportion of wind dispersed individuals for both trees
and saplings reached the proportion of wind-dispersed trees observed in non-nuclear tree
plots within 50 m of the nuclear trees.
(a)
22
(b)
Fig. 1.4 Plot of a binomial dummy variable for dispersal syndromes (0 = anemochorous,
1 = all other dispersal modes), plotted against the distance the individual is from a nuclear
tree for (a) trees (>10 cm DBH) and (b) saplings (<10 cm DBH). Dark line represents
generalized linear model best fit for the data, and the gray areas bounded by dashed lines
are the 95% confidence intervals. Horizontal bars indicate proportion non-anemochorous
species in plots not adjacent to nuclear trees.
Aboveground Biomass and Sapling Density
The top candidate model for aboveground biomass included only time since fire
(Fig. 1.5) and was given 34% of model weight. Time since fire occurred in all the models
given weight and was the only statistically significant variable (z = 2.879, p < 0.01) in the
conditional-averaged model. We compared the AGB values from our plots to published
models (Poorter et al. 2016) of forest biomass as a function of age at Neotropical sites.
Using time since fire as a substitute for stand age, and found that our data was consistent
23
with range of values of the data used to calibrate the model for Neotropical secondary
forests biomass accumulation with forest age.
We found that that both distance to nearest old-growth forest fragment and time
since fire, as well as their interaction predicted variation in the number of saplings in each
plot (Fig. 1.6). The candidate model with just these three terms was given half the weight
in the averaged model, and the second best model was 2 AIC units lower than the top
candidate model. All candidate models given weight included all three terms.
Fig. 1.5 Aboveground biomass of trees (>10 cm DBH) in forest patches plotted against
time since the most recent fire.
24
DISCUSSION
Drivers of Community Composition
Second-growth dry forest in Santa Rosa displayed high degrees of heterogeneity
in tree and sapling biomass, species richness, and functional composition, belying the
frequent treatment of regenerating forests as homogenous. Fire has long been understood
to be both a pasture management tool and a natural force maintaining savannahs in humid
biomes, and not surprising then, time since most recent fire was the sole significant
explanatory variable for estimated forest biomass. Fire history was also an important
correlate of sapling density and species richness of both tree and sapling assemblages.
Aside from fire, Janzen (1988a) hypothesized two dominant forest regeneration
pathways that underpinned the heterogeneity of recovering dry forests of ACG, based
largely on anecdotal observations of very young regeneration. Thirty years later, our data
partially confirm that both of these pathways continue to have a legacy effects on what
are now intermediate aged forests. First, wind-dispersed species constitute the large
majority of trees in these forests despite the rarity of this dispersal syndrome in old-
growth forests. On this basis alone, we can firmly reject a purely neutral model for dry
forest community assembly. Even within the sapling assemblages, wind-dispersed
species are strongly over-represented. However, the directional distance effect that
Janzen observed seems to have largely disappeared with time. We found no effect of
distance downwind on either forest biomass, sapling number, nor any of our community
composition variables, though none of our plots were more than 500 m downwind of an
old-growth forest patch.
25
In regards to Janzen’s second hypothesized regeneration pathway, our data show a
continuing signal of the nuclear tree effect on species composition in 30 year old
regenerating forest system, a first for any forest system. To our knowledge these data are
also the first to show a continued nuclear tree impact on the sapling class within a closed
canopy forest systems of any age. The nucleation effect, however, seemed to dissipate
within a relatively small radius of the nuclear trees. According to our statistical model,
while roughly 60% of the trees within 10 m of a nuclear tree are non-wind dispersed
species, at a distance of 40-50 m from a nuclear tree less than 20% of trees are non-wind
dispersed species. Our sample size of nuclear trees and haphazard discovery of nuclear
tree species identity prevented us from ascertaining whether the species identity or
functional traits of the nuclear tree affect composition of surrounding tree and sapling
assemblages, though these variables have been found to be important in sapling
recruitment under pasture trees in the region (Derroire et al. 2016).
26
Fig. 1.6 Number of saplings (tree species <10 cm DBH) in forest patches plotted against
the distance to the nearest old growth forest patch. Points are scaled by the time since the
most recent fire
Implications for Conservation
The results of this study carry a number of implications for the conservation of
tropical dry forest, one of the most heavily converted biomes in the species-rich tropics,
as well as for climate mitigation efforts. The first is simply that when fires can be
controlled, passive restoration efforts can result in a rapid return of carbon-sequestering
forest and shading out of fire-prone invasive grasses. Given that more passive approaches
typically are far cheaper than more labor-intensive active restoration approaches based on
planting seedlings or direct seeding, the former may offer a better means of leveraging
limited conservation resources for the restoration of dry forests with mesic soil
conditions, at least when return of forest structure or carbon sequestration is the primary
27
goal. Chao estimates of total sapling species richness in regenerating forest in our study
system are account for more than half of the total number of trees and treelet species
present in the dry forests of lowland Guanacaste (Janzen & Liesner 1980), at least after
species believed to be riparian specialists are not considered. This observation is a source
of cautious optimism that passive approaches will eventually result in forests with a
diversity approaching that of extant old-growth.
This optimism is in part predicated on the presence of nuclear trees, which we
found play an outsize (albeit highly localized) role as recruiting foci for animal-dispersed
tree species. If a rapid return of floral alpha diversity or functional diversity (especially
in terms of dispersal syndrome diversity) is the primary restoration objective, our data
suggest some method of increasing the number of non-anemochorous species may be
required (e.g. enrichment planting or seeding or introduction of artificial perch sites for
volant seed dispersers) in areas where the density of nuclear trees falls below 1 per
hectare.
Furthermore, the clear importance of nuclear trees for increasing beta diversity
and functional diversity of regenerating forest trees and saplings in Santa Rosa suggests a
policy prescription for ensuring young forests are maximally diverse: offer financial
incentives or establish regulations mandating a minimum density of shade trees be
maintained in actively managed cattle pastures. Low densities of trees within pastures in
Costa Rica and elsewhere in the Neotropics have not been found to decrease livestock
yields, and indeed can increase pasture health in some cases when pasture trees are
nitrogen fixing species (Apolinário et al. 2015). In Costa Rica and elsewhere, long-term
degradation of soils on pastures has traditionally resulted in the eventual shifting of
28
pastures and fallowing of land (hence the mosaic pattern of the agroscape in much of
seasonally-dry Mesoamerica). Our data suggest that ensuring that nuclear trees are
scattered across the landscape would increase the conservation value of young forest in
the years following a fallow or abandonment, which may be particularly important for the
overall conservation value of these mosaic landscapes since young forests are often re-
cleared (Schwartz et al. 2017) before maturing to the point that they can provide valuable
habitat to species of conservation concern.
Increasing the abundance of animal-dispersed species in young forest also likely
means higher densities of fruit resources for frugivorous species, which should positively
reinforce seed rain of animal dispersed species in the future. Indeed, in the absence of
nuclear trees, second-growth dry forests will become dominated by wind-dispersed
species, which may explain previous work suggesting that all three of Santa Rosa’s
primate species (Alouatta palliata, Ateles geoffroyi, and Cebus capucinus) have lower
densities in regenerating forest compared to old-growth forest (Sorensen & Fedigan
2000). Plainly, further investigation is needed to draw direct links between the functional
composition of second-growth forest and the presence or density of frugivores.
Limitations and Future Directions
Our findings demonstrate both the existence of second-growth forest
heterogeneity in structure and composition, as well as several likely sources of that
heterogeneity, but we are limited in our ability to confidently state that all unexplained
variation in our dataset is the result of stochastic processes. While we attempted to
control for edaphic conditions by limiting samples to one soil formation, we did not
29
rigorously characterize the soil traits in our plots, which have been shown to be important
for floral composition in tropical dry forests (Becknell & Powers 2014). Likewise, since
fires seemed to almost completely eliminate living aboveground woody biomass, it is
impossible to tease apart the influence of fire per se, from climatic variation during the
early years of forest regeneration.
Given the complex multi-causality of the ecological processes we sought to
investigate, we are cautious about applying our findings to other study systems that
experience environmental variation beyond what is experienced in Santa Rosa. In
particular, we would feel remiss if we failed to point out that even within ACG, rainfall
gradients and soil conditions can independently retard forest recovery. Forest cover has
been slow to return to the Santa Elena peninsula, which are characterized by serpentine
soils, even in unburned areas. This is also the case in the heavily degraded soils in the
transitional wet-dry forests in the north of the park (Treuer et al 2017) and areas
underlain by vertisol soils in Sector Horizontes (Werden et al 2017). Conservationists
found it necessary to plant seedlings into cattle pastures in the wetter sectors of the
conservation area following little spontaneous recruitment of woody species (Janzen
2000).
There may also be additional subtleties to the processes we identify with our data
that we are unable to tease out, particularly with regards to the nuclear tree effect. Recent
research examining saplings growing beneath adult trees in actively managed cattle
pastures near Santa Rosa has suggested the deterministic properties of the nuclear tree
effect may vary depending on the functional traits of the nuclear tree (Derroire et al
2016). Unfortunately, our ability to examine the downstream effects of nuclear tree
30
species identity on intermediate-aged forest is not possible given our limited sample size
and the haphazard way in which nuclear trees were selected for inclusion in this study (it
was not possible to determine species identity of nuclear trees from the 1988 aerial
imagery). Future research could take advantage of our finding of a highly localized
nucleation effect to justify small vegetation plots and sampling of nuclear trees that have
greater proximity to each other.
Finally, this study examines only woody plant assemblages. Further work is
needed to address how variation in the floral composition of second-growth forests
affects the heterogeneity of faunal communities. This is particularly important from an
applied ecology perspective. More research on drivers of faunal heterogeneity in second-
growth forests is needed to determine how and when forest restoration should be
optimally deployed as a conservation tool for threatened and endangered tropical dry
forest fauna.
31
WORKS CITED
Allen, W. (2001) Green Phoenix: Restoring the Tropical Forests of Guanacaste, Costa
Rica. Oxford University Press, USA.
Apolinário, V.X.O., Dubeux, J.C.B., Lira, M.A., Ferreira, R.L.C., Mello, A.C.L., Santos,
M.V.F., Sampaio, E.V.S.B. & Muir, J.P. (2015) Tree Legumes Provide Marketable
Wood and Add Nitrogen in Warm-Climate Silvopasture Systems. Agronomy
Journal, 107, 1915.
Arroyo-Mora, J.P., Sanchez-Azofeifa, G.A., Kalacska, M.E.R., Rivard, B., Calvo-
Alvarado, J.C. & Janzen, D.H. (2005) Secondary Forest Detection in a Neotropical
Dry Forest Landscape Using Landsat 7 ETM+ and IKONOS Imagery1. Biotropica,
37, 497–507.
Batterman, S.A., Hedin, L.O., van Breugel, M., Ransijn, J., Craven, D.J. & Hall, J.S.
(2013) Key role of symbiotic dinitrogen fixation in tropical forest secondary
succession. Nature, 502, 224–7.
Becknell, J.M. & Powers, J.S. (2014) Stand age and soils as drivers of plant functional
traits and aboveground biomass in secondary tropical dry forest. Canadian Journal
of Forestry Research.
Boysen, L.R., Lucht, W., Gerten, D., Heck, V., Lenton, T.M. & Schellnhuber, H.J. (2017)
The limits to global-warming mitigation by terrestrial carbon removal. Earth’s
Future, 5, 463–474.
Castillo, M., Rivard, B., Sánchez-Azofeifa, A., Calvo-Alvarado, J. & Dubayah, R. (2012)
LIDAR remote sensing for secondary Tropical Dry Forest identification. Remote
Sensing of Environment, 121, 132–143.
32
Chave, J., Andalo, C., Brown, S., Cairns, M.A., Chambers, ... & Yamakura, T. (2005)
Tree allometry and improved estimation of carbon stocks and balance in tropical
forests. Oecologia, 145, 87–99.
Chave, J., Coomes, D., Jansen, S., Lewis, S.L., Swenson, N.G. & Zanne, A.E. (2009)
Towards a worldwide wood economics spectrum. Ecology Letters, 12, 351–366.
Chazdon, R.L. (2014) Second Growth. University of Chicago Press.
Chazdon, R.L., Peres, C.A., Dent, D., Sheil, D., Lugo, A.E., Lamb, D., Stork, N.E. &
Miller, S.E. (2009) The potential for species conservation in tropical secondary
forests. Conservation biology : the journal of the Society for Conservation Biology,
23, 1406–17.
Crouzeilles, R., Ferreira, M.S., Chazdon, R.L., Lindenmayer, D.B., Sansevero, J.B.B.,
Monteiro, L., Iribarrem, A., Latawiec, A.E. & Strassburg, B.B.N. (2017) Ecological
restoration success is higher for natural regeneration than for active restoration in
tropical forests. Science Advances, 3, e1701345.
Dent, D.H. & Joseph Wright, S. (2009) The future of tropical species in secondary
forests: A quantitative review. Biological Conservation, 142, 2833–2843.
Derroire, G., Coe, R., Healey, J.R. & Wulf, M. (2016) Isolated trees as nuclei of
regeneration in tropical pastures: testing the importance of niche-based and
landscape factors. Journal of Vegetation Science, 27.
Dunn, R.R. (2004) Recovery of faunal communities during tropical forest regeneration.
Conservation Biology, 18, 302–309.
ESRI. (2011) ArcGIS Desktop.
Gardner, T.A., Barlow, J., Chazdon, R., Ewers, R.M., Harvey, C.A., Peres, C.A. & Sodhi,
33
N.S. (2009) Prospects for tropical forest biodiversity in a human-modified world.
Ecology Letters, 12, 561–582.
Gasser, T., Guivarch, C., Tachiiri, K., Jones, C.D. & Ciais, P. (2015) Negative emissions
physically needed to keep global warming below 2 °C. Nature Communications, 6,
7958.
Guevara, S. & Laborde, J. (1993) Monitoring seed dispersal at isolated standing trees in
tropical pastures: consequences for local species availability. Frugivory and seed
dispersal: ecological and evolutionary aspects (eds T.H. Fleming & A. Estrada), p.
Springer Netherlands, Dordrecht.
Guevara, S., Meave, J., Moreno-Casasola, P. & Laborde, J. (1992) Floristic composition
and structure of vegetation under isolated trees in neotropical pastures. Journal of
Vegetation Science, 3, 655–664.
Guevara, S., Purata, S.E. & Van der Maarel, E. (1986) The role of remnant forest trees in
tropical secondary succession. Vegetatio, 66, 77–84.
Janzen, D.J. (1988a) Guanacaste National Park: tropical ecological and biocultural
restoration. Rehabilitating Damaged Ecosystems, Vol II (ed J. Cairns), pp. 143–192.
CRC Press, Boca Raton, FL.
Janzen, D.H. (1988b) Management of Habitat Fragments in a Tropical Dry Forest:
Growth. Annals of the Missouri Botanical Garden, 75, 105.
Janzen, D.H. (1988c) Management of habitat fragments in a tropical dry forest: growth.
Annals of the Missouri botanical garden, 75, 105–116.
Janzen, D.H. (2000) Costa Rica’s Area de Conservación Guanacaste: A long march to
survival through non-damaging biodevelopment. Biodiversity, 1, 7–20.
34
Janzen, D.H. & Hallwachs, W. (2011) Joining inventory by parataxonomists with DNA
barcoding of a large complex tropical conserved wildland in northwestern costa rica.
PloS one, 6, e18123.
Janzen, D.H. & Liesner, R. (1980) ANNOTATED CHECK - LIST OF PLANTS OF
LOWLAND GUANACASTE PROVINCE , COSTA RICA , EXCLUSIVE OF
GRASSES AND NON - VASCULAR CRYPTOGAMS. , 18, 15–90.
Lebrija-Trejos, E., Pérez-García, E.A., Meave, J.A., Bongers, F. & Poorter, L. (2010)
Functional traits and environmental filtering drive community assembly in a species-
rich tropical system. Ecology, 91, 386–398.
Poorter, L., Bongers, F., Aide, T.M., Almeyda Zambrano, A.M., Balvanera, P., Becknell,
… & Rozendaal, D.M.A. (2016) Biomass resilience of Neotropical secondary
forests. Nature, 530, 211–214.
Portillo-Quintero, C.A. & Sánchez-Azofeifa, G.A. (2010) Extent and conservation of
tropical dry forests in the Americas. Biological Conservation, 143, 144–155.
Possingham, H.P., Bode, M. & Klein, C.J. (2015) Optimal conservation outcomes require
both restoration and protection. PLoS biology, 13, e1002052.
Powers, J.S., Becknell, J.M., Irving, J. & Pèrez-Aviles, D. (2009) Diversity and structure
of regenerating tropical dry forests in Costa Rica: Geographic patterns and
environmental drivers. Forest Ecology and Management, 258, 959–970.
Sandor, M.E. & Chazdon, R.L. (2014) Remnant Trees Affect Species Composition but
Not Structure of Tropical Second-Growth Forest (ed N Farwig). PLoS ONE, 9,
e83284.
Schlawin, J.R. & Zahawi, R.A. (2008) “Nucleating” succession in recovering neotropical
35
wet forests: The legacy of remnant trees. Journal of Vegetation Science, 19, 485–
492.
Schwartz, N.B., Uriarte, M., DeFries, R., Gutierrez-Velez, V.H. & Pinedo-Vasquez, M.A.
(2017) Land-use dynamics influence estimates of carbon sequestration potential in
tropical second-growth forest. Environmental Research Letters, 12, 74023.
Sloan, S., Jenkins, C.N., Joppa, L.N., Gaveau, D.L.A. & Laurance, W.F. (2014)
Remaining natural vegetation in the global biodiversity hotspots. Biological
Conservation, 177, 12–24.
Slocum, M.G. & Horvitz, C.C. (2000) Seed arrival under different genera of trees in a
neotropical pasture. Plant Ecology, 149, 51–62.
Sorensen, T.C. & Fedigan, L.M. (2000) Distribution of three monkey species along a
gradient of regenerating tropical dry forest. Biological Conservation, 92, 227–240.
Toh, I., Gillespie, M. & Lamb, D. (1999) The Role of Isolated Trees in Facilitating Tree
Seedling Recruitment at a Degraded Sub-Tropical Rainforest Site. Restoration
Ecology, 7, 288–297.
Treuer, T.L.H., Choi, J.J., Janzen, D.H., Hallwachs, W., Peréz-Aviles, D., Dobson, A.P.,
Powers, J.S., Shanks, L.C., Werden, L.K. & Wilcove, D.S. (2017) Low-cost
agricultural waste accelerates tropical forest regeneration. Restoration Ecology.
Zanne, A.E., Lopez-Gonzalez, G., Coomes, D.A., Ilic, J., Jansen, S., Lewis, S.L., Miller,
R.B., Swenson, N.G., Wiemann, M.C. & Chave, J. (2009) Data from: Towards a
worldwide wood economics spectrum.
36
CHAPTER 2
Low-Cost Agricultural Waste Accelerates Tropical Forest Regeneration
Author names and affiliations:
Timothy L. H. Treuera, Jonathan J. Choia, Daniel H. Janzenb, Winnie Hallwachsb,
Daniel Perez Avilesc, Andrew P. Dobsona, Jennifer S. Powersc, Laura C. Shanksd, Leland
K. Werdene, David S. Wilcovea,f
aPrinceton University, Department of Ecology and Evolutionary Biology, 106a Guyot
Hall, Princeton University, Princeton, NJ, USA 08544
bUniversity of Pennsylvania, Department of Biology, 433 S University Ave, Philadelphia,
PA 19104
cUniversity of Minnesota, College of Biological Sciences, 1987 Upper Buford Cir, St.
Paul, MN 55108
dBeloit College, 700 College St. Beloit, WI 53511
eUniversity of Minnesota, Program in Plant Biological Sciences, 1445 Gortner Avenue,
250 Biological Sciences, Saint Paul, MN 55108
fPrinceton University, Woodrow Wilson School of Public and International Affairs,
Princeton, NJ 08540
37
Abstract
Lower-cost tropical forest restoration methods, particularly those framed as win-win
business-protected area partnerships, could dramatically increase the scale of tropical
forest restoration activities, thereby providing a variety of societal and ecosystem
benefits, including slowing both global biodiversity loss and climate change. Here we
describe the long-term regenerative effects of a direct application of agricultural waste on
tropical dry forest. In 1998, as part of an innovative agricultural waste disposal service
contract, an estimated 12,000 Mg of processed orange peels and pulp were applied to a 3
ha portion of a former cattle pasture with compacted, rocky, nutrient-poor soils
characteristic of prolonged fire-based land management and overgrazing in Área de
Conservación Guanacaste, NW Costa Rica. After 16 years, the experimental plot showed
a threefold increase in woody plant species richness, a tripling of tree species evenness
(Shannon Index), and a 176% increase in aboveground woody biomass over an adjacent
control plot. Hemispheric photography showed significant increases in canopy closure in
the area where orange waste was applied relative to control. Orange waste deposition
significantly elevated levels of soil macronutrients and important micronutrients in
samples taken two years and sixteen years after initial orange waste application. Our
results point to promising opportunities for valuable synergisms between agricultural
waste disposal and tropical forest restoration and carbon sequestration.
38
Implications for Practice:
● Agroindustry and other sectors in the tropics often produce large quantities of
nutrient rich byproducts or waste streams, which in some cases require high net-
cost disposal or processing.
● In countries with strong waste disposal laws, cost-negative ‘win-win’ restoration
projects through creative partnerships between the private sector and restoration
ecologists can be achieved.
● Such initiatives also potentially result in significantly accelerated carbon
sequestration.
● Documentation of the history, as well as restoration and carbon benefits of one
such initiative using orange waste are provided as a model for future
undertakings.
● While aggressive safeguards should be taken to mitigate unintended
consequences, at least in the case discussed here, concerns over negative
environmental impacts associated with agricultural waste use proved unfounded.
Keywords: Area de Conservacion Guanacaste; carbon sequestration; Citrus; Costa Rica;
ecological restoration; fertilization; invasive grass; reforestation;
39
INTRODUCTION
Improved methods for restoring tropical forests are important for meeting global
conservation (Possingham et al.2015) and climate change amelioration goals (Locatelli et
al.2015), given the capacity of second growth forests to support biodiversity (Chazdon et
al.2009; Dunn 2004; Bowen et al.2007) and to sequester atmospheric carbon dioxide
(Poorter et al.2016). Tropical forest area is declining (Hansen et al.2013; Asner et
al.2009) and calls have been made for large-scale reforestation of degraded lands
(Chazdon 2008; Chazdon 2014; Dobson 1997; Janzen, 1988a,b, Lamb et al.2005).
Tropical forest restoration, however, is often expensive (Lamb et al.2005), particularly
when land-use has resulted in increased soil compaction, decreased soil organic carbon,
and decreased nutrient-storing capacity (Ohsowski et al.2012; Hamza & Anderson 2005),
and when native tree saplings and seedlings must compete intensively with invasive
plants (Holl et al.2000; D’Antonio & Meyerson 2002). To date, however, few studies
have considered the possibility that large-scale application of agrowaste can catalyze
successional processes on degraded lands through amelioration of the abiotic conditions
and/or competition that prevents woody seedling establishment. A Web of Science search
with the terms ‘forest restoration’ and ‘agricultural waste’ or ‘agricultural byproduct’ or
‘crop residue’ yielded results related only to the use of municipal solid wastes and/or
sewage sludge (Shiralipour et al.1992; Cellier et al.2012), various forms of biochar
derived from agricultural sources (Thomas and Gale 2015), and ‘Bokashi’ (fermented
agricultural byproducts: Jaramillo-López et al.2015). Additional gray literature reports of
the use of composts, biosolids, and sewage sludge for soil remediation or forest
restoration also exist (EPA 2007; Rathinavelu and Graziosi 2005), but the authors could
40
not find well-documented examples of the use of direct application of agricultural waste
products for forest restoration.
However, a lack of peer-reviewed studies databased on Web of Science or other
readily accessible gray literature does not imply that trials or experiments on the use of
direct application of agricultural waste for forest restoration have not been conducted.
Here we report on an orange waste biodegradation project in Área de
Conservación Guanacaste (ACG) in northwestern Costa Rica (Jimenez 1998, 1999,
Janzen 2000, Escofet 2000, Daily and Ellison 2002) and document how this management
project, conducted through an agricultural waste disposal service contract, has changed
soil conditions and led to accelerated forest regeneration (in terms of woody biomass
recovery and tree species accumulation) on heavily degraded, abandoned pasture
characterized by compacted, rocky, nutrient-poor soils. The site was previously not
allowed to regenerate for a century or more because of active grazing and prescribed
burning prior to its incorporation into Guanacaste National Park in 1989 (later formally
incorporated into the larger ACG in 1994 around which time all remaining cattle were
fully removed from the park). These edaphic conditions combined with the continued
presence of invasive jaragua (Hyparrhenia rufa) grass are thought to be the primary
barriers to rapid natural regeneration in this area.
Project History
In 1991-1992, Del Oro S.A., an orange juice company, established thousands of
hectares of orange (Citrus × sinensis) plantations near the town of Santa Cecilia,
Guanacaste Province in northwestern Costa Rica. These plantations abut ACG
41
(http://www.acguanacaste.ac.cr) and occupy a once-forested ecosystem that is an
interface between Costa Rican lowland rain forest and dry forest (Janzen and Hallwachs
2016). By 1995, the first juice oranges were available and D. Janzen (hereafter DHJ)
asked Del Oro about the plans for disposing of the orange waste from its newly
constructed extraction plant (Janzen 2000). The orange waste was a product of both
machine-removal of juice as well as a second machine processing to remove most of the
essential oils from the rind, a common step taken with citrus crops (Weiss 1997). The
company replied it was going to construct a multi-million dollar drying and pelleting
plant to make cattle feed of the waste (Daily & Ellison 2002). From the ACG viewpoint,
the orange waste seemed to be an ideal food source for one or more of the estimated
375,000 species living in ACG (D. Janzen 2017, University of Pennsylvania, personal
communication). Seeking win-win partnerships with surrounding landholders, DHJ
offered a different plan for the orange waste: biodegrade it on recently incorporated
degraded pastureland within ACG. In return for this agricultural waste management, Del
Oro could donate its forested land at the margins of ACG that it had no intention of
cultivating, and eventually provide cash payments. An experiment was born. No active
planting or deliberate seeding of the site was planned as the necessary and sufficient
justification for the collaboration was anticipated win-win biodegradation of agricultural
waste; forest restoration was a secondary consideration albeit anticipated as a significant
ancillary benefit.
On 14 May 1996, at the beginning of the rainy season following the usual 5-
month dry season, Del Oro donated 100 dumptruck loads of orange processing waste
(~1,200 Mg), which was spread on about 0.25 ha of centuries-old ACG pasture (termed
42
‘Modulo I’). At that time, it was densely covered with introduced and ungrazed African
pasture grasses (primarily Hyparrhenia rufa) dotted with a few species of native shrubs.
18 months later, with no further treatment, the deposition had created a deep black loam
soil, and all grasses (whose roots had presumably been killed by the anoxic conditions
created by the orange waste) had been replaced by broad-leaved herbs (see Fig. S1). The
primary biodegraders of the orange waste were the larvae of three species of hoverflies
(Syrphidae, unknown species), an abundant soldier fly (Stratiomyidae, Hermetia
illucens), and their accompanying fungi and microbes (Janzen, personal observation,
Jimenez 1999), all common members of the decomposition process for fallen wild fruit
crops in the adjacent ACG forests.
With these experimental results of the pilot study in hand, ACG developed a
formal contract with Del Oro to biodegrade 1,000 truckloads (~12,000 Mg) of processed
orange waste per year for 20 years in exchange for 1,600 ha of intact primary forest land
owned by Del Oro that lay contiguous with ACG forest at 400-700 m elevation on the
slopes of Volcan Orosi and Orosilito (Janzen 2000, Daily & Ellison 2002). In the
beginning of the 1998 dry season (January), the first 1,000 truckloads were delivered to a
3 ha patch of highly degraded ACG pasture (termed ‘Modulo II’), a few km east of
Modulo I. This patch was selected for its convenient accessibility to trucks carrying
orange waste from a roughly 1 km by 1 km (100 ha) extent of largely homogenous
former pasture. The waste was left to biodegrade without further treatment. The project
was terminated after this step due to a complex series of politicized events that took place
over the following three years. These events began with a lawsuit filed by a competing
orange processing company on the grounds that Del Oro and ACG staff had sullied a
43
national park (Escofet 2000). However, the anticipated biodegradation process at Modulo
II continued and resulted in the disappearance of the orange waste. For additional
historical details on the Modulo II site and its context within the history of ACG as a
landscape-level forest restoration project see Janzen (2000), Daily and Ellison (2002),
and Allen (2001). While the orange waste project at Modulo II represents an unreplicated
treatment, such uniqueness has famously not prevented deep insights from analogous
ecological studies (e.g. Silvertown et al.2006; Whittaker et al.1989; Savchenko 1995).
The treated area of Modulo II is in fact slightly larger than the entire Park Grass
Experiment, arguably the longest-running and among the most important ecological
experiments ever conducted (Silverton et al.2006).
Here, we explore the current outcome of the Modulo II biodegradation project in
terms of impacts on soil chemistry, specifically concentration and bulk density of
nutrients, and forest recovery, specifically species richness and evenness of trees, canopy
closure, aboveground biomass, and numbers of saplings. Our point of comparison is
adjacent untreated, abandoned pasture. We choose to compare the Modulo II site to
adjacent untreated pasture rather than old-growth forest (by using compositional
similarity or other metrics benchmarked against old growth forest) given the lack of
suitable, non-riparian old-growth forest to serve as a baseline of comparison. Moreover,
we feel this comparison best reflects the actual opportunities that may arise if orange
waste is found to be benefit reforestation. We evaluate three hypotheses: deposition of
orange waste resulted in (1) an increase in quantity and availability of key soil macro-
and micronutrients; (2) an increase in species richness and evenness of tree species and
44
higher abundances of tree saplings; and (3) increased aboveground woody biomass and
greater canopy closure.
METHODS
Study Site and Biodegradation
The study site Modulo II is located on the former La Guitarra ranchland at the
east end of ACG’s Sector El Hacha (N 11.028°, W 85.523°, 290 m above sea level). The
site is located at the northwestern base of Volcan Orosí and at the eastern edge of the
ACG dry forest where it begins to intergrade with the ACG rainforest. Species
composition of the forest fragments in the surrounding landscape show a transition
between dry forest and rainforest flora (Janzen and Jimenez unpublished). The ranchland
of La Guitarra occupied ~1,000 ha and is believed to have been cleared in the 1600s or
1700s, and with the exception of stream buffers, remained unforested until the advent of
this project. Modulo II was located in the northeast corner of a ~100 ha block of
contiguous former pasture within La Guitarra. Before application of orange mulch in
1998, the site was covered primarily by Hyparrhenia rufa, and dotted with occasional
shrubby trees, primarily Curatella americana and Byrsonima crassifolia. No differences
in vegetation structure or composition were noted between the treatment area and the
surrounding ~100 ha of pasture of Modulo II prior to application of orange waste (D.
Janzen 2017, University of Pennsylvania, personal communication).
In 1998, 1,000 truckloads of orange waste (described above) were applied to a 3
ha plot (hereafter ‘orange waste treatment’) on the eastern side of a single-track dirt
access road at Modulo II. The organic material was spread into a layer ~0.1-0.5 m thick
by Del Oro using heavy machinery (Mata 1998), with an estimated weight at the time of
45
application of ~400 kg m-2 of which 320 kg m-2 was water and 80 kg m-2 was organic
waste (Universidad Nacional 1999), though DHJ believes the true mass of orange waste
was about half that value (D. Janzen 2017, University of Pennsylvania, personal
communication). Chemical analyses determined that the organic waste was 13%
cellulose, 8% protein, 68% carbohydrate, 4% fats and 5% ash (Universidad Nacional
1999). Nutrient surveys of the orange waste found 14.0 g Ca, 9.7 g K, 0.9 g Mg, and 1.2
g P per 1.0 kg of dry orange waste (Del Oro 1998).
Four months after initial deposition, there was still a layer of 0.1-0.2 m of organic
matter at the site (Janzen, personal observation; Mata 1998). At no point were seedlings
planted or attempts made to increase seed rain at the site, though the orange waste was
anticipated to remove grasses believed to be competing with seedlings. For all variables,
we compare measurements made within the 3 ha site receiving agrowaste to the adjacent
untreated abandoned pasture (hereafter ‘control’). Only the 3 ha closest to the treatment
site were surveyed, despite the 100 ha extent of the largely homogenous untreated former
pasture contiguous with the treatment site. In the years following the application of
orange waste, careful monitoring did not document any fires at either the control or
treatment sites.
Soil Sampling
We used two sets of soil samples to quantify initial and persistent changes in soil
chemistry resulting from orange waste deposition. The first set of samples was collected
and analyzed in 2000 by LS and the second set was collected in 2014 by JJC. Samples
were analyzed using different but comparable methods. In 2000, six composite samples
46
of twenty sub-samples each were taken from the orange waste treatment site. Six
composite control samples of twenty sub-samples each were then taken from untreated
pasture to the north and east of the orange waste treatment plot (see Appendix S1 for
details on the exact sampling scheme). pH, organic matter, and concentrations of
extractable Al3+, P, Ca, Mg, K, Cu, Fe and Zn were measured for each sample from
control and treatment plots. Sampling protocol details can be found in Appendix S1. Data
from control and treatment samples were compared using single-tailed Student’s t-tests in
Statgraphics (Rockville, MD).
A different soil sampling protocol was implemented in July 2014 by JJC and JSP
because they were unaware of the 2000 soil sample collection. Because only some of the
exact boundaries of the orange waste treatment area were clearly demarcated in 2014, soil
samples were taken from a 50 m by 50 m grid within the central core of the orange waste
treatment area. An identical grid was created on the opposite side of the access road in
the control to create a sampling design that captured a similar amount of landscape
heterogeneity.
As opposed to twelve composite samples of twenty subsamples from 2000, in
2014 eighteen composite soil samples were taken of nine subsamples each. These 2014
subsamples were collected to a depth of 0.1 m within a 1 m2 area next to each of nine
equidistant points in the grid mentioned above (see Appendix S1 for additional details).
Soils were analyzed for percent carbon, nitrogen and Mehlich III-extractable elements at
the Research Analytical Laboratory at the University of Minnesota. Welch’s unequal
variance t-tests and Wilcoxon rank-sum tests were used to compare the data depending
on the normality. As a drying oven was not available in 2014, the sampling grids were
47
revisited in July 2016 and twelve additional soil samples were taken (six from the orange
waste treatment site and six from the control) using a 0.1 m tall, 0.07 m diameter soil core
ring, and oven-dried at 100 C for 24 hours prior to weighing to determine bulk density.
Bulk densities were used to estimate total nutrient pools for each nutrient in the top 0.1 m
of the treatment and control sites (see Appendix S1 for additional details).
Vegetation
To quantify changes in vegetation structure and composition resulting from the
orange waste deposition, three 100 m by 6 m transects were established within the orange
waste treatment area at a distance 50 m, 75 m and 100 m from the access road dividing
the control and orange waste treatment plots in June 2014. An equivalent set of transects
was established in the control pasture on the opposite side of the road. Vegetation was
sampled following the approach of Powers et al. (2009). All trees larger than 5 cm
diameter at breast height (DBH) and taller than 1.3 m within 3 m of the centerline of each
transect were tagged, DBH was measured, and identified to species by JJC, DPA, and
TT. All saplings < 5 cm DBH and taller than 1.3 m that were growing within 3 m of the
centerline of each transect were measured for DBH, but were not tagged or identified, to
account for their aboveground biomass. All size measurements were completed during
the first two weeks of July 2014. Individual-based rarefaction curves were constructed to
quantify differences in species richness of trees.
Aboveground biomass was calculated for control and orange waste treatment sites
from transect DBH data using allometric scaling equations from van Breugel et al. (2011)
and wood density measurements from Powers & Tiffin (2010) and the Dryad Wood
48
Density Database (Chave et al.2009). Owing to the unreplicated nature of the Modulo II
management application, we used the data available to calculate the probability that one
would observe a difference in aboveground biomass as great as was actually observed
assuming the trees at the site were randomly distributed. This was done by constructing a
null model, wherein the estimated biomass of each measured tree was randomly assigned
to 'orange waste treatment' and 'control' populations with 50% probability. We conducted
1,000,000 such trials and ranked our observed difference by percentile as a non-
parametric test of significance.
To further determine the degree to which orange waste deposition resulted in
forest regeneration after 16 years, solar radiation indices, percent of visible sky, and leaf
area indices were determined using HemiView software and images taken with a fisheye
lens on 11 July 2014 (Rich et al.1999). Photos (n = 66) were taken 1.3 m off the ground
every 10 m along each transect within each set of three transects in both the waste
application plot and the control plot. A tripod was used for stabilization and a spherical
level was used to ensure that the camera was level. The photos were taken during the late
morning and early afternoon at times when the sky was deeply overcast to reduce glare in
the photos. Wilcoxon rank-sum tests were used to compare data.
49
Fig. 2.1: Aerial imagery of orange peel fertilized treatment area (mosaic of >10 m trees
and dense mats of herbaceous shrubs and vines to right of dirt road) and unfertilized
control (rocky expanse of grass with scattered ~2 m tall trees to left of dirt road) taken by
quadcopter drone in July 2015.
RESULTS
Soils
The application of orange waste led to dramatic differences in soil available
nutrients in both 2000 and 2014 as reflected in the differences between orange waste
treatment and control samples. In 2000, soils in the orange waste treatment site showed
increases in pH relative to the control (Student’s one-tail t-test, n = 6, p < 0.01, 10.9%
increase), and significantly higher concentrations, relative to the control, of extractable K,
Ca, Cu, Fe, and Zn. The initial increases in nutrient availability were largely maintained
fourteen years later. Moreover, orange waste deposition resulted in significant increases
50
in the macronutrients N (Welch t-test, n = 18, p < 0.001, 28.3% increase) and Mehlich III
extractable P (Wilcoxon, p < 0.001, 157.8% increase), and micronutrients Mg (Welch, p
= 0.002, 62.9% increase) and Mn (Welch, p = 0.012, 77.0% increase). Finally, orange
peel deposition resulted in a decreased C:N ratio (Welch, p < 0.001, 17.4% decrease),
when comparing 2014 orange waste treatment and control samples.
Bulk density was significantly higher in the control than in the treatment (Welch,
p = 0.022, 14.0%). However, even when correcting for the reduced mass in the top 0.1 m
of soil at the control and orange waste treatment sites, the significant disparities in
individual nutrients noted above persisted, albeit the magnitude of the difference was
reduced.
Table 2.1: Comparison of species richness, diversity, and evenness indicators for tree
species in control and treatment in 2014.
Vegetation
Deposition of orange peels resulted in differences in vegetation cover 16 years
later that were readily visible to the naked eye (Fig. 2.1). Within the total surveyed area
of 1800 m2 in the control site, we found 149 trees with a DBH greater than 5 cm from
eight different species from seven different families, compared to 133 trees, representing
51
twenty-four species from twenty families in the equivalent area of the treatment. These
133 trees along the treatment transects solely consist of new arrivals to the plot, as post-
orange waste deposition monitoring documented a die-off of all trees present at the time
of deposition, presumably through asphyxiation of roots. Of the 149 control plot trees,
134 (89.9%) were either Curatella americana or Byrsonima crassifolia (Fig. 2.2) both
species are associated with heavily degraded cattle pastures in this region (Condit et al.
2011). In comparison, six treatment plot species had more than 10 individuals and a
maximum abundance of just 16 individuals and included two species (Fig. 2.2), Trichilia
martiana (12 individuals) and Xylopia frutescens (11 individuals) that are associated with
advanced secondary growth or mature forest (Condit et al. 2011). The other four most
common treatment trees were Cercropia peltata (16 individuals), Guazuma ulmifolia (15
individuals), Cochlospermum vitifolium (13 individuals), and Curatella americana (11
individuals) (Fig. 2.2). The Shannon Index value for transects in the orange waste
treatment area was roughly triple the value of the control transects (Table 2.1), and
differences in species richness were taken to be statistically significant (p < 0.05) based
on non-overlapping 95% confidence intervals of rarefaction curves (Fig. 2.3). There were
820 saplings in the treatment compared to 353 in the control.
52
Fig. 2.2: Relative abundance of tree species in orange waste fertilized (treatment) and
unfertilized (control) transects.
Despite containing 10% fewer total stems > 5 cm DBH, the estimated
aboveground woody biomass of trees and saplings within the orange waste treatment was
nearly triple (73.69 Mg ha-1) that of the control (26.73 Mg ha-1). This observed difference
in biomass (46.96 Mg ha-1) was higher than the difference observed in 99.87% of null
model trials (Fig. 2.4). The difference in biomass remained significant after removing an
outlier tree from the treatment dataset and rerunning the model.
Canopy variables indicated a higher degree of canopy closure in the treatment
transects than in the control transects (Fig. 2.5). Leaf area index (LAI) and percent visible
sky calculated from hemispherical photography along the treatment transects (mean +/-
standard error: 2.184 +/- 0.996 and 17.2% +/- 11.3% respectively) were significantly
higher (Wilcoxon, p < 0.001), than LAI and percent visible sky along control transects
(0.355 +/- 0.367 and 59.2% +/- 15.9% respectively).
53
Fig. 2.3: Individual-based rarefaction and extrapolation of tree species richness in
treatment (pink) and control (blue). Shaded area indicates the 95% confidence interval for
the extrapolated or interpolated species richness.
DISCUSSION
Our results provide nuance and detail to what was overwhelmingly obvious during
informal surveys in 1999 and 2003: depositing orange waste on this degraded and
abandoned pastureland greatly accelerated the return of tropical forest, as measured by
lasting increases in soil nutrient availability, tree biomass, tree species richness, and
canopy closure. The clear implication is that deposition of agricultural waste could serve
as a tool for effective, low-cost tropical forest restoration, with a particularly important
potential role at low-fertility sites. As far as the authors are aware, this is the first
demonstration in the scientific literature of the forest restoration potential of direct
application of agricultural waste, without involving composting (Shiralipour et al.1992;
U.S. EPA 2007), pyrolyzation (Thomas and Gale 2015), or fermentation (Jaramillo-
54
López et al.2015). Direct application of agricultural waste, of course, assumes that
conditions for forest recovery are otherwise suitable (e.g. nearby seed sources, protection
from fires) and that fertilization with agrowaste is solving a disposal problem, rather than
competing with some other more lucrative downstream use for the waste (e.g.
Rathinavelu and Graziosi 2005), as well as a favorable socio-political environment. The
degree to which non-orange agrowastes (or orange wastes with essential oils present)
could be used to achieve similar results depends on the specific mechanisms by which
forest recovery is accelerated (in particular, grass suppression versus fertilization versus a
synergy of the two). It is also not clear from our study to what extent the removal of
grazers and suppression of fire are required to achieve successful restoration using
agricultural waste.
55
a)
56
b)
Fig. 2.4: (a) A plot of the distribution of the difference between 'treatment' and 'control'
aboveground biomass (AGB) in one million null model runs where trees measured in
2014 were randomly assigned into 'treatment' and 'control' designations after a single
Ficus sp. individual was removed from the dataset (the DBH of the individual was 15
standard deviations above the median tree DBH). The red line indicates observed
difference in biomass between actual treatment and control of 22.79 Mg ha-1, greater than
99.75% of null model trials. (b) Violin density plots of stem size distributions in
treatment (left) and control (right) with outlier Ficus sp. individual removed from
treatment dataset.
Soil Properties and Nutrients
The edaphic characteristics we measured showed dramatic changes toward
increased fertility as a result of orange waste deposition. Key macro- and micronutrients
(N, P, K, Ca, Mg, Mn, Cu, Fe, Zn) showed significantly elevated concentrations in
topsoil in the treatment site relative to the control 16 years after the initial orange waste
deposition. The addition of nutrient rich organic material likely played an important role
in accelerating the recovery of aboveground biomass and increasing the diversity of
57
woody plants species in the orange waste treatment site relative to the control. However,
the relative contributions of direct fertilization effects, indirect fertilization effects (e.g.
attracting or enabling the presence of facilitating species), and non-fertilization effects
(e.g. grass suppression) cannot easily be teased apart. Nonetheless, the soil sample results
from 2000 and 2014 hold several important clues for how orange waste deposition may
have improved soil conditions.
The reduction in soil acidity observed in 2000 in the treatment samples relative to
control samples suggests that the incorporation of Ca2+ and K+ cations may have
increased the pH of the soil by competing with H+ ions for adsorption sites (Gardiner &
Miller 2008). These two nutrients occur in high concentrations in biodegraded orange
peels (Del Oro 1998) and were found at elevated levels in the treatment soil samples in
2000 and 2014. The persistence of Ca2+ and K+ suggests that the cation exchange capacity
of the fertilized soils was not a short-lived effect. This makes the addition of orange peels
particularly beneficial for the dystrophic and acidic soils characteristic both of the
Modulo II site and many cleared tropical forests around the world (Guariguata and
Ostertag 2001).
One difference in macronutrients between orange waste treatment and control
sites in the 2000 and 2014 surveys worthy of discussion is phosphorous. In 2014 there
was 221.5% more P in top 0.1 m of soil in the orange waste treatment site than in the
control after correcting for bulk density differences. While the difference in P between
the two sites is striking, the total additional amount of P in the top 0.1 m of soil at the
treatment site is as little as 0.3% of the P originally present in the deposited orange waste,
a much lower remaining proportion than for Ca, Mg, or even leaching-prone K.
58
Combined with the dearth of nitrogen-fixing species common to young tropical forests
(Batterman et al.2013), this is suggestive of a P-limited system.
In summary, the effect of the orange peel deposition on edaphic conditions was
dramatic and could serve as a reasonable partial explanation for the difference in tree
species composition and aboveground biomass between orange waste treatment and
control. Soil fertility showed dramatic signs of improvement both 2 years and 16 years
after fertilization.
a)
59
b)
Fig. 2.5: (a) Violin density plots of percent visible sky in treatment and control transects
and (b) of leaf area index (LAI) in treatment and control transects determined using
hemispherical photography.
Vegetation
A dramatic increase in species richness and evenness as a result of the orange
waste deposition is unmistakable and would only further increase with the inclusion of
woody shrubs (e.g., Vachellia spp), vines, epiphytes and understory herbaceous species,
which we did not quantify. This expectation is supported by the presence of taller trees
that are better able to support lianas, epiphytes, and shade-tolerant plants (Choi & Treuer
unpublished data) within the orange waste treatment area.
When assessing secondary forests that are the product of restoration efforts,
several authors have called for careful attention not just to the conditions following
restoration, but also the trajectory that the forest is likely to follow, with respect to both
flora (Brown & Lugo 1990; Chazdon et al.2016) and fauna (Dent 2010). Orange waste
60
application resulted in a considerable increase in the system’s fire resilience via
suppression of highly inflammable grasses. Additionally, the more than doubling in
sapling number in the treatment area relative to the control suggest that differences in
aboveground biomass are likely to be maintained into the future. Finally, the presence of
Cecropia peltata and Ficus sp. individuals in the treatment but not control area is
important as both are known to provide important fruit resources to many forest dwelling
animals (Fleming & Williams 1990).
Management and Policy Implications
When the contract between Del Oro and ACG was voided by the Costa Rican
government Comptroller General in 1999 and ACG staff given an impossible-to-execute
order by the Sala Cuarta to remove the orange waste that had long since degraded,
substantive (as opposed to aesthetic) concerns with the biodegradation of the orange
waste reportedly centered on the notion that the mulch would become a breeding ground
for pests or pathogens or that there would be significant leaching of problematic
compounds into surrounding waterways, most prominently the essential oil D-limonene,
which was claimed to be a carcinogen (Escofet 2000). These concerns turned out to be
baseless; an independent team of scientists dismissed concerns over pollution and threats
to nearby producers as outlandish (Escofet 2000) and D-limonene has been found not to
be carcinogenic (Asamoto et al.2005). Nevertheless, subsequent attempts to revive or
establish similar restoration projects using agricultural waste have stalled due to the
potential partners not wishing to risk highly politicized litigation again. These
ameliorated concerns combined with the overwhelmingly positive results of orange waste
61
application on forest restoration suggest that agricultural waste could have considerable
potential as a management tool for forest restoration, though certainly careful
consideration of social, political, and idiosyncratic environmental conditions (e.g.
potential for harsh chemical or biohazardous runoff into local waterways) is warranted.
The potential harm from pesticides or other problematic compounds in particular merit
careful consideration and safeguards. Assuming these conditions can be met, further
explorations of using agricultural wastes for restoration should be encouraged and
potentially subsidized through existing or future payments for ecosystem services
schemes, such as already exist in Costa Rica (Daily & Ellison 2002), rather than
aggressively prohibited.
When agroindustry produces nutrient-rich, but costly-to-dispose-of waste streams
(as in the case of oranges in Costa Rica), there is an opportunity for low-cost (or indeed,
cost-negative), scalable, biodiversity-friendly, carbon-sequestering ecological restoration.
Given that the scale of biodiversity-friendly restoration activities is typically limited by
cost (Lamb et al.2005) or by sociopolitical prohibitions as is the case described here, the
results of this management project suggest that this general approach should be widely
trialed in a variety of settings, using a variety of agricultural inputs. While the
agroindustry may well be aware these wastes could serve to achieve biodiversity and
climate mitigation goals via accelerated forest regeneration, as this study underscores,
achieving positive outcomes requires outreach on the part of restoration ecologists as well
as favorable regulatory regimes.
62
WORKS CITED
Allen W (2001) Green Phoenix: Restoring the Tropical Forests of Guanacaste, Costa
Rica. Oxford University Press, USA
Asamoto M, Ota T, Toriyama-Baba H, Hokaiwado N, Naito A, Tsuda H (2002)
Mammary Carcinomas Induced in Human c-Ha- ras Proto-oncogene Transgenic
Rats Are Estrogen-independent, but Responsive to d -Limonene Treatment.
Japanese Journal of Cancer Research 93:32–35
Asner GP, Rudel TK, Aide TM, Defries R, Emerson R (2009) A contemporary
assessment of change in humid tropical forests. Conservation biology : the journal
of the Society for Conservation Biology 23:1386–95
Batterman S a, Hedin LO, van Breugel M, Ransijn J, Craven DJ, Hall JS (2013) Key role
of symbiotic dinitrogen fixation in tropical forest secondary succession. Nature
502:224–7
Bond WJ (2008) What Limits Trees in C 4 Grasslands and Savannas? Annual Review of
Ecology, Evolution, and Systematics 39:641–659
Bowen ME, McAlpine C, House APN, Smith GC (2007) Regrowth forests on abandoned
agricultural land: A review of their habitat values for recovering forest fauna.
Biological Conservation 140:273–296
Brown S, Lugo AE (1990) Tropical secondary forests. Journal of Tropical Ecology 6:1–
32
Carter M (1993) Soil sampling and methods of analysis. Lewis Publishers, USA
Chave J, Coomes D, Jansen S, Lewis SL, Swenson NG, Zanne AE (2009) Towards a
worldwide wood economics spectrum. Ecology Letters 12:351–366
63
Chazdon RL (2008) Beyond deforestation: restoring forests and ecosystem services on
degraded lands. Science (New York, N.Y.) 320:1458–60
Chazdon RL (2014) Second Growth. University of Chicago Press
Chazdon RL, Brancalion PHS, Laestadius L, Bennett-Curry A, Buckingham K, Kumar C,
Moll-Rocek J, Vieira ICG, Wilson SJ (2016) When is a forest a forest? Forest
concepts and definitions in the era of forest and landscape restoration. Ambio 1–
13
Chazdon RL, Peres CA, Dent D, Sheil D, Lugo AE, Lamb D, Stork NE, Miller SE (2009)
The potential for species conservation in tropical secondary forests. Conservation
biology : the journal of the Society for Conservation Biology 23:1406–17
Condit R, Perez R, Daguerre N (2011) Trees of Panama and Costa Rica. Princeton
University Press
Daily GC, Ellison K (2002) Costa Rica: Paying Mother Nature to Multitask. In: The New
Economy of Nature: The Quest to Make Conservation Profitable. Island Press pp.
165–189.
D’Antonio C, Meyerson LA (2002) Exotic Plant Species as Problems and Solutions in
Ecological Restoration: A Synthesis. Restoration Ecology 10:703–713
Del Oro (1998) Compendido de documentos relacionada con las actividades de carácter
ambiental de Del Oro. Guanacaste Costa Rica
Dent DH (2010) Defining the conservation value of secondary tropical forests. Animal
Conservation 13:14–15
Dobson AP (1997) Hopes for the Future: Restoration Ecology and Conservation Biology.
Science 277:515–522
64
Dunn RR (2004) Recovery of faunal communities during tropical forest regeneration.
Conservation Biology 18:302–309
Escofet G (2000) Costa Rican orange-peel project turns sour. Ecoamericas
Fleming TH, Williams CF (2009) Phenology, seed dispersal, and recruitment in Cecropia
peltata (Moraceae) in Costa Rican tropical dry forest. Journal of Tropical Ecology
6:163
Gardiner D, Miller R (2008) Soils in our environment. Pearson/Prentice Hall, Upper
Saddle River, NJ
Guariguata MR, Ostertag R (2001) Neotropical secondary forest succession: changes in
structural and functional characteristics. Forest Ecology and Management
148:185–206
Hamza MA, Anderson WK (2005) Soil compaction in cropping systems. Soil and Tillage
Research 82:121–145
Hansen MC, Potapov P V, Moore R, Hancher M, Turubanova SA, Tyukavina A et al.
(2013) High-resolution global maps of 21st-century forest cover change. Science
(New York, N.Y.) 342:850–3
Holl KD, Loik ME, Lin EHV, Samuels IA (2000) Tropical Montane Forest Restoration in
Costa Rica: Overcoming Barriers to Dispersal and Establishment. Restoration
Ecology 8:339–349
Janzen DJ (1988) Guanacaste National Park: tropical ecological and biocultural
restoration. In: Rehabilitating Damaged Ecosystems. Cairns, J, editor. CRC Press,
Boca Raton, FL pp. 143–192.
65
Janzen DH (2000) Costa Rica’s Area de Conservación Guanacaste: A long march to
survival through non-damaging biodevelopment. Biodiversity 1:7–20
Janzen DH (1988) Management of habitat fragments in a tropical dry forest: growth.
Annals of the Missouri botanical garden 75:105–116
Janzen DH, Hallwachs W (2016) Biodiversity Conservation History and Future in Costa
Rica: The Case of Area de Conservacion Guanacaste (ACG). In: Costa Rican
Ecosystems. Kappelle, M, editor. University of Chicago Press, Chicago pp. 290–
341.
Jimenez G (1999) Resultos preliminares del proyecto: manejo y tratamiento de pulpa de
naranja. Rothschildia 6
Lathwell D, Grove T (1986) Soil-Plant Relationships in the Tropics. Annual Review of
Ecology and Systematic 17:1–16
Locatelli B, Catterall CP, Imbach P, Kumar C, Lasco R, Marín-Spiotta E, et al. (2015)
Tropical reforestation and climate change: beyond carbon. Restoration Ecology
23:337–343
Mata A (1998) Visita a la empresora Del Oro y al Parque Nacional de Guanacaste (3 y 4
de Augusto de 1998). Área de Conservación de Guanacaste, Liberia, Costa Rica
Ohsowski BM, Klironomos JN, Dunfield KE, Hart MM (2012) The potential of soil
amendments for restoring severely disturbed grasslands. Applied Soil Ecology
60:77–83
Poorter L, Bongers F, Aide TM, Almeyda Zambrano AM, Balvanera P, Becknell JM, et
al. (2016) Biomass resilience of Neotropical secondary forests. Nature 530:211–
214
66
Possingham HP, Bode M, Klein CJ (2015) Optimal conservation outcomes require both
restoration and protection. PLoS biology 13:e1002052
Powers JS, Becknell JM, Irving J, Pèrez-Aviles D (2009) Diversity and structure of
regenerating tropical dry forests in Costa Rica: Geographic patterns and
environmental drivers. Forest Ecology and Management 258:959–970
Powers JS, Tiffin P (2010) Plant functional type classifications in tropical dry forests in
Costa Rica: leaf habit versus taxonomic approaches. Functional Ecology 24:927–
936
Rich P, Wood J, Vieglais D, Burek K, Webb N (1999) Hemiview User Manual.
Savchenko VK (Vladimir K (1995) The ecology of the Chernobyl catastrophe : scientific
outlines of an international programme of collaborative research. UNESCO
Silvertown J, Poulton P, Johnston E, Edwards G, Heard M, Biss PM (2006) The Park
Grass Experiment 1856-2006: its contribution to ecology. Journal of Ecology
94:801–814
van Breugel M, Ransijn J, Craven D, Bongers F, Hall JS (2011) Estimating carbon stock
in secondary forests: Decisions and uncertainties associated with allometric
biomass models. Forest Ecology and Management 262:1648–1657
Weiss EA (1997) Essential oil crops. Cab International, Wallingford, UK.
Whittaker RJ, Bush MB, Richards K (1989) Plant Recolonization and Vegetation
Succession on the Krakatau Islands, Indonesia. Ecological Monographs 59:59–
123
67
CHAPTER 3
Organic Wastes and Tropical Forest Restoration
Author List:
Choi, Jonathan J.,1 Treuer, Timothy L. H.,2 Werden, Leland K.,3 Wilcove, David S.,2,4
1: Duke Marine Geospatial Ecology Lab, Duke University Nicholas School of the
Environment, 9 Circuit Drive, Durham, NC 27710
2: Department of Ecology and Evolutionary Biology, Princeton University, 106A Guyot
Lane, Guyot Hall, Princeton, NJ 08541
3: Lyon Arboretum, University of Hawai'i, 3860 Manoa Rd, Honolulu, HI 96822
4: Woodrow Wilson School of Public and International Affairs, Princeton University, 20
Prospect Ave, Robertson Hall, Princeton, NJ 08541
Keywords: Agricultural waste; forest restoration; soil; carbon offsets; soil amendments;
arrested succession; fire traps; nutrient depletion
68
Abstract
In a recent publication, we documented the benefits of using agricultural waste
(specifically, leftover orange peels from a commercial orange juice factory) to promote
forest recovery at a site in Costa Rica. While we showed unambiguously positive impacts
on soil conditions, forest biomass, and tree diversity, our ability to infer mechanisms
behind this recovery was limited because the project was never replicated. It appears our
work is one of only a handful of peer-reviewed studies testing the use of unprocessed
agricultural waste as part of a tropical forest restoration initiative. We argue that
regardless of the mechanism, there are first-principle reasons to expect that minimally
processed (and thus low-cost) agricultural wastes could be utilized to accelerate tropical
forest restoration in a variety of contexts, potentially creating a new class of biodiversity-
friendly carbon offsets that may address previous concerns about linking tropical forestry
to global carbon markets. We outline research initiatives that could lead to a richer
understanding of when and where it is safe and effective to utilize agricultural and other
wastes in tropical forest restoration endeavors.
Keywords: agricultural waste, forest restoration, soil, carbon offsets, soil amendments,
arrested succession, fire traps, nutrient depletion
69
In 1998, scientists and managers of Area de Conservacion Guanacaste (ACG) in
northwest Costa Rica looked on as trucks filled with orange peels drove across a wooden
bridge and down a narrow dirt access road. The trucks unloaded what would eventually
amount to an estimated 12,000 tonnes of organic byproducts from a nearby jui- cing and
essential oil extraction facility onto a 3-hectare corner of an abandoned cattle pasture
overgrown with jaragua (Hyparrhenia rufa), an invasive, fire-adapted grass that had long
restricted regrowth.
This unusual activity was the product of a ground- breaking 20-year contract
between ACG and the juice company, Del Oro, wherein the company received free waste
disposal services (obviating the need for capital intensive investments in waste-
processing equipment and infrastructure) in exchange for giving the park mul- tiple
blocks of hilly, still-forested land unsuitable for cul- tivation. Scientific advisers to the
park (most prominently Daniel Janzen and Winnie Hallwachs) conducted a pro- mising
pilot trial 2 years before and were confident the orange peels would readily degrade and
promote vegeta- tion regrowth with minimal ecological risk to the park or the
surrounding community. This formalized (and quite ahead of its time) ecosystem services
agreement seemed to offer a compelling model for future mutually beneficial
arrangements between agroindustry and con- servation initiatives.
Then came the lawsuit. Alleging damage to a national park, a rival juice company
successfully litigated an immediate termination to the project, voiding the con- tract
between ACG and Del Oro. Janzen, Hallwachs, and others continued to visit the site long
enough to know that the orange waste had initiated a regime shift in vegetation cover
from grass to woody plants, but no attempt to rigorously quantify these changes was
70
made until 2014, when we conducted a variety of soil, vegetation and other surveys. We
recently published these findings as a case study (Treuer et al., 2017). While we found
dramatic improvements in aboveground biomass, soil condition, and tree diversity, we
were unable to confidently ascribe a mechanism by which the deposition of orange waste
accelerated forest recovery, due to the lack of replication of the treatment and the nature
of the original experimental design, which was structured as a proof of concept rather
than as a mechanistic study.
While we were greatly impressed by the rapid recovery at the site in 2014, we
were also surprised to find that ours was one of the only studies that we could find in
which a minimally processed organic waste was used as an active tropical forest
restoration approach. We believe that there are first-principle reasons that future
endeavors using such wastes would be highly effective in not only accelerating forest
regeneration in a variety of contexts across the tropics, but also in mitigating arrested
succession. We discuss those reasons below and conclude with a brief discussion of a
promising research directions and the potential to incorporate these restoration methods
into carbon markets.
Mitigating arrested or delayed succession
Across the tropics, trends in globalization, urbanization, agricultural
intensification, and traditional shifting-cultivation practices have driven forest regrowth
on abandoned agricultural and pastoral lands. However, the pace of forest regeneration on
abandoned agricultural lands is hardly uniform and varies with the broad diversity in
tropical forest type (Holl, 2007; Janzen, 1988), soil conditions (Powers et al 2009),
rainfall gradients (Poorter et al., 2016), presence of weedy or invasive vegetation (Calvo-
71
Alvarado, McLennan, Sánchez-Azofeifa, & Garvin, 2009), fire (D’Antonio & Vitousek,
1992), and land use history (Crouzeilles et al., 2016; Holl & Zahawi, 2014). In some
instances, these and other conditions (e.g. insufficient seed rain, seedling and sapling
mortality, and slow woody biomass accumulation) present enough of a barrier to forest
recovery as to fully arrest succession, at least along timescales that are relevant to human
and conservation needs (D’Antonio & Vitousek, 1992; Holl, 2007; Holl, Loik, Lin, &
Samuels, 2000). Of these variables, we believe there are two conditions that present
particularly difficult obstacles to regeneration in ACG and across in the tropics.
The first of these conditions is the nutrient poor nature of many tropical forest
soils. Ecologists have long noted that tropical forest soils are depauperate in key
nutrients, with a high proportion found in living tissues that quickly get recycled upon
decomposition (Vitousek & Sanford, 1986). During the process of land conversion, often
via slash and burn agriculture, nutrients are lost from the ecosystem via the harvesting
and removal of nutrient-rich plant tissue, the volatilization of nitrogen by fire, and
increased runoff (Griscom & Ashton, 2011). Nutrient deficiency decreases the
establishment and growth of tree species during succession (Davidson et al., 2004;
Kitajima & Tilman, 1996), and has been documented in many settings, including at least
one site with nutrient-rich soils (Chou, Hedin, & Pacala, 2018), and is also evidenced by
high rates of symbiotic nitrogen fixation early in succession (Batterman et al., 2013).
The second challenge--the presence of weedy, often non-native vegetation that
competes with seedlings for resources—interacts with the challenges posed by nutrient
limitation. In treeless cattle pastures, aggressive herbaceous vegetation (e.g. perennial
grasses and bracken ferns) can grow rapidly, competing for resources and shading out
72
later successional species (Calvo-Alvarado et al., 2009; Shoo & Catterall, 2013). Such
vegetation fuel hot fires that kill seedlings and saplings -- the so-called ‘fire trap’ that
drives the bistability of savannas and forests in certain climates (sensu Pellegrini 2016).
Availability of nitrogen and phosphorus, or even micronutrients like Ca2+ and Mg2+
(Hoffmann et al., 2009), by increasing tree seedling growth rates, can allow systems to
escape the fire trap (Pellegrini, 2016) .
Both of these challenges can potentially be overcome through traditional tropical
forest restoration techniques involving fertilizer and herbicide application (Griscom &
Ashton, 2011), but these approaches are typically expensive and can incur negative
carbon and health outcomes. Scientists have engaged in a wide variety of regenerative
efforts, including the application of manure, compost, or biochar, as well as direct
seeding or seedling planting (Griscom & Ashton, 2011). However, we believe the
application of many types of agricultural wastes may achieve the same results at a far
lower material and labor cost. Indeed, we suspect the application of orange waste at our
site in Costa Rica led to forest recovery by mitigating these two interacting challenges.
Within a couple of months of their application, the orange waste had killed off the grasses
(likely via asphyxiation) and broken down into a nutrient rich organic layer. These
fertilized soils, long after catalyzing extensive regrowth, still hold more nutrients than the
surrounding area, even after accounting for their lower bulk density.
Promising Directions
While whole industries now exist to convert various agricultural and other
nutrient-rich wastes into useful products, these ventures are often capital-intensive or
73
yield products with limited markets, conditions that may prevent widespread adoption in
tropical developing nations. In the case of our study, it was largely the costs of turning
the orange peels into a benign product, in this case cattle feed, that drove interest in the
waste disposal contract with ACG (Janzen pers. Comm.). After media coverage of our
study, we were contacted by numerous citrus growers interested in exploring similar
arrangements, suggesting that for this particular type of waste, lucrative markets for
waste derivatives are not universal and alternative disposal costs are high (Fig. 3.1).
Figure 3.1 – Orange peels gathered for disposal in Ghana. Photo courtesy of Lyndon
Estes.
Beyond citrus waste, future research should examine other organic wastes,
including non-citrus fruits grown for juice or oil extraction as well as non-marketed
fleshy fruits like the skins of coffee fruit or cashew fruits, for their tropical forest
74
restoration potential. Agricultural waste has long been used in forestry settings to
ameliorate soil conditions on the scale of individual trees (Harris, 1992). The application
of Biochars, manures, and agricultural waste residues in applied forest restoration settings
has also demonstrated great potential to increase biomass production and improve a suite
of soil characteristics (Box 3.1). Beyond plant-based agriculture by-products, organic
municipal waste, chicken litter, and sewage from concentrated feedlots all are nutrient-
wastes that at least in some contexts are costly to process otherwise.
In all cases, but particularly with organic wastes with the potential to negatively
impact downstream communities (Herrera et al., 2017), scientifically guided and
carefully managed enterprises are essential. Pilot studies are necessary both to establish
that organisms present in the surrounding environment are capable of breaking down the
Box 3.1 – Effective use of processed agricultural waste to ameliorate soil
conditions and catalyze succession
Biochars: Agricultural waste residues converted to Biochars can be high in soluble
nutrients (phosphorus and potassium), and can retain water, cations and anions. A
meta-analysis showed over 40% mean increases in biomass production in restoration
applications after Biochar applications (Thomas and Gale 2015). However, the
effectiveness of these soil amendments deserves further investigation as Biochars can
have both positive and negative impacts on forest regrowth (Thomas and Gale 2015).
For example, a forest restoration project in the ACG used agricultural waste to
ameliorate degraded soil and showed no impact of Biochar application on tree
seedling growth or survival (Werden et al. 2018).
Composts and manures: The application of urban household compost to abandoned
agricultural land can led to increased plant recruitment from existing seed bank and
increased plant grow rates (Ros et al. 2003). The application of composts can also
stimulate microbial activity and renew soil nutrient cycles by increasing nitrogen
mineralization, microbial respiration, and soil carbon storage (Harris 2003).Many
studies have also found manure application to increase total soil organic carbon and
water holding capacity of degraded soils (Khaleel et al. 1981).
75
organic waste and that runoff will not pose a health risk to local waterways and
populations. Likewise, lifecycle analysis of wastes and methane and nitrous oxide
emissions are needed to ensure that negative climate impacts do not offset the carbon
sequestration benefits of enhanced tree growth. In addition, projects guided by input from
local stakeholders should have greater odds of acceptance than those led by private
enterprises without meaningful public input, which can trigger accusations of pollution,
profiteering or other malfeasance.
Ensuring the efficient application of organic waste for the maximum
environmental impact should be a major focus of future research efforts. In our study,
orange waste from several hundred hectares of productive orchards was used to
accelerate forest succession on just 3 hectares. Additional research should determine the
quantity of agricultural waste needed in a particular area to jumpstart regrowth as well as
the minimum spatial scale of application needed to prevent the reinvasion of grasses.
Research should also address the marginal impact of additional waste per area on forest
biomass and species diversity.
Other important research directions include whether waste-application techniques
can be deployed successfully in other climates. Most of the citrus growers we were
contacted by hailed from Mediterranean biomes, where forest regeneration may not be
nutrient limited, and organisms that break down agricultural wastes may be less likely to
occur in needed densities (though increased soil water retention may offer a separate
mechanism for organic wastes to speed vegetation regrowth). In all biomes, it is also
worth comparing how effective waste application is relative to other more established
techniques.
76
Climate mitigation
Improving the carbon sequestration potential of pastures and fields to mitigate
climate change has been the subject of considerable recent research (Ryals, Kaiser, Torn,
Berhe, & Silver, 2014) and has implications for existing international carbon regimes,
such as REDD+ and the Bonn Challenge. However, many schemes suffer from
difficulties in demonstrating the two pillars of carbon mitigation, namely additionality
and a lack of leakage. From an economic perspective, the laws of supply and demand and
the inherent fungibility of timber sources from different locales means that scaling up
payments to prevent logging in one place will only lead to timber harvest in another.
Likewise, payments to set aside areas for forest regeneration may end up going to
landowners who would have let forest recover anyway. Accelerating natural forest
regeneration offers a potentially scalable mode of carbon offsetting, but with fewer
concerns of leakage or missing additionality.
Incorporating the use of organic wastes as a form of carbon offsetting into carbon
markets could divert otherwise only marginally beneficial organic wastes, such as oil
palm fibers that are often burned for power at oil extraction facilities, to projects related
to forest restoration, all while providing other positive externalities like improved local
air quality and enhanced biodiversity. As such, ecologists should work with
environmental economists to better understand how organic wastes might be incorporated
into carbon market schemes.
Ultimately, our study has demonstrated large potential for low-cost agricultural
waste to catalyze tropical forest recovery and further research to optimize the
effectiveness of its application should be encouraged.
77
WORKS CITED
Batterman, S. A., Hedin, L. O., Van Breugel, M., Ransijn, J., Craven, D. J., & Hall, J. S.
(2013). Key role of symbiotic dinitrogen fixation in tropical forest secondary
succession. Nature, 502(7470), 224–227. https://doi.org/10.1038/nature12525
Calvo-Alvarado, J., McLennan, B., Sánchez-Azofeifa, A., & Garvin, T. (2009).
Deforestation and forest restoration in Guanacaste, Costa Rica: Putting conservation
policies in context. Forest Ecology and Management, 258(6), 931–940.
https://doi.org/10.1016/j.foreco.2008.10.035
Chou, C. B., Hedin, L. O., & Pacala, S. W. (2018). Functional groups, species and light
interact with nutrient limitation during tropical rainforest sapling bottleneck. Journal
of Ecology, 106(1), 157–167. https://doi.org/10.1111/1365-2745.12823
Crouzeilles, R., Curran, M., Ferreira, M. S., Lindenmayer, D. B., Grelle, C. E. V., & Rey
Benayas, J. M. (2016). A global meta-Analysis on the ecological drivers of forest
restoration success. Nature Communications, 7(May), 1–8.
https://doi.org/10.1038/ncomms11666
D’Antonio, C. M., & Vitousek, P. M. (1992). Biological Invasions by Exotic Grasses, the
Grass/Fire Cycle, and Global Change. Annual Review of Ecology and Systematics,
23(1), 63–87. https://doi.org/10.1146/annurev.es.23.110192.000431
Davidson, E. A., Reis de Carvalho, C. J., Vieira, I. C. G., Figueiredo, R. D. O., Moutinho,
P., Yoko Ishida, F., … Tuma Sabá, R. (2004). Nitrogen and Phosphorus Limitation
of Biomass Growth in a Tropical Secondary Forest. Ecological Applications,
14(sp4), 150–163. https://doi.org/10.1890/01-6006
78
Griscom, H. P., & Ashton, M. S. (2011). Restoration of dry tropical forests in Central
America: A review of pattern and process. Forest Ecology and Management,
261(10), 1564–1579. https://doi.org/10.1016/j.foreco.2010.08.027
Harris, R. W. (1992). Arboriculture: integrated management of landscape trees, shrubs
and vines (2nd ed.). Upper Saddle River, New Jersey: Prentice-Hall International.
Herrera, D., Ellis, A., Fisher, B., Golden, C. D., Johnson, K., Mulligan, M., … Ricketts,
T. H. (2017). Upstream watershed condition predicts rural children’s health across
35 developing countries. Nature Communications, 8(1).
https://doi.org/10.1038/s41467-017-00775-2
Hoffmann, W. A., Adasme, R., Haridasan, M., De Carvalho, M. T., Geiger, E. L.,
Pereira, M. A. B., … Franco, A. C. (2009). Tree topkill, not mortality, governs the
dynamics of savanna-forest boundaries under frequent fire in central Brazil.
Ecology, 90(5), 1326–1337. https://doi.org/10.1890/08-0741.1
Holl, K. D. (2007). Oldfield vegetation succession in the Neotropics in the Neotropics. In
R. J. Hobbs & V. A. Cramer (Eds.), Old Fields: Dynamics and Restoration of
Abandoned Farmland. Washington DC: Island Press.
Holl, K. D., Loik, M. E., Lin, E. H. V., & Samuels, I. A. (2000). Tropical Montane Forest
Restoration in Costa Rica: Overcoming Barriers to Dispersal and Establishment.
Restoration Ecology, 8(4), 339–349.
Holl, K. D., & Zahawi, R. A. (2014). Factors explaining variability in woody above-
ground biomass accumulation in restored tropical forest. Forest Ecology and
Management, 319, 36–43. https://doi.org/10.1016/j.foreco.2014.01.024
79
Janzen, D. H. (1988). Tropical Dry Forests. In E. O. Wilson (Ed.), Biodiversity.
Washington DC: The National Academies Press.
https://doi.org/https://doi.org/10.17226/989
Kitajima, K., & Tilman, D. (1996). Seed Banks and Seedling Establishment on an
Experimental Productivity Greadient. Oikos, 76(2), 381–391.
Pellegrini, A. F. A. (2016). Nutrient limitation in tropical savannas across multiple scales
and mechanisms. Ecology, 97(2), 313–324. https://doi.org/10.1002/ecm.1283
Poorter, H., Fiorani, F., Pieruschka, R., Wojciechowski, T., van der Putten, W. H.,
Kleyer, M., … Postma, J. (2016). Pampered inside, pestered outside? Differences
and similarities between plants growing in controlled conditions and in the field.
New Phytologist, 212(4), 838–855. https://doi.org/10.1111/nph.14243
Ryals, R., Kaiser, M., Torn, M. S., Berhe, A. A., & Silver, W. L. (2014). Impacts of
organic matter amendments on carbon and nitrogen dynamics in grassland soils. Soil
Biology and Biochemistry, 68, 52–61. https://doi.org/10.1016/j.soilbio.2013.09.011
Shoo, L. P., & Catterall, C. P. (2013). Stimulating natural regeneration of tropical forest
on degraded land: Approaches, outcomes, and information gaps. Restoration
Ecology, 21(6), 670–677. https://doi.org/10.1111/rec.12048
Treuer, T. L. H., Choi, J. J., Janzen, D. H., Hallwachs, W., Peréz-Aviles, D., Dobson, A.
P., … Wilcove, D. S. (2017). Low-cost agricultural waste accelerates tropical forest
regeneration. Restoration Ecology, 26(2), 275–283.
https://doi.org/10.1111/rec.12565
80
Vitousek, P. M., & Sanford, R. L. (1986). Nutrient Cycling in Moist Tropical Forest.
Annual Review of Ecology and Systematics, 17, 137–167.
https://doi.org/0.1146/annurev.es.17.110186.001033
81
CHAPTER 4
Where the Wild Things Are: Second-growth Forest Heterogeneity Reflected in Terrestrial
Vertebrate Community Composition but Not Site Occupancy
Author List: Timothy L.H. Treuer, Alexander Gow, Fangyuan Hua, Justin Becknell,
Andrew P. Dobson, Jennifer Powers, David S. Wilcove
Corresponding Author: Timothy L.H. Treuer, [email protected], 106a Guyot Hall,
Princeton, NJ 08544
82
Abstract
The conservation value of second-growth forest habitats for wildlife in the tropics has
been hotly contested, with different taxa exhibiting divergent habitat usage patterns under
varying sociopolitical contexts. For heavily degraded and fragmented regions, including
many biodiversity hotspots, this debate is anything but academic; limited resources must
be partitioned between conservation of remaining fragments and restoration of habitat.
The degree to which forest species can utilize regenerating habitat makes a difference in
how the two strategies should be prioritized. This study seeks to establish patterns of
habitat usage for a data-deficient group of organisms in a particular setting—terrestrial
vertebrates in a Mesoamerican dry forest—while also assessing how frequently ignored
dimensions of second-growth forest heterogeneity might be reflected in the site
occupancy of various species. We recorded 26 species of terrestrial vertebrates using 40
camera traps deployed for an average of 181 days in the second-growth dry forests of
Área de Conservación Guanacaste (ACG), Costa Rica. We obtained enough capture
events for 12 species to model their second-growth forest usage under and occupancy
modelling framework. We recorded 62 capture events for the endangered Baird’s tapir
(Tapirus bairdii), of which 60 occurred in forest less than or equal to 30 years in stand
age. Overall, our findings fail to suggest that young second-growth dry forest habitat is
less valuable for the conservation of terrestrial vertebrates in this system than old-growth
forest, implying that strategies to increase the size of existing dry forest fragments
through passive restoration may be an effective use of conservation funds within this
system.
83
Highlights:
● We conducted a 7,255 camera trap-day study of a secondary tropical dry forest
system.
● At 40 sites we recorded 2,186 capture events and 26 species of terrestrial
vertebrates.
● 60 out of 62 capture events of endangered Baird’s tapir occurred in ≤30 year
regrowth.
● Stand age was a weak occupancy predictor for 10 of 12 most common species.
Keywords: camera trap; occupancy model; tropical forest; second-growth forest;
succession; dry forest; Tapirus bairdii
84
INTRODUCTION
Restoration of habitat is essential for achieving many conservation goals,
particularly in heavily degraded and fragmented landscapes characteristic of many
tropical forest biodiversity hotspots (Myers et al. 2000). Possingham et al. (2015) present
a theoretical model that goes as far as suggesting that some portion of funding for
biodiversity preservation ought to be utilized for ecological restoration even before
remaining habitat has been fully protected, because of the value of habitat creation for
mitigating extinction debt. However, the optimal value of restoration for the conservation
of faunal assemblages depends heavily on the ability of animals to utilize early
successional habitat. Despite substantial effort to elucidate the value of second-growth
tropical forest for biological conservation, an incomplete picture remains (Bowen et al.
2007; Dent & Wright 2009; Chazdon 2014). Capacity to utilize regenerating habitat
varies by taxa, even within a single system (Gardner et al. 2008).
Moreover, second-growth forests are not monolithic. Though stand age is
frequently included as an explanatory variable in analyses of community composition and
diversity of regenerating systems, other drivers of forest heterogeneity are often
overlooked. Many of these are orthogonal axes of environmental variation that shape
important elements of second-growth forest structure or composition, such as edaphic
conditions (Powers et al. 2009; Becknell & Powers 2014), legacy trees (Schlawin &
Zahawi 2008; Sandor & Chazdon 2014; Treuer et al. in prep), landscape composition
(Lamb 2014), and restoration treatment (e.g. Treuer et al. 2017). In most cases, it is less
clear how the heterogeneity in forest condition that is independent of stand age translates
into faunal community composition. However, it stands to reason that profound
85
differences in host plant or fruit resource availability would translate into differences in
faunal abundance or occupancy of second-growth forest patches.
Terrestrial vertebrates in tropical forests present a difficult community to study
because of their elusiveness and low population densities. That is particularly true of
large terrestrial vertebrates, which, in addition to their charismatic nature, also perform
important rolls in tropical forests. Large-bodied herbivores can play important roles in
ecosystem functioning by dispersing large-seeded plant species (Cordeiro et al 2009;
Moran et al. 2009; Sethi and Howe 2009). Top predators are an important group
ecologically, frequently acting as keystone species (Estes et al. 2011). Their absence can
potentially trigger trophic cascades impacting forest regeneration, which have been
documented in a number of forest systems spanning temperate and tropical forest
ecosystems (Estes et al. 2011).
Both large herbivores and top predators have particularly large habitat needs
(Lamb 2014), and because of that deserve special attention in when determining success
of landscape-scale habitat restoration as well as the value of second-growth forests for
achieving conservation goals. While their low population densities have made them a
challenging group to study as indicator species, emerging technologies have opened up
new opportunities for their study. Camera traps help overcome the challenge of observer
effects on shy fauna while multiplying sampling effort of researchers to build meaningful
datasets for rare species (Rowcliffe & Carbone 2008; Tobler et al. 2008). They are a
relatively inexpensive and standardized tool that can continue increasing the scale of
ecological data collection broadly (Estes et al. 2018). More specifically, camera traps can
aid in resolving the usage patterns of second-growth habitat by non-volant mammals,
86
which in Central American forests have been characterized as readily occupying (Romero
et al. 2016) and displaying strongly reduced species richness (Estrada et al. 1994) in
second-growth forest habitats.
Most previous work on second-growth forest habitat usage in the Mesoamerican
biodiversity hotspot has focused on moist forests (Estrada et al. 1994; Romero et al.
2016). Though undoubtedly important from a conservation perspective, these forests have
seen lower conversion rates to other land uses than the tropical dry forests of the same
ecoregion (Portillo-Quintero & Sánchez-Azofeifa 2010).
As arguably the world’s largest tropical forest restoration project (Allen 2001),
Área de Conservación Guanacaste (ACG) in Costa Rica offers an ideal and deeply
heterogeneous (Powers et al. 2009; Castillo-Núñez et al. 2011; Becknell & Powers 2014)
study system for examining the utilization of second-growth habitats that have been
unusually well characterized for any tropical forest, let alone understudied seasonally dry
systems.
In this study, we sought to answer three questions related to terrestrial vertebrates
in this second-growth tropical dry forest system. First, how do species richness, diversity,
and composition of this community vary with forest age and vegetation type? Second,
how does the heterogeneity of these second-growth forests shape the occupancy of this
community? Finally, how do terrestrial vertebrates at risk of extinction fare in second-
growth dry forest?
87
METHODS
Site Description
This study was conducted in Parque Nacional Santa Rosa (SRNP) (10°48′53″N,
85°36′54″W, elevation ~200m), a constituent Sector of Área de Conservación
Guanacaste (ACG) as well as a small portion of the adjacent Sector Santa Elena
(10°55'30"N 85°36'31"W) in northwest Costa Rica. SRNP has an unusual history for a
protected area. It is thought to have been actively managed as a cattle-producing hacienda
for five centuries (Allen 2001) prior to its expansion from a 1,000 ha national monument
in 1971. The mosaic pattern of ranches in the region, in particular the persistence of
pockets of old-growth forest, maintained a high diversity of flora and fauna on a
landscape scale (Janzen 2000). Cattle were fully removed from SRNP by 1978 (Janzen
1988), although a small number of free-ranging horses were present until the early 1990s
(Janzen 2000). Near the Santa Elena Field Station, horses used by park rangers still
occasionally range into the surrounding forests.
Fire suppression in SRNP began in earnest in the mid-1980s, but occasionally
fires have spilled over from neighboring ranches or have been ignited by humans within
the park resulting in variable fire history for many areas of regenerating forest in SRNP.
The result is a complex mosaic of stand ages across the former pastures. Comparing fire
records obtained from meticulous park records (W. Medina Sandoval pers. comm.) with
results of previous vegetation surveys confirms that so long as grasses persist at the time
of fires, few small trees are able to survive the fires when they occur. As a result, time
since most recent fire is a fair approximation of stand age in this system.
88
We established camera traps in areas that have previously been classified as
‘Santa Rosa Tropical Dry Forest’ (SRTDF) and ‘Santa Rosa Oak Forest’ (SROAK)
(Powers et al). The former is characterized by mesic soil conditions and more diverse tree
flora and the latter by rockier, thinner soils with lower cation exchange capacity and high
densities of Quercus oleoides (Becknell & Powers 2014). The climate of this system is
strongly seasonal, with a windy dry season lasting from approximately December to the
middle of May, during which time little if any of the 900 to 2600 mm of annual rainfall
(30 year average of 1765 mm) falls (Arroyo-Mora et al. 2005; Becknell & Powers 2014).
El Niño conditions both preceded our study and persisted for the duration of it, which
resulted in anomalously low rainfall in 2014 of 1112.9 mm (+9.9 mm SD) and 600.8 mm
(+7.6 mm SD) during 2015, corresponding to reduced gross primary productivity of 13%
and 42% respectively (Castro et al. 2018). Fig. 4.1 shows aerial imagery of the study
region.
Sampling Design
Camera trap locations were established immediately adjacent to each of the forty
vegetation survey plots that form a network spanning the Santa Rosa and Santa Elena
sectors of ACG. These plots were established to conduct floristic surveys across a range
of stand ages, soil types, and other suspected sources of floristic heterogeneity. Twelve of
these plots were previously established by researchers from the University of Minnesota
(Powers et al. 2009; Becknell & Powers 2014) while the remaining twenty-eight were
established by Treuer et al (Treuer et al. in review). All plots measured 20 m by 50 m.
The species identity and diameter at breast height (DBH) of all trees >10cm DBH were
89
recorded in 2015. The twelve University of Minnesota plots were split evenly between
SRTDF and SROAK forest patches, with stand ages ranging from 25 to 80 years (Powers
et al. 2009). The latter twenty-eight plots were established entirely in SRTDF locations
with stand ages ranging from 8 to 30 years, all on former cattle pasture. These plots were
sited to capture a range of distances from old-growth forest fragments and were split
between two different treatments, with half established immediately adjacent to former
pasture shade trees and the other half established at least 50 m from such trees in what
was once open pasture. Within these twenty-eight plots, a small number of species of
wind-dispersed trees were found to constitute about 90% of trees in former open pasture,
but less than 50% of trees within 10 m of former pasture shade trees, suggesting that
these trees serve as attractants for seed dispersing animals and foci of animal-dispersed
plant species recruitment in this system. More information about these plots and the
floristic surveys that occured in them can be found in Powers et al. (2009), Becknell &
Powers (2014) and Treuer et al. (in review).
Adjacent to each plot a Bushnell Trophy Cam HD camera trap was affixed to a
tree at approximately 50 cm off the ground, and angled slightly downward so that the top
of the field of view was approximately flush with the horizon. Each site was chosen so as
to provide a relatively clear field of view for 5 m in front of each camera trap. Ground
cover vegetation was cleared with a machete when the traps were installed in June 2015,
and again when they were checked in August 2015. Camera traps were then taken down
in January 2016. Only animals photographed within 5 m of the cameras were recorded in
our dataset.
90
Camera traps were set to low sensitivity to minimize false triggering events due to
wind, and were set to take a burst of three photos when triggered, with a minimum of 10
seconds between triggering events. Particularly at more open sites, there were thousands
of false triggering events apparently caused by wind. The large majority of animals
captured in photos was readily identifiable to species, but in cases where photos were too
blurred by motion or proximity to the camera, no capture event was recorded. To
distinguish between ocelots (Leopardus pardalis) and margays (Leopardus wiedii) expert
advice was solicited from experts on the Facebook group Estudios de Camaras Trampas
en Costa Rica. For two other pairs of congeneric species it proved too difficult to
consistently distinguish all photos, so the sister species were lumped together. These pairs
were the Plain Chachalaca (Ortalis vetula) and the Grey-headed Chachalaca (Ortalis
cinereiceps) and the common opossum (Didelphis marsupialis) and Virginia opossum
(Didelphis virginiana) which were difficult to distinguish because the camera trap flash
generally washed out the color of their pelage, and the distinguishing color patterning on
their tails was often obscured or blurred. The smallest animal included in this analysis
was the spotted skunk (Spilogale putorius), though smaller rodents occasionally triggered
some camera traps. Capture events for ground-feeding birds (great curassow, Crax rubra;
thicket tinamou, Crypturellus cinnamomeus; white-tipped dove, Leptotila verreaux;
chacalaca, Ortalis spp.; crested guan, Penelope purpurascens; and lesser ground-cuckoo,
Morococcyx erythropygus) were included in the analyses, but capture events of birds in
flight or perched were excluded.
Community composition similarity was determined through the non-metric
multidimensional scaling of the matrix of species abundances at all sites. Several metrics
91
of conservation relevance were calculated. These include observed species richness,
evenness (Shannon Index), and an estimate of total diversity (Chao richness estimator), as
well as the incidences of Baird’s tapir (Tapirus bairdii), meso- and large-carnivores.
These values were calculated for the data as a whole as well as for the data after
subsetting it into age classes. These age classes included young forest of less than 25
years in stand age, intermediate forest of between 25 and 40 years in stand age, and old-
growth forest, which was greater than 40 years in age. The conservation metrics were
also calculated for the twenty-eight sites established by Treuer after separating them into
the subset of sites that were adjacent to nuclear trees and those that were greater than 50
m from a nuclear tree (hereafter, ‘Control’ sites).
Fig. 4.1 An aerial image of the study region from 1988 showing light tan pastures and
green and brown forest patches. Purple dots show locations of camera trap sites.
92
Statistical modelling
To better account for the imperfect ability of camera traps to detect all species
within a given patch of forest (i.e. their high false negative rate), we followed the
example of past camera trapping studies (O’Connell & Bailey 2011) and constructed
occupancy models for the 12 species for which we recorded at least ten capture events
across our study. These models provide a framework for separately assessing covariates
of detection (factors that affect the likelihood of a species being detected at a site given
that it does occur there) and covariates of occupancy (factors that shape the likelihood
that a site with a given set of characteristics is occupied by a given species). For each
species we assembled a comprehensive list of candidate models fit to the data using
maximum likelihood estimates and used a model averaging framework to characterize the
importance of variables in shaping detection and occupancy. We utilized models
predicated on the assumption of closure (i.e. that there was no significant migration in
and out of the system), which we felt was justified based on the stable incidence rate of
capture events for these species over our study period, the year round residency of these
species within this system, and the relatively short duration of our eight month sampling
period which excluded the driest parts of the dry season when water stress is highest and
animals might temporarily migrate out of the study region.
Occupancy models are sensitive to ‘zero-inflation’ or over-abundance of non-
detection events in datasets of rare species, but because the camera traps were continually
recording over the entirely study interval, we were able to establish survey sampling
periods by condensing daily presence absence data for each species into 10 and 20 day
93
sampling intervals as needed to allow model convergence (Karanth et al. 2011; Hamel et
al. 2013). We used the occu function in the R package unmarked to construct the models.
We included three covariates of detection in our models based on a priori
reasoning about what might be important for influencing detection probability. First, we
included aggregate rainfall during each sampling window because we thought that rain
might affect activity levels or movement patterns, particularly during this low rainfall
year. Next, we included an index of understory density surrounding the camera trap site,
under the premise that the cleared patches in front of each camera might constitute
differentially attractive spaces to move through depending on how dense the rest of the
understory was. This index ranged from 1 to 4, with the former indicating an open
understory that would allow relatively free movement in any direction and the latter
indicating a person would have difficulty walking in any direction without pushing aside
vegetation or using a machete. Lastly, we included a proxy variable for the proportion of
vegetation in the area immediately surrounding the camera trap positions under the
suspicion that the density of animal dispersed vegetation surrounding the cameras might
relate to the likelihood of an animal passing in front of the camera. This proxy variable
consisted of the number of large trees that appear in aerial photographs of the sites in
1980 within a 50 m radius of camera trap positions. This distance was chosen because of
the intensely localized effect of these trees on regenerating forest composition (Treuer et
al. in review). For camera trap positions that were located in areas that were pasture in the
photographs, the maximum number of large trees within 50 m was 3, and for the
remaining camera traps that were positioned in areas that were closed canopy forest in the
94
photographs, we recorded as 4s, since it was impossible to count the number of large tree
crowns at these sites in the aerial photos.
After accounting for the collinearity of potential site-level variables, we narrowed
down the list of covariates of occupancy in our complete model to just three. The first of
these variables was the stand age of the patch of forest where the camera trap was located
(this variable strongly correlated with total basal area of the adjacent plot next to each
camera trap). The second variable included was the proportion of area within a 500 m
radius of each camera trap that was made up of old growth forest (as determined by
examination of the 1980s aerial photos). The third was the linear distance from each
camera trap site to the nearest road.
RESULTS
A total of 7,255 camera trap-days worth of surveying was conducted at 40 sites,
yielding 2,186 capture events of 26 species of terrestrial vertebrates, including 19 species
of mammals, six species of birds, and two reptile species. There were several camera
failure events (had each trap run continuously for the entire study, there would have been
8,800 camera trap-days), including at least one case of deactivation by a poacher. Slightly
more than half of all capture events (1,136) were for Central American white-tailed deer
(Odocoileus virginianus truei). The only fully terrestrial species thought to occur in our
study area that we did not record at least once were the northern naked-tailed armadillo
(Cabassous centralis), northern raccoon (Procyon lotor), paca (Agouti paca), and hooded
skunk (Mephitis macroura). We captured images of all four of the Central American
95
terrestrial felids, though we only had a single capture event for jaguarundi (Herpailurus
yagouaroundi).
Table 4.1 A summary of camera trapping effort in this study showing divergent patterns
in large and meso-carnivore incidence across stand ages and between sites near and far
from nuclear trees in intermediate growth forest. Differences in Shannon Indices were
significantly different for young forest and intermediate and advanced forest sites,
however differences in richness and Chao richness estimators were not.
Table 4.1 summarizes a number of metrics of conservation relevance across the
entire data set and within different age classes of forest, as well as in the subset of young
and intermediate aged sites that do and do not occur with 50 meters of a nuclear tree. To
account for differential sampling effort across these treatments we constructed rarefaction
curves (Fig. 4.2), and found substantial overlap in 95% confidence intervals for all
subsets of the data, suggesting no significant difference in the species richness between
any two subsets of forest type. However, sampling was insufficient in advanced and
young forest stands to confidently imply an asymptotic species richness for these forest
classes.
Terrestrial vertebrate community composition showed modest convergence in
composition with stand age class (Fig. 4.3). Grey fox (Urocyon cinereoargenteus),
eastern cottontail (Sylvilagus floridanus), and nine-banded armadillo (Dasypus
96
novemcinctus) all were captured at least five times in young or intermediate forest age
classes without a single appearance on an advanced regrowth camera trap.
a)
97
b)
Fig. 4.2 Rarefaction and extrapolation curves for camera trap data after dividing into (a)
subsets of just traps within 50 m of a nuclear tree and those further than 50 m from a
nuclear tree, and (b) all traps sorted by age class (young forest ≤25 years old;
intermediate ≥25 and ≤40 years; old ≥40 years). In neither case was there significant
differences in gamma diversity between treatment types.
Occupancy models for the dozen species with at least ten detection events
converged in all but two cases, tayras (Eira barbara) and nine-banded armadillos
(Dasypusn novemcinctus) However for five species successfully running the models
required condensing observation intervals from 14 days to 28 days. Fig. 4 visualizes the
top model for each species and 95% confidence intervals for the covariates of occupancy
included in the top models.
98
We recorded 62 detection events for T. bairdii. Of these, 60 detection events
occurred at sites adjacent to plots with a stand age of 30 years or less. However, neither
stand age nor the percentage of old growth forest within a 500 m radius of the site
appeared in any of the top five models for T. bairdii occupancy. Three of these models
contained no covariates of site occupancy at all, while the other two only included
distance to nearest road, which had a 95% confidence interval overlapping zero in both
models. In the averaged model, distance to nearest road had a weak positive relationship
with T. bairdii occupancy, while percent old growth forest in the surrounding 500 m had
a modest negative relationship and stand age had a weak positive relationship with tapir
occupancy. None of these variables were significant in the component models at an alpha
level of p < 0.05.
99
Fig. 4.3 A non-metric multidimensional scaling plot based on species composition of
each camera trap site. Color shows age class of the forest and shape of the icons shows
whether the site is within 50 m of a nuclear tree (circles are used to represent forest ≥40
years, which was classified as advanced regrowth and not characterized as either nuclear
or control sites). Older growth forest showed moderate convergence in composition,
though did not significantly diverge from young or intermediate forest in either alpha or
gamma species richness.
100
DISCUSSION
Richness and Diversity
Terrestrial vertebrate species richness did not vary significantly across stand age
classes or between young and intermediate forest sites that varied in their proximity to
nuclear trees. However, species evenness at intermediate sites was double that of young
forest sites; intermediate and advanced stand ages did not differ in species richness.
Second-growth forest patches in areas with a higher diversity and abundance of
animal-dispersed tree species (due to proximity to nuclear trees—see Chapter 1) did not
see significantly higher diversity of terrestrial vertebrates than patches that were
surrounded by higher densities of wind-dispersed species, and had strongly overlapping
NMDS plots suggesting similar overall composition. However, the incidence of large and
meso-carnivores and tapirs varied by a factor of two between these two classes of forest
patches (tapirs and meso-carnivores appeared more frequently at camera trap sites within
50 m of nuclear trees, while large carnivores appeared more frequently at sites that are
dominated by wind-dispersed tree species. Causality is impossible to infer from this
study, and we suggest this surprising finding be subject to further investigation.
Occupancy Models
Our occupancy models overall suggest that stand age is not a strong driver of
second-growth forest occupancy of terrestrial vertebrates in this system. Because these
models are predicated on a binary determinations of occupancy (occupied vs. not
occupied), it is premature to conclude that younger second-growth habitat has equal value
to older growth forest. Indeed, for the three primate species in ACG dry forest, detailed
101
study has suggested that while they were readily occupy most closed-canopy second-
growth forests, they will do so at lower densities than in advanced regrowth (Sorensen &
Fedigan 2000). The same may hold true for several important large terrestrial vertebrate
species. Indeed, the incidence of large and meso-carnivores increased with stand age.
Nonetheless, these young stands provided at least some habitat for the terrestrial
vertebrates. They appear to have genuine value for wildlife, under the assumption that
their presence at these sites provided them with net benefits (which we cannot determine
with the data at hand).
Baird’s Tapir
Baird’s tapir (Tapirus bairdii) deserves particular note among species in the study
area for two reasons. First, they are the largest extant native terrestrial animal and an
important disperser of large-seeded tree species (O’Farrill et al. 2013). They are also the
only large terrestrial vertebrate in these study system that is listed by the International
Union for the Conservation of Nature as endangered. In fact, they were determined to be
the 34th most important conservation case among a thousand recently assessed mammal
species when ranked by phylogenetic uniqueness and global population trends (Isaac et
al. 2007).
T. bairdii populations are declining, but they have also shown an ability to utilize
second-growth forest habitats in moist and montane forests (Foerster & Vaughan 2002;
Tobler 2002; Carbajal-Borges, Godínez-Gómez & Mendoza 2014), suggesting that
habitat or biological corridor creation could be an effective tool for quickly improving
their conservation outlook. To our knowledge, however, no study has addressed whether
102
stand age is a driver of probability of habitat occupancy by T. bairdii a regenerating dry
forest system.
Our results suggest that at the very least there is little reason to think that second
growth forest <40 years of age is inferior to advanced regrowth as tapir habitat. Indeed,
the incidence of tapir capture events was more than five times higher in young and
intermediate forest than more advanced regrowth. This suggests that second-growth dry
forest, like second-growth moist and montane forest, provides important potential habitat
for the species. Given the relative ease of passive restoration efforts in dry forest relative
to wetter tropical forests in ACG (Janzen 2000), passive restoration of dry forests could
be a critically valuable tool for slowing the decline and aiding in the recovery of Baird
tapir numbers globally. This is particularly true in settings like Costa Rica, where hunting
bans are in place and enforced.
Overall, our results suggest that second-growth tropical dry forest has a high
conservation value for terrestrial vertebrates in Costa Rica, and that environmentally-
driven differences observed in floral community composition and diversity (see Chapter
1) do not translate into similarly dramatic differences in terrestrial faunal composition
and diversity. The presence of the endangered tapir in even the youngest forest patches,
as well as the occupancy of a variety of closed canopy second-growth sites by top
carnivores suggests that these regenerating ecosystems have a valuable role in conserving
terrestrial vertebrates. Landscape scale restoration should be regarded as a practical tool
for staving off stochastic extirpations of key species (i.e. staving of ‘relaxation’ or loss of
species to extinction debt), and in doing so prevent trophic downgrading and its attendant
103
consequences for people and ecosystems (Terborgh et al. 2001; Estes et al. 2011). This
finding should buoy conservationists working in Mesoamerican dry forest regions, as
these habitat has shown a rapid ability to regenerate spontaneously so long as fires are
brought under control (see Chapter 1).
104
WORKS CITED
Allen, W. (2001) Green Phoenix: Restoring the Tropical Forests of Guanacaste, Costa
Rica. Oxford University Press, USA.
Arroyo-Mora, J.P., Sanchez-Azofeifa, G.A., Kalacska, M.E.R., Rivard, B., Calvo-
Alvarado, J.C. & Janzen, D.H. (2005) Secondary Forest Detection in a Neotropical
Dry Forest Landscape Using Landsat 7 ETM+ and IKONOS Imagery1. Biotropica,
37, 497–507.
Becknell, J.M. & Powers, J.S. (2014) Stand age and soils as drivers of plant functional
traits and aboveground biomass in secondary tropical dry forest. Canadian Journal
of Forestry Research.
Bowen, M.E., McAlpine, C. a., House, A.P.N. & Smith, G.C. (2007) Regrowth forests on
abandoned agricultural land: A review of their habitat values for recovering forest
fauna. Biological Conservation, 140, 273–296.
Carbajal-Borges, J.P., Godínez-Gómez, O. & Mendoza, E. (2014) Density, Abundance
and Activity Patterns of the Endangered Tapirus Bairdii in One of its Last
Strongholds in Southern Mexico. Tropical Conservation Science, 7, 100–114.
Castillo-Núñez, M., Sánchez-Azofeifa, G. a, Croitoru, a, Rivard, B., Calvo-Alvarado, J.
& Dubayah, R.O. (2011) Delineation of secondary succession mechanisms for
tropical dry forests using LiDAR. Remote Sensing of Environment, 115, 2217–2231.
Castro, S.M., Sanchez-Azofeifa, G.A. & Sato, H. (2018) Environmental Research Letters
Effect of drought on productivity in a Costa Rican tropical dry forest Effect of
drought on productivity in a Costa Rican tropical dry forest. Environ. Res. Lett, 13.
Chazdon, R.L. (2014) Second Growth. University of Chicago Press.
105
Dent, D.H. & Joseph Wright, S. (2009) The future of tropical species in secondary
forests: A quantitative review. Biological Conservation, 142, 2833–2843.
Estes, L., Elsen, P., Treuer, T.L.H., Ahmed, L., Caylor, K., Chang, J., Choi, J.J. & Ellis,
E.C. (2018) The Spatial and Temporal Domains of Modern Ecology. Nature
Ecology & Evolution.
Estes, J.A., Terborgh, J., Brashares, J.S., Power, M.E., Berger, J., Bond, W.J., Carpenter,
S.R., Essington, T.E., Holt, R.D., Jackson, J.B.C., Marquis, R.J., Oksanen, L.,
Oksanen, T., Paine, R.T., Pikitch, E.K., Ripple, W.J., Sandin, S.A., Scheffer, M.,
Schoener, T.W., Shurin, J.B., Sinclair, A.R.E., Soulé, M.E., Virtanen, R. & Wardle,
D.A. (2011) Trophic downgrading of planet Earth. Science (New York, N.Y.), 333,
301–6.
Estrada, A., Coates-Estrada, R. & Meritt, D. (1994) Non flying mammals and landscape
changes in the tropical rain forest region of Los Tuxtlas, Mexico. Ecography, 17,
229–241.
Foerster, C.R. & Vaughan, C. (2002) Home Range, Habitat Use, and Activity of Baird’s
Tapir in Costa Rica1. Biotropica, 34, 423–437.
Gardner, T.A., Barlow, J., Araujo, I.S., Avila-Pires, T.C., Bonaldo, A.B., Costa, J.E.,
Esposito, M.C., Ferreira, L. V, Hawes, J., Hernandez, M.I.M., Hoogmoed, M.S.,
Leite, R.N., Lo-Man-Hung, N.F., Malcolm, J.R., Martins, M.B., Mestre, L.A.M.,
Miranda-Santos, R., Overal, W.L., Parry, L., Peters, S.L., Ribeiro-Junior, M.A., da
Silva, M.N.F., da Silva Motta, C. & Peres, C.A. (2008) The cost-effectiveness of
biodiversity surveys in tropical forests. Ecology letters, 11, 139–50.
Hamel, S., Killengreen, S.T., Henden, J.-A., Eide, N.E., Roed-Eriksen, L., Ims, R.A. &
106
Yoccoz, N.G. (2013) Towards good practice guidance in using camera-traps in
ecology: influence of sampling design on validity of ecological inferences (ed RB
O’Hara). Methods in Ecology and Evolution, 4, 105–113.
Isaac, N.J.B., Turvey, S.T., Collen, B., Waterman, C. & Baillie, J.E.M. (2007) Mammals
on the EDGE: Conservation Priorities Based on Threat and Phylogeny (ed W Reid).
PLoS ONE, 2, e296.
Janzen, D.H. (1988) Management of habitat fragments in a tropical dry forest: growth.
Annals of the Missouri botanical garden, 75, 105–116.
Janzen, D.H. (2000) Costa Rica’s Area de Conservación Guanacaste: A long march to
survival through non-damaging biodevelopment. Biodiversity, 1, 7–20.
Karanth, K.U., Gopalaswamy, A.M., Kumar, N.S., Vaidyanathan, S., Nichols, J.D. &
MacKenzie, D.I. (2011) Monitoring carnivore populations at the landscape scale:
occupancy modelling of tigers from sign surveys. Journal of Applied Ecology, 48,
1048–1056.
Lamb, D. (2014) Large-Scale Forest Restoration. Routledge.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A. & Kent, J. (2000)
Biodiversity hotspots for conservation priorities. Nature, 403, 853–8.
O’Connell, A.F. & Bailey, L.L. (2011) Inference for Occupancy and Occupancy
Dynamics. Camera Traps in Animal Ecology, pp. 191–204. Springer Japan, Tokyo.
O’FARRILL, G., GALETTI, M. & CAMPOS-ARCEIZ, A. (2013) Frugivory and seed
dispersal by tapirs: an insight on their ecological role. Integrative Zoology, 8, 4–17.
Portillo-Quintero, C.A. & Sánchez-Azofeifa, G.A. (2010) Extent and conservation of
tropical dry forests in the Americas. Biological Conservation, 143, 144–155.
107
Possingham, H.P., Bode, M. & Klein, C.J. (2015) Optimal conservation outcomes require
both restoration and protection. PLoS biology, 13, e1002052.
Powers, J.S., Becknell, J.M., Irving, J. & Pèrez-Aviles, D. (2009) Diversity and structure
of regenerating tropical dry forests in Costa Rica: Geographic patterns and
environmental drivers. Forest Ecology and Management, 258, 959–970.
Romero, A., Timm, R.M., Gerow, K.G. & McClearn, D. (2016) Nonvolant mammalian
populations in primary and secondary Central American rainforests as revealed by
transect surveys. Journal of Mammalogy, 97, 331–346.
Rowcliffe, J.M. & Carbone, C. (2008) Surveys using camera traps: are we looking to a
brighter future? Animal Conservation, 11, 185–186.
Sandor, M.E. & Chazdon, R.L. (2014) Remnant Trees Affect Species Composition but
Not Structure of Tropical Second-Growth Forest (ed N Farwig). PLoS ONE, 9,
e83284.
Schlawin, J.R. & Zahawi, R.A. (2008) 'Nucleating’ succession in recovering neotropical
wet forests: The legacy of remnant trees. Journal of Vegetation Science, 19, 485–
492.
Sorensen, T.C. & Fedigan, L.M. (2000) Distribution of three monkey species along a
gradient of regenerating tropical dry forest. Biological Conservation, 92, 227–240.
Terborgh, J., Lopez, L., Nuñez, P., Rao, M., Shahabuddin, G., Orihuela, G., Riveros, M.,
Ascanio, R., Adler, G.H., Lambert, T.D. & Balbas, L. (2001) Ecological meltdown
in predator-free forest fragments. Science (New York, N.Y.), 294, 1923–6.
Tobler, M.W. (2002) Habitat Use and Diet of Baird’s Tapirs (Tapirus bairdii) in a
Montane Cloud Forest of the Cordillera de Talamanca, Costa Rica1. Biotropica, 34,
108
468–474.
Tobler, M.W., Carrillo-Percastegui, S.E., Leite Pitman, R., Mares, R. & Powell, G.
(2008) An evaluation of camera traps for inventorying large- and medium-sized
terrestrial rainforest mammals. Animal Conservation, 11, 169–178.
Treuer, T.L.H., Choi, J.J., Janzen, D.H., Hallwachs, W., Peréz-Aviles, D., Dobson, A.P.,
Powers, J.S., Shanks, L.C., Werden, L.K. & Wilcove, D.S. (2017) Low-cost
agricultural waste accelerates tropical forest regeneration. Restoration Ecology.
109
CHAPTER 5
Leveraging Productivity: An Exploration of the Efficiency and Expediency of a Partial
Closure of High Seas Fisheries
Author List: Timothy L.H. Treuer, T. Wangyal Shawa, Michael Oppenheimer
Corresponding Author:
Timothy L.H. Treuer, [email protected]
110
Abstract:
International management efforts for fisheries in the high seas have a dismal record of
accomplishment at preventing overexploitation of fish stocks, leading a number of
authors to call for a full closure of the high seas to all commercial fishing. While there is
a strong case to be made that the global benefits of such a regime are strongly net positive
in theory (largely through increased catch in Exclusive Economic Zones), crafting such a
treaty faces many challenges. These challenges include the extreme optics of closing 46%
of the Earth’s surface to exploitation in direct contravention of centuries of customary
law and decades of international hard law. Calls for a global ban have also seemingly
overlooked the heavily skewed nature of the distribution of net primary productivity
(NPP) in the high seas, as well as chlorophyll-a, a measurement even more strongly
correlated with fisheries catch potential. This paper fills a gap in the scientific literature
by creating a dataset of long-term measurements of the distribution of both NPP and
chlorophyll-a within just the world’s high seas, in order to better understand how well a
partial closure of the high seas could serve to achieve the goals of a full closure, should
the latter prove untenable. Using modeled data of high seas NPP from 2003-2015 this
analysis shows that just 11.55% of the high seas not covered by sea ice contains 25% of
the NPP, and that the area with the highest decile of productivity falls into 18 compact
regions that disproportionately occur in areas immediately adjacent to existing EEZs. The
results for chlorophyll-a are even more extreme in skew: from 2009-2013, 10% of the
high seas contained 29.61% of the chlorophyll-a. The implications for high seas fisheries
management are discussed. This paper assesses the viability of one promising pathway to
a partial closure regime involving an amendment to the United Nations Fish Stocks
111
Agreement. Particular attention is paid to incentive structures of such an amendment to
ensure ratification by key international players and enforcement mechanisms to ensure
compliance.
Highlights:
Distributions of net primary productivity and chlorophyll-a in the high seas are
skewed.
22.22% of NPP and 29.61% of chlorophyll-a occurs in 10% of the high seas
respectively.
Top decile of productivity falls into 18 compact regions adjacent to EEZs.
An agreement that closes high NPP regions is a tractable coordination problem.
An amendment to UNFSA provides a promising structure for such a regime.
Keywords: High seas; fisheries; skewness; maritime law; NPP; chlorophyll-a; MPAs
112
INTRODUCTION
High seas fisheries
The sustainability of fishing in international waters has been long been questioned
(Selak 1952). These concerns predate even the modern designation of international
waters as ‘high seas’ under the United Nations Convention on the Law of the Sea
(UNCLOS). Indeed, overfishing of stocks beyond national jurisdictions was a significant
motivation for the establishment of Exclusive Economic Zones (EEZs), which define the
boundary of the high seas (Nandan 1987). Despite the relatively recent advent of
technology that has facilitated industrial-scale exploitation of the high seas (Swartz et al.
2010; Cullis-Suzuki and Pauly 2010), these historical concerns of overexploitation
correctly presaged our current crisis of declining or depleted high seas fish stocks. As
little as 10% of the biomass of large predatory fish from pre-industrial times remained in
2003 (Myers & Worm 2003), and there have been precipitous declines in the populations
of certain high seas fish, including pelagic species such as many tunas and mackerels
(Juan-Jordá et al 2011) and slow-growing benthic species such as orange roughy
(Hoplostethus atlanticus), oreo dories (Allocyttus spp., Neocyttus spp. and Pseudocyttus
spp.), and toothfishes (Dissostichus spp.--commonly marketed as ‘Chilean seabass’)
(Maguire et al. 2006). Moreover, the state of high seas fisheries may be worse than they
seem; there simply are not adequate data to assess the population trends for a number of
species currently being exploited in the high seas, but indirect evidence suggests they are
being harvested unsustainably (Maguire et al. 2006).
Belying this seeming lack of management of high seas fish stocks, are the
presence of Regional Fisheries Management Organizations (RFMOs) covering nearly all
113
parts of the high seas. Since 2001, RFMOs have been endowed with the power of
sustainably managing high seas fisheries through the implementation of the United
Nations Fish Stock Agreement (UNFSA). The systematic RFMO failures are well known
(Cullis-Suzuki and Pauly 2010) and have persisted despite clear articulation of best
practices (Lodge et al. 2007; Ban et al. 2010). These failures are perhaps best captured in
the finding that two-thirds of RFMO managed stocks fished on the high seas are either
overexploited or depleted (Cullis-Suzuki and Pauly 2010).
A high seas closure?
Frustration with RFMO management and their slow pace of adopting best
practices has prompted researchers to explore the potential of closing the high seas to
fishing entirely (e.g. Sumaila et al. 2014; White and Costello 2014). Counterintuitively,
some evidence suggests a full closure could significantly increase the profitability, total
catch, and fish biomass simultaneously under a range of stock types. This would be
achieved through a combination of increased catch per unit effort resulting from more
fishing nearshore, spillover effects from well protected populations, and resolution of
game-theoretic challenges inherent in managing stocks that span multiple sovereign
regions (White and Costello 2014). Further, such a closure would serve to address
massive global inequalities in the revenue from transboundary fish harvests, since
relatively few distant water fishing nations (DWFNs) currently reap the lion’s share of
take from the high seas (Sumaila et al. 2014). Finally, a total closure of the high seas to
fishing would certainly satisfy calls to increase the amount of pelagic protected areas--
arguably the least protected ecosystems on Earth (Game et al. 2009). Though a total high
114
seas closure appears sensible, negotiating and implementing one would require a steep
uphill battle.
Three primary challenges hamstring an effective international agreement for
closing the high seas to fishing. First, by definition, the high seas fall outside of any
sovereign jurisdiction. Changes to existing international regulatory regimes require either
new agreements or a consensus amendment to the UNFSA, thus allowing the few
countries responsible for the vast majority of high seas fishing to simply opt out of a
closure unilaterally (to say nothing of the significant challenges inherent in any sort of
multilateral coordination of common-pool resource exploitation, particularly ones with
prisoner’s dilemma dynamics). DWFNs would be likely to do so, because the loss of
access to high seas fisheries would not be counterbalanced by increased catch within their
own EEZs under reasonable assumptions (Sumaila et al. 2014).
Second, illegal, unreported, and unregulated (IUU) fishing is already a major
problem on the high seas despite guaranteed legal access to stocks under the UNFSA (Le
Gallic and Cox 2006; Haward 2004; Agnew et al. 2009). Closure of the high seas to
fishing could significantly increase IUU fishing by exacerbating the existing overcapacity
of the world’s deep water fishing fleet (Pauly et al. 2002; Beddington et al. 2007). Given
the unprecedented nature of the enforcement challenge, the need for novel enforcement
techniques has been acknowledged, even by proponents of a total closure (White and
Costello 2014). It is unclear whether proposed techniques (Erceg 2006; Stokke 2009; Le
Gallic and Cox 2006) can be applied at the scale of 58% of the world’s oceans.
The final major hurdle facing a high seas closure is the most nebulous but may be
the most vexing: the psychological barrier of closing ~42% of the Earth’s surface to
115
renewable resource extraction in direct contravention of hundreds of year of customary
international law. Cullis-Suzuki and Pauly (2010) contend that RFMO attempts to
prevent tragedy of the commons on the high seas are hamstrung by the persistent
paradigm of Grotius’ Mare Liberum (1609)--the idea of freedom of the seas. Efforts to
negotiate and enforce a full high seas closure would no doubt be forestalled to an even
greater extent. The importance of this psychological status quo should not be overlooked
when considering the feasibility of achieving global consensus on a moratorium on
fishing in the high seas.
A partial high seas closure
Calls for a global high seas closure overlook a fundamental characteristic of the
world’s oceans: high heterogeneity in biological activity (Fig. 5.1). Protecting the regions
of the high seas with the highest capacity for producing commercially harvested fish
stocks could mitigate a portion of each of the three primary challenges to a total high seas
closure, while still protecting a disproportionately large number of fish. One
straightforward way to pursue this goal would be to prioritize the closure of areas with
the highest concentrations of chlorophyll-a, which can be remotely sensed from satellite
measurements and correlates well with fish catch potential globally (Stock et al. 2016).
Another strategy would be to prioritize based on rates of net primary productivity (NPP),
which represents the amount of carbon dioxide being fixed into organic carbon through
photosynthesis after accounting for the respiration of photosynthesizing organisms. NPP
is correlated with catch potential (Ware and Thomson 2005; Chassot et al. 2007;
Sherman et al. 2009; Chassot et al. 2010; Friedland et al. 2012), albeit to a far weaker
extent than chlorophyll-a concentrations (Friedland et al. 2012; Stock et al. 2016).
116
Nevertheless, NPP has the advantage of representing an ecologically relevant flux (i.e.
the amount of chemical energy flowing into an ecosystem), and has implications for long-
term carbon storage in the oceans. Halting fishing in regions of the high seas with high
NPP or high chlorophyll-a is likely to have larger positive spillover benefits to fishers
within EEZs than protecting a random, but otherwise equivalent region.
A survey of the existing literature on calls for a partial high seas closure resulted
only in papers that explore or call for implementing new high seas marine protected areas
(MPAs) at a scale commensurate with existing large MPAs (e.g. Corrigan and Kershaw
2008; Game et al. 2009). While useful, such large-scale MPAs (>100,000 km2) would be
dwarfed two orders of magnitude by a 10% high seas closure. Moreover, there have not
been calls in the literature for high seas conservation areas to be prioritized by their NPP
or chlorophyll-a concentrations, or even publically available datasets of productivity
within just the high seas.
This paper seeks to explore the potential of a partial closure of the high seas with
the highest NPP or highest chlorophyll-a by asking three key questions: (1) how are the
most productive regions of the high seas distributed spatially and how much total high
seas NPP or chlorophyll-a would a 10%, 25% or 40% closure capture? (2) To what
degree does the spatial distribution of these regions facilitate or present challenges to
achieving a closure of fishing activities through negotiation of an international
agreement? (3) How could a partial closure of the high seas be achieved given its
distribution?
117
METHODS
Data sources
Data on ocean productivity were obtained from Ocean Productivity (2017). The
data were monthly estimates from January 2003 to December 2015 (inclusive) of ocean
NPP (in mg C/m2/day) for a gridded (1080 x 2160 pixels) equidistant cylindrical
projection of the Earth’s surface using a Vertically Generalized Production Model
(VGPM) following the method of Behrenfeld and Falkowski (1997). Inputs into the
model were MODIS surface chlorophyll concentrations, MODIS 4-micron sea surface
temperature data, and MODIS cloud-corrected incident daily photosynthetically active
radiation. A complete description of the model can be found at
http://www.science.oregonstate.edu/ocean.productivity/vgpm.model.php. Missing values,
often associated with sea ice conditions, were assumed to be zero to not inflate mean
annual productivity estimates. As a sensitivity analysis, zeros were replaced with the
minimum value for the grid cell over the 13 year data span and the analysis rerun. The
results of this sensitivity check were not qualitatively different.
Monthly datasets were downloaded and processed in R version 3.02 (Core Team
2013) to find mean NPP values for each grid cell for the 13 year interval and exported to
ArcMap 10.4.1 (ESRI 2017).
Data on estimated chlorophyll-a concentrations through the world’s oceans were
obtained for 2009 through 2013 from NASA Earth Observations. This dataset was
produced through measurements from the MODIS instruments on the satellites Aqua and
Terra. Documentation on how these concentrations were calculated can be found in Hu et
al. (2012).
118
These data were clipped by masking points occurring over continents, a 150 km
around the continents (this was done because territorial waters are not considered part of
EEZs), and EEZs. The raster layers for the continents and EEZs were downloaded from
ESRI and the World Wildlife Fund respectively. The resultant dataset thus represented
only grid cells with midpoints occurring over the high seas. These data were exported
back to R and a cumulative distribution function was plotted for the data after ranking the
cells in descending order of NPP and chlorophyll-a.
Fig. 5.1. Heat map of marine fishing vessel activity over one year beginning May 24th
2012 based on satellite telemetry of AIS transponders, showing strongly non-uniform
distribution of fishing activity across the high seas. Figure modified from Global Fishing
Watch (2018).
RESULTS
The mean and median values for the high seas dataset were 300.8 and 275.0 mg C/m2/day
respectively, with a range of 1-2033 mg C/m2/day, indicating significant rightward skew
of the distribution of productivity (the highest productivity grid cells were higher than if
119
the distribution were normal with a similar mean and standard deviation) (Fig. 5.2). As a
result only 11.22% of grid cells contained a full quarter of the total NPP of the high seas
(Fig. 5.3).
Fig. 4 shows the distribution of NPP across the high seas with values binned into
deciles. The grid cells in the highest decile of productivity represent 22.22% of total high
seas productivity and were almost entirely grouped into 18 discrete areas of contiguous
points. These zones of high productivity (hereafter NPP hotspots) have a median distance
to the nearest EEZ of less than 150 km.
a)
120
b)
Fig. 5.2. Histograms of the distribution of (a) NPP (mg C/m2/year) and (b) chlorphyll-a on
the high seas
DISCUSSION
Interpretation of the distribution of high seas NPP and chlorophyll-a
The results demonstrate that a partial closure of the high seas has the potential to
capture many of the benefits of a total closure, and thereby serves as an effective and
achievable alternative should negotiation or implementation of the latter ultimately prove
untenable. First, the significant right skew combined with the high variance in the
distribution of NPP and chlorophyll-a in the high seas means more than a quarter of the
fish producing potential of the high seas could likely be protected by closing just 10% of
the total area.
121
Second, the most productive decile of the high seas is strongly, but not completely
underdispersed. The mostly-clumped nature of the highest NPP grid cells seen in Fig. 5.4
is fortuitous since discrete, contiguous regions are easier to implement, manage, and
protect than diffuse patches scattered throughout the high seas. That there are twenty
distinct regions spread across all oceans should also facilitate a partial closure, since the
benefits and burdens are likely to play out more evenly than if all areas occurred in one
ocean alone. This suggests that a closure of these regions ought to be negotiated
simultaneously as a single amendment rather than a piecemeal process that would have
otherwise similar fishing nations from different regions facing divergent costs and
benefits.
Finally, the proximity of the highest productivity decile of the high seas to EEZs
is also beneficial for maximizing the positive impacts of a high sea closure. This allows
for both the dispersal of juveniles from unfished source populations with greater numbers
of large females (that are disproportionately important for producing the next generation
of offspring) into adjacent EEZs, as well as migration of those large females into depleted
EEZs (Pauly et al. 2002).
122
a)
123
b)
Fig. 5.3. a) Cumulative distribution function for high sas NPP based on rank-ordered
Carbon-based Production Model pixels from Ocean Productivity (2017). Red line
indicates percent (11.22%) of pixels needed to capture 25% of total High Seas
productivity. b) CDF for high seas chlorophyll-a with red lines indicated 10%, 25%, and
40% of pixels.
Regime structure
The framework provided by the UNFSA’s amendment process is as practical a
vehicle as any for implementing a partial closure of the high seas, given the agreement’s
established history of transboundary fisheries management. Article 45 of the UNFSA
spells out the process: a nation requests a convention to consider an amendment and if
half of the parties to the agreement acquiesce within sixth months, then the convention is
held. This should not be difficult to achieve, given the relatively small number of nations
124
that actively exploit high seas fisheries, and the large number of states standing to benefit
from any amount of high seas closure (Sumaila et al. 2014).
Amendments to the UNFSA have the option of stipulating for themselves the
number of ratifications needed before it enters into force. In this case the total number of
signatories is less important than getting key players to fall in line. The most important,
but also most challenging, ratifications to obtain will be from distant water fishing
nations, including China, Japan, Spain, and South Korea. Some of these nations struggle
to even remove harmful subsidies for their high seas fishing fleets (Sumaila et al. 2010),
so they would seem unlikely to accede to any closure of the high seas. China would seem
particularly disinclined, given its EEZ would be unlikely to see significant spillover from
newly protected high seas. Bringing DWFNs on board would most likely require some
sort of side payment. These could take the form of a provision in the amendment that
tweaks the allocation of fishing rights by RFMOs, requiring that fewer rights go to
nations experiencing stock gains within their EEZs as a result of closure of high seas
areas, and additional rights be allocated to states harmed by the closures. Since highly
migratory fish that move between various EEZs are often quite lucrative fisheries when
fished locally, and distant water fishing in the high seas in fact has low profitable to begin
with (Sumaila et al. 2010), it should not in theory take tremendous modification to the
allocation formulae to incentivize a self-interested rational actor to agree to such terms.
The exact balance of catch allocations by RFMOs would require considerable
negotiation.
125
a)
126
b)
Fig. 5.4. a) Distribution of average annual Net Primary Productivity (NPP) across the
high seas from 2003-2015 using a Vertically Generalized Production Model (VGPM)
(Behrenfeld and Falkowski, 1997; source data from Ocean Productivity (2017)) showing
a strongly skewed distribution between high productivity and low productivity regions of
the High Seas. The darkest color shows the ten percent of the High Seas with the highest
NPP over the period. b) Distribution of chlorophyll-a from 2009-2013 in the high seas.
Enforcement
Two additional challenges to implementing a partial high seas closure amendment
to the UNFSA merit attention. First, while a partial closure may be easier to enforce than
a total closure, it is still impractical to consider patrolling 10% of the high seas (~4% of
the Earth’s surface). Ensuring that a partial closure of the high seas does not lead to a
spike in IUU fishing requires new monitoring and enforcement techniques. One
127
promising avenue involves satellite based tracking of individual vessels based on their
onboard automatic identification systems (AIS).
Recently, the NGO Oceana, Google, and the Leonardo DiCaprio Foundation
launched a joint venture called Global Fishing Watch aimed at curbing IUU fishing
through tracking of AIS transponders (Gibbs 2014). AIS transponders are required by the
International Maritime Organization on all fishing vessels with a gross tonnage of 300 or
more traveling in international waters, and can be monitored by satellites (Figure 5.1a).
This system is well suited to track large fishing vessels as they fish illegally either by
noting when they are engaged in movement patterns indicative of fishing in off-limits
areas, or by noting when transponders are turned off or scrambled as a vessel approaches
an area where it is not allowed to fish (Gibbs 2014). While this system has a blind spot
for smaller vessels that lack transponders entirely, this does not include many vessels that
would be fishing in the rough conditions of the deep water high seas. This system could
also be potentially augmented by visual tracking using high resolution satellite imagery
and deep learning algorithms to facilitate processing of enormous datasets.
Beyond preventing IUU fishing, a partial high seas closure regime would also
need a way to prevent rogue legal fishing by individual firms. Such a scenario could take
place by firms adopting a so-called ‘Flag-of-Convenience’. By paying a nominal fee,
fishing vessels could fly a flag of a state that failed to sign on to the UNFSA amendment
and thereby legally be allowed to fish in closed high seas areas. Such circumvention
could rapidly lead to the disintegration of the agreement. Preventing this could be
achieved through a provision in the amendment requiring nations to deny landing of fish
at ports under their jurisdiction to vessels flying the flags of non-signatory nations.
128
Depending on the capacity of the tracking system discussed above, denial of landing
rights could be limited to individual vessels rather than all vessels flying a particular flag.
Transshipment of fish may need to be curtailed if it is used to ‘launder’ illegal catch from
the high seas, but such measures are already being implemented in the ports of many
developed nations (Swan 2006; Stokke 2009). Furthermore measures in the amendment
requiring additional, relatively painless changes to docking processes for developing
nations to curtail transshipments (Swan 2006) should be palatable to developing nations
given they stand to benefit the most from a partial closure regime (Sumaila et al. 2015).
Limitations and next steps
There are several features of a regime to partially close the high seas that remain
to be assessed or investigated before any particular implementation of a closure would be
prudent. The first is the use of bioeconomic models (a la Sumaila et al. 2015) to simulate
the end effects of various permutations of a partial high seas closure on global fisheries
profits, catch, and stocks. NPP is not a perfect proxy for fish biomass production
potential, and such models provide a strong independent objective criteria for prioritizing
high seas regions for closure. Further, both the bioeconomic models and models of NPP
should be run under a range of climate scenarios to ensure that productivity hot spots do
not shift substantially with projected warming in sea surface temperatures driven by
global climate change.
Finally, it is important to note that while a partial high seas closure could be
strongly beneficial for achieving a larger pie split more evenly between the nations of the
world, it will not come close to resolving the global overfishing crisis. 90% of wild
129
caught marine fish are caught within EEZs, and many of the most significantly
overexploited high seas fish are highly migratory species that move between the high
seas and EEZs. Even a total closure of the high seas would not prevent a species like
bluefin tuna from being overexploited, without considerable improvement to RFMO
practices and within-EEZ reforms. A partial high seas closure, however, seems to
represent a step in the right direction.
Conclusion
This study highlights the potential for a partial closure of the high seas through an
amendment to the UNFSA to provide significant global benefits to fishing nations, while
forestalling further degradation of high seas fish stocks and allowing recovery of
damaged ecosystems and depleted stocks. By focusing on areas of high NPP, such an
agreement would overcome many of the challenges preventing a full high seas closure,
while still delivering the greatest benefits possible through the creation of what would
amount to twenty massive scale marine protected areas covering all regions of the high
seas with the exception of the Central Arctic.
130
WORKS CITED
Agnew DJ et al. (2009) Estimating the Worldwide Extent of Illegal Fishing Sandin, SA,
editor. PLoS ONE 4:e4570
Ban NC et al. (2014) Systematic Conservation Planning: A Better Recipe for Managing
the High Seas for Biodiversity Conservation and Sustainable Use. Conservation
Letters 7:41–54
Beddington JR, Agnew DJ, Clark CW (2007) Current Problems in the Management of
Marine Fisheries. Science 316
Behrenfeld MJ, Falkowski PG (1997) Photosynthetic rates derived from satellite-based
chlorophyll concentration. Limnology And Oceanography 42
Core Team (2013) R v. 3.0.3.
Cullis-Suzuki S, Pauly D (2010) Failing the high seas: A global evaluation of regional
fisheries management organizations. Marine Policy 34:1036–1042
Erceg D (2006) Deterring IUU fishing through state control over nationals. Marine Policy
30:173–179
Ewell C et al. (2017) Potential ecological and social benefits of a moratorium on
transshipment on the high seas. Marine Policy 81:293–300
Gallic B Le, Cox A (2006) An economic analysis of illegal, unreported and unregulated
(IUU) fishing: Key drivers and possible solutions. Marine Policy 30:689–695
Global Fishing Watch (2017) Introducing Global Fishing Watch.
http://globalfishingwatch.org/
Grotius H (1609) Mare liberum sive de jure quod Batavis competit ad Indicana
commercia dissertatio.
131
Haward M (2004) IUU Fishing: Contemporary Practice, Oceans management in the 21st
century: institutional frameworks and responses. In: Elferink, AO & Rothwell, D,
editors. Martinus Nijhoff, Leiden pp. 87–106.
IMO Resolution A.917(22) Guidelines for the onboard operational use of shipborne
automatic identification systems (AIS).
Juan-Jordá MJ et al. (2011) Global population trajectories of tunas and their relatives.
Proceedings of the National Academy of Sciences of the United States of America
108:20650–5
Kulmala S, Laukkanen M, Michielsens C (2008) Reconciling economic and biological
modeling of migratory fish stocks: optimal management of the Atlantic salmon
fishery in the Baltic Sea. Ecological Economics
Leenhardt P et al. (2013) The rise of large-scale marine protected areas: Conservation or
geopolitics? Ocean and Coastal Management
Lodge MW et al. (2007) Recommended Best Practices for Regional Fisheries
Management Organizations. London
Maguire JJ et al. (2006) The state of world highly migratory, straddling and other high
seas fishery resources and associated species. Rome
MarineTraffic (2017) MarineTraffic: Global Ship Tracking Intelligence | AIS Marine
Traffic. https://www.marinetraffic.com/
Myers RA, Worm B (2003) Rapid worldwide depletion of predatory fish communities.
Nature 423:280–283
Nandan SN (1987) The Exclusive Economic Zone: A Historical Perspective.
132
Ocean Productivity (2017) Ocean Productivity: Online VGPM Data.
http://orca.science.oregonstate.edu/1080.by.2160.monthly.xyz.vgpm.m.chl.m.sst.
php
O’Leary BC et al. (2012) The first network of marine protected areas (MPAs) in the high
seas: The process, the challenges and where next. Marine Policy 36:598–605
Pauly D et al. (2002) Towards sustainability in world fisheries. Nature 418:689–695
Pikitch EK et al. (2004) Ecosystem-Based Fishery Management. Science 305
Selak CB (1952) The Proposed International Convention for the High Seas Fisheries of
the North Pacific Ocean on JSTOR. American Journal of International Law
46:323–330
Stock CA et al. (2017) Reconciling fisheries catch and ocean productivity. Proceedings of
the National Academy of Sciences of the United States of America 114:E1441–
E1449
Stokke OS (2009) Trade measures and the combat of IUU fishing: Institutional interplay
and effective governance in the Northeast Atlantic. Marine Policy 33:339–349
Sumaila UR et al. (2015) Winners and losers in a world where the high seas is closed to
fishing. Scientific Reports 5:8481
Sumaila UR et al. (2010) Subsidies to high seas bottom trawl fleets and the sustainability
of deep-sea demersal fish stocks. Marine Policy 34:495–497
Swan J (2006) Port State Measures to Combat IUU Fishing: International and Regional
Developments. Sustainable Development Law & Policy 7
Swartz W et al. (2010) The Spatial Expansion and Ecological Footprint of Fisheries
(1950 to Present) Sandin, SA, editor. PLoS ONE 5:e15143
133
United Nations (1995) The United Nations Agreement for the Implementation of the
Provisions of the United Nations Convention on the Law of the Sea of 10
December 1982 relating to the Conservation and Management of Straddling Fish
Stocks and Highly Migratory Fish Stocks.
White C et al. (2014) Close the High Seas to Fishing? PLoS Biology 12:e1001826
Worm B et al. (2009) Rebuilding Global Fisheries. Science 325
134
CONCLUSION
Title: Ecological Kintsugi: Scaling up Habitat Restoration Efforts by Shifting Paradigms
Author List: Timothy L.H. Treuer
Kintsugi is the Japanese art of repairing broken ceramics using adhesive lacquer
containing powdered gold, silver, or platinum (Fig. 1). Practitioners fit shards of broken
pots back together with the metallic glue. The art form embraces wabi-sabi, a
philosophy-infused tradition of Japanese aesthetics that embodies the Buddhist belief in
the impermanence, suffering, and non-selfness of all things. Kintsugi recreates the lost
function of an object, but it does so by gilding the scars of past damage, rather recreating
an original, blemish-free piece.
If we hope to achieve our biggest conservation goals—in particular, increasing the
scale of habitat resuscitation—we need to embrace ecological kintsugi.
135
Fig. 1 This fractured ceramic bowl was repaired through kintsugi. Photo credit
Haragayato.
The classic practice of habitat restoration starts by picking a baseline or reference
ecosystem and set it as a target toward which restorationists must set their aim. Wise
hands in the discipline, such as Clewell and Aronson (2007), have laudably counseled
practitioners that they must think of themselves not as painters recreating a static image,
but as doctors giving a patient the tools to self-heal. However, too often the emphasis of
restoration ecology is on one-fell-swoop interventions, particularly when plans are
developed through the gauze of bureaucracy. Even the term ‘restoration’ invites the
mentality of recreating what once was.
This mentality creates intertwining problems. It emphasizes working backwards
from a final state, too often tempting practitioners into pursuing overly engineered
136
strategies that seek an ultimate solution in a single leap. It also encourages a myopic
focus on the restored area itself, generally failing to emphasize the role of existing pieces,
like pockets of remaining habitat, or the desirable organisms that are already present in
the degraded system. What’s more, the references states are often quixotic targets,
unachievable in a world of changing climate, ubiquitous pollution, or extinct species.
The overarching issue is not that ecological restoration is harmful, but that the
existing paradigm encourages a troublingly inefficient use of limited resources for
conservation. In most cases, the gold-standard practices of today’s restoration ecology
cannot scale up to meet the conservation challenges we face without a dramatic and
unlikely increase in funding. Carpeting a degraded forest landscape in seedlings is
expensive, even where labor is relatively cheap. Even in Indonesia, the median price of
restoring a hectare of forest through plantation-style plantings of native seedlings is more
than $2,000. At those prices, restoring the amount of tropical forest lost annually to
deforestation and degradation would cost tens of billions of dollars a year, and that’s
without any investment in land rights or mitigation of the conditions that drove the forest
loss in the first place.
The tenets of ecological kintsugi
The metaphor of kintsugi contains three valuable lessons for shifting the paradigm
of restoration ecology to scale-up the landscape transformations that it can achieve. The
first lesson is that recovery starts by leveraging the literal and figurative fragments of the
original habitat. Heavily fragmented landscapes often retain high numbers of species,
particularly when loss of the matrix between the fragments was recent. They also can act
137
as propagule sources, creators of favorable microclimates, or otherwise serve as nuclei of
natural regeneration. Passive regeneration can often be achieved through these biological
resources alone.
The degraded lands themselves can also be seen as figurative fragments for
ecological kintsugi. Restoration too often takes for granted what is present in a degraded
landscape. Following massive beetle kill, some pine forests in Western North America
are clear-cut by salvage loggers, removing the surviving trees that may harbor genetic
resistance to the insect pests. In the tropics, some restoration projects start by clearing
woody regrowth along with invasive grasses and bracken ferns, despite their potential to
facilitate the dispersal or survival of later successional stage species. Despite this, a
recent meta-analysis found that restoration projects in tropical forests that relied on
entirely passive restoration methods had more positive outcomes than those that involved
active replanting of seedlings.
The second lesson of ecological kintsugi is to treat material beyond the original
fragments as an expensive addition to invest in judiciously. When possible, funding
should be spread over a large area using less intensive techniques that harness a systems
natural capacity to recover. More exhaustive and expensive efforts should be saved for
when they are highest impact, such as establishing biological corridors or restoring
critical ecosystem services. As Crouzeilles et al. (2017) suggest, more expensive and
intensive interventions may not yield superior results for tropical forests. In disturbed
systems, hysteretic alternative stable states appear relatively rare; in most cases when the
forces causing disturbances are removed, ecosystems are able to recover many of their
original properties on their own. Two common exceptions are fire-maintained alternative
138
stable states in terrestrial systems and alternative states generated by the local loss of key
species. Fortunately, medium-term fire management and reintroductions are typically
inexpensive on a landscape scale, relative to many other restoration techniques.
The final lesson of ecological kintsugi is to focus on reestablishing integrity and
function of the system, rather than a particular appearance or composition. Kintsugi
centers on acknowledging and embracing the traumas and history of an object while
restoring its usefulness. Recreating exactly what was lost is an impossible task and a
distraction from the myriad other victories that come through revitalizing degraded
ecosystems. Long before an ecosystem is fully restored, it is likely to recover many of its
valuable functions, such as providing habitat for endangered species, storing carbon,
stabilizing soil, and serving as a repository for genetic diversity. Perhaps most
importantly of all, a mentality of ecological kintsugi lets us celebrate the beauty of our
own efforts to heal the world we broke, regardless of what flaws remain. Enthusiasm for
environmental causes wanes when tasks are overwhelming or seemingly hopeless.
Ecological kintsugi offers a more robust permission structure than reference-state
restoration to toast our successes.
Scaling up by enhancing resilience
This dissertation explored several examples ecological kintsugi thinking, and the
increased scale of ecosystem recover that it can engender. In Chapters 1 and 4, it
documented the landscape transformation of the former cattle haciendas of seasonally-
dry Guanacaste. They showed that stopping the fires was enough of a biological
intervention to put the mosaic dry forest system on an irresistible path to closed canopy
139
cover of a diverse forest. By revealing the importance of nuclear trees, Chapter 1 also
suggested an excellent opportunity for a high impact but relatively easy intervention to
ensure that the trajectory of recovery starts off in the best way possible, namely ensuring
areas slated for restoration have sufficient attractants to seed dispersing animals.
Chapter 2 did not even leave ACG to offer another blueprint for how to scale up
our efforts to repair damaged ecosystems. The application of orange waste to 3 ha of
transitional wet-dry forest in ACG revealed just how much a simple low-cost intervention
could accelerate recovery of a degraded tropical system. Even if it was tragically
unreplicated both in ACG because of the lawsuit (see Chapter 2) and in other settings, it
still serves as a demonstration of how much damage can be undone without resorting to
expensive restoration approaches. Chapter 3 explores this idea further, articulating where
and how organic wastes may be useful as a tool for pushing a degraded tropical forest
system out of an alternative stable state maintained by a synergy of fire, depleted soils,
and invasive vegetation.
Another corner of ACG harbors an additional model for ecological kintsugi for
tropical forests. To re-establish a biological corridor between the cloud and rain forests of
two areas of the conservation area (the relatively intact forests of the volcanoes Rincon de
la Vieja and Cacao), park staff planted large blocks of gmelina trees (Gmelina arborea)
to act as a nurse crop. They shaded out invasive grass, and provided favorable
microclimates for native seedlings and saplings. Gmelina seemed like an unlikely
candidate to be a habitat restoring workhorse, as it has long been villainized for its role as
a pulp tree grown in plantations planted on cleared rainforests.
140
Beyond ACG, examples of ecological kintsugi also abound. Chapter 5 highlights
the potential for unlocking habitat recovery at an entirely unprecedented scale, charting
the potential for the recovery of 4%, 10% or even 16% of the Earth’s surface via
protection of stretches of the high seas. In Gorongosa National Park in Mozambique, the
vast majority of mammalian life disappeared during two decades of civil strife, but now
many species and much of the original ecosystem functioning has been restored thanks to
a few targeted reintroductions and a major push to end poaching. Recently leopards were
spotted in the park, suggesting that active reintroduction of species may not be entirely
necessary.
Stopping the persecution of keystone species has also proven sufficient to break
out of alternative degraded states in two diverse marine systems. Across the Pacific
Northwest, the recovery of sea otters from hunting for their fur has flipped many locales
from urchin barrens to kelp forests. In Hawaii, a site called Kahekili on the island of
Maui has demonstrated that just stopping the harvest of herbivorous fish could
significantly increase coral cover in less than a decade.
These examples of ecological kintsugi highlight the power of relatively mild
interventions that can unleash the resilience of nature in both terrestrial and marine
settings. So long as we exist in a world with limited resources for resuscitating degraded
habitats, these approaches offer the promising avenues for achieving our conservation
goals while revitalizing the ecosystem services we depend on.
141
WORKS CITED
Clewell, A. & Aronson, J. (2008) Ecological Restoration. Island Press.
Crouzeilles, R., Ferreira, M.S., Chazdon, R.L., Lindenmayer, D.B., Sansevero, J.B.B.,
Monteiro, L., Iribarrem, A., Latawiec, A.E. & Strassburg, B.B.N. (2017) Ecological
restoration success is higher for natural regeneration than for active restoration in
tropical forests. Science Advances, 3, e1701345.
Graham, V., Laurance, S.G. & Grech, A. (2016) Environmental Research Letters A
comparative assessment of the financial costs and carbon benefits of REDD+
strategies in Southeast Asia Related content Spatially explicit estimates of forest
carbon emissions, mitigation costs and REDD+ opportunities in Indonesia.