ChapterPDF Available


We synthesize findings from the world’s largest and longest-running experimental study of habitat fragmentation, in central Amazonia. Over the past 36 years, 11 forest fragments ranging from 1 ha to 100 ha in size have experienced a wide array of ecological changes. Edge effects have been a dominant driver of fragment dynamics, strongly affecting forest microclimate, tree mortality, carbon storage and fauna. The matrix of vegetation surrounding fragments has changed markedly over time (evolving from large cattle pastures to mosaics of abandoned pasture and secondary regrowth forest), and this, in turn, has strongly influenced the dynamics of fragments and faunal communities. Both rare weather events and apparent global-change drivers have significantly influenced forest structure and dynamics across the entire study area, both in forest fragments and in nearby intact forest. Such large-scale drivers are likely to interact synergistically with habitat fragmentation.
The text that follows is a PREPRINT
O texto que segue é um PREPRINT.
Please cite as:
Favor citar como:
Laurance, W.F.; J.L.C. Camargo, P.M.
Fearnside, T.E. Lovejoy, G.B.
Williamson, R.C.G. Mesquita, C.F.J.
Meyer, P.E.D. Bobrowiec, & S.G.W.
Laurance. 2016. An Amazonian forest
and its fragments as a laboratory of
global change. pp. 407-439. In: L. Nagy,
B. Forsberg, P. Artaxo (eds.) Interactions
Between Biosphere, Atmosphere and
Human Land Use in the Amazon Basin..
Springer (Ecological Studies 227), Berlin,
Alemanha. 478 pp.
doi (whole book): 10.1007/978-3-662-49902-3_16
ISBN (Hard Cover): 978-3-662-49902-3;
ISBN (Ebook): 978-3-662-49900-9.
Copyright: Springer Science + Business Media
The original publication is available from:
A publicação original está disponível de:
Tiergartenstraße 17
69121 Heidelberg
An Amazonian forest and its fragments as a 1
laboratory of global change 2
William F. Laurance1,*, José L. C. Camargo2, Philip M. Fearnside3, Thomas E. Lovejoy2,4, 4
G. Bruce Williamson5, Rita C. G. Mesquita6, Christoph F. J. Meyer2,7, Paulo E. D. 5
Bobrowiec2,8 & Susan G. W. Laurance1,* 6
7 1Centre for Tropical Environmental and Sustainability Science (TESS) and College of Marine 8
and Environmental Science, James Cook University, Cairns, Queensland 4878, Australia, 9
*corresponding authors: and 10 2Biological Dynamics of Forest Fragments Project, National Institute for Amazonian 11
Research (INPA) and Smithsonian Tropical Research Institute, Manaus, AM 69067-375, 12
Brazil, 13 3Department of Environmental Dynamics, National Institute for Amazonian Research 14
(INPA), Manaus, AM 69067-375, Brazil, 15 4Department of Environmental Science and Policy, George Mason University, Fairfax, 16
VA 22030, USA, 17 5Department of Biological Sciences, Louisiana State University, Baton Rouge, LA, 70803, 18
USA, 19 6Long-term Ecological Research (PELD), National Institute for Amazonian Research (INPA), 20
Manaus, AM 69067-375, Brazil, 21 7Center for Environmental Biology, University of Lisbon, Campo Grande C2, 1749-016 22
Lisbon, Portugal; E-mail: 23 8Coordination of Biodiversity, National Institute for Amazonian Research (INPA), Manaus, 24
AM 69067-375, Brazil, 25
Abstract 26
We synthesize findings from the world’s largest and longest-running experimental study of 27
habitat fragmentation, in central Amazonia. Over the past 36 years, 11 forest fragments 28
ranging from 1-100 ha in size have experienced a wide array of ecological changes. Edge 29
effects have been a dominant driver of fragment dynamics, strongly affecting forest 30
microclimate, tree mortality, carbon storage, and fauna. The matrix of vegetation surrounding 31
fragments has changed markedly over time (evolving from large cattle pastures to mosaics of 32
abandoned pasture and secondary regrowth forest), and this, in turn, has strongly influenced 33
the dynamics of fragments and faunal communities. Both rare weather events and apparent 34
global-change drivers have significantly influenced forest structure and dynamics across the 35
entire study area, both in forest fragments and in nearby intact forest. Such large-scale drivers 36
are likely to interact synergistically with habitat fragmentation. 37
Keywords: Amazonia, Biodiversity, Climate change, Edge effects, Ecosystem services, 39
Environmental synergisms, Global change, Habitat fragmentation, Rainforest 40
1. Introduction 43
The Biological Dynamics of Forest Fragments Project (BDFFP) is the world’s largest and 44
longest-running experimental study of habitat fragmentation1 (Box 1). Located in central 45
Amazonia near the city of Manaus, the BDFFP has evolved since its inception in 1979 into a 46
major epicenter for long-term research. The BDFFP’s research mission has gradually 47
broadened to include not only long-term studies of forest fragmentation but also important 48
work on global-change phenomena and a variety of basic research topics. 49
Here we highlight some key contributions of this singular project to the study of land-50
use change and regional- and global-scale drivers in central Amazonia, at the heart of the 51
world’s largest tropical forest. 52
1.1 Amazonia and global change 54
Amazonia stands at the intersection of several key questions for global change, both for study 55
and for action. It is believed to be one of the regions that will be most impacted by projected 56
climate changes (Salazar et al. 2007, Dai 2012, IPCC 2013). It has the potential to contribute 57
significantly to efforts to mitigate climate change during the narrow window of time that we 58
have to avert ‘dangerous’ global warming (Fearnside 2000, 2012). It is also one of the places 59
where sharply reducing greenhouse-gas emissions—by limiting forest loss and degradation—60
could deliver the greatest global benefits for humankind (Stickler et al. 2009). 61
The rapid loss and fragmentation of old-growth forests are among the greatest threats 62
to tropical biodiversity (Lovejoy et al. 1986, Sodhi et al. 2004, Laurance and Peres 2006, 63
Gibson et al. 2011). More than half of all remaining tropical forest occurs in the Amazon 64
Basin, which is being seriously altered by large-scale agriculture (Fearnside, 2001a, Gibbs et 65
al. 2010), industrial logging (Asner et al. 2005), proliferating roads and energy infrastructure 66
(Laurance et al. 2001a, Fearnside 2002, 2007, Killeen 2007), increasing biofuel production 67
(Butler and Laurance 2009), and oil, gas, and mining developments (Finer et al. 2008). 68
The exploitation of Amazonia is driving forest fragmentation on a vast spatial scale. 69
By the early 1990s, the area of Amazonian forest that was fragmented (<100 km2) or 70
vulnerable to edge effects (<1 km from edge) was over 150% greater than the area that had 71
been deforested (Skole and Tucker 1993). From 1999 to 2002, deforestation and logging in 72
Brazilian Amazonia respectively created ~32,000 and ~38,000 km of new forest edge 73
annually (Broadbent et al. 2008). Prevailing land uses in Amazonia, such as cattle ranching 74
and small-scale farming, produce landscapes dominated by small (<400 ha) and irregularly 75
shaped forest fragments (Cochrane and Laurance 2002, Broadbent et al. 2008). Such 76
fragments are highly vulnerable to edge effects, fires, and other deleterious consequences of 77
forest fragmentation (Laurance et al. 2002, Barlow et al. 2006, Cochrane and Laurance 2008).78
While model predictions for future climate in Amazonia vary considerably, it is 79
expected that parts of the region will be hotter and drier under expected global warming (Dai 80
2012, IPCC 2013). What this warming portends for affected areas of Amazonian forest is a 81
matter of some controversy. Disastrous die-offs projected by the UK Meteorological Office 82
Hadley Centre at atmospheric CO2 concentrations about twice those in the pre-industrial 83
atmosphere (Cox et al. 2000, 2004) have now been countered by a new model version from 84
the same group indicating the Amazon forest almost entirely intact, even with up to four times 85
the pre-industrial CO2 concentration (Cox et al. 2013, Good et al. 2013, Huntingford et al. 86
2013). The main difference is inclusion of CO2-fertilization effects, making the trees grow 87
faster, resist stress better, and close their stomata more frequently such that they use and need 88
less water. 89
1.2 Contributions of the BDFFP to global-change research 91
The BDFFP, with 36 years of research in fragmented and continuous forest in central 92
Amazonia (Box 1; Fig. 1), has been contributing to quantifying the interactions of land use 93
and global climate change. BDFFP studies have assessed the vulnerability of the forest to 94
changes in meteorological parameters (Laurance et al. 2009a), including those that are 95
aggravated by fragmentation (Laurance 2004). The long-term monitoring of thousands of 96
individual forest trees, and of populations of various other plant and animal species in the 97
same locations, provides the potential for early detection of global environmental changes. 98
The BDFFP is a source of invaluable long-term datasets. These include high-quality 99
estimates of Amazon forest biomass and carbon stocks (Phillips et al. 1998, Baker et al. 100
2004). The project also contributes greatly to knowledge of the diversity of species and their 101
relationships in an Amazon forest ecosystem (Laurance et al. 2010a, ter Steege et al. 2013). 102
Biodiversity and ecosystem processes represent part of what is lost when the forest is 103
destroyed or degraded, whether by direct human action, by climate change, or by the 104
interaction of both together. Understanding these processes is essential for assessing not only 105
the vulnerability of forests, but also their potential resilience in the face of global change and 106
the rate of recovery following perturbation (Williamson et al. 2014). 107
The BDFFP has made a substantial contribution to debate over climatic influences on 108
the Amazon via its monitoring of lianas and forest dynamics (Laurance et al. 2014a, 2014b). 109
Lianas evidently make better use of rising CO2 than do trees (e.g., Condon et al. 1992), and 110
contribute significantly to tree damage and mortality (Ingwell et al. 2010). They also form 111
heavily vine-dominated ‘liana forests’ in drier parts of Amazonia (Fearnside 2013). BDFFP 112
plots show a marked increase in liana abundance and biomass between censuses in 1997-99 113
and 2012 (Laurance et al. 2014). Liana increases have also been found in tropical forests in 114
western Amazonia, central America, the Guianas, and elsewhere, with rising CO2 levels being 115
one of the more likely explanations (see Laurance et al. 2014 and references therein). This 116
negative effect of CO2 enrichment is not included in the Hadley Centre models, and would 117
likely cancel out some of the benefits indicated in a high-CO2 future. 118
BDFFP data have helped to identify the direct effects of a warmer, drier climate on the 119
forest. The microclimate at forest edges is significantly hotter and drier than that in the 120
continuous forest (Kapos 1989, Kapos et al. 1993, Camargo and Kapos 1995). Canopy trees 121
are vulnerable to changing microclimates on forest edges during the dry season, with 122
desiccation detected up to 2 km from clearings (Briant et al. 2010). At the BDFFP, edge-123
associated tree mortality and ‘biomass collapse’ have been extensively documented (Laurance 124
et al. 1997, 1998, 2000, Nascimento and Laurance 2004). Because the entire forest can be 125
expected to face comparable conditions under projected climate change, the dead trees in the 126
BDFFP fragment edges stand as a clear warning of the power of these changes. 127
Better estimates of how the forest will fare under changed climate are essential for 128
many reasons, including providing the scientific basis needed to convince both world leaders 129
and the general public that containing climate change is worth the cost. But just as basic is the 130
question: what should we do about climate change once the world finally decides to act? The 131
role of tropical forests is critical to this debate, as they contain a large stock of carbon that 132
could either be substantially released by deforestation, logging, and fire, or conserved for their 133
crucial environmental values. The ways that avoiding these emissions could be incorporated 134
into global mitigation efforts, how carbon benefits would be rewarded, and how they should 135
be calculated have been the subject of longstanding controversy dividing environmental 136
groups, national governments, and scientists (see Fearnside 2001b, 2012). 137
One aspect of this discussion to which the BDFFP makes an important contribution is 138
in reducing the uncertainty surrounding biomass and carbon-stock estimates for Amazon 139
forest. The BDFFP tree survey is much more complete than, for example, the 3000 1-ha plots 140
surveyed by the RADAMBRASIL Project (Nogueira et al. 2008). The BDFFP has much 141
better species identifications and includes data on other forest components, such as palms, 142
lianas, strangler figs, understory vegetation, and dead vegetation (necromass). Correct species 143
identification allows better matching with plant functional traits such as wood density and tree 144
form (e.g., Fearnside 1997, Nogueira et al. 2005, 2007, Chave et al. 2006). 145
Crucially, the BDFFP forest data allow one to see the variability in biomass from one 146
hectare to another. The mean aboveground biomass of live trees across 69 1-ha plots was 356 147
± 47 Mg ha-1 (Laurance et al. 1999). This great variability indicates the need for many plots, 148
rather than relying on only a few plots of 1 ha or less scattered around the region as the basis 149
for calibrating satellite imagery for biomass mapping and for estimating greenhouse-gas 150
emissions from deforestation (see Fearnside 2014). 151
2. Long-term studies of forest fragmentation 153
The BDFFP’s original mission focuses on assessing the effects of forest fragmentation on 154
Amazonian forests and fauna, and on important ecological and ecosystem processes. Here we 155
summarize some key conservation lessons that have been gleaned to date. 156
2.1. Sample effects are important in Amazonia 158
Many species in Amazonian forests are rare or patchily distributed. This phenomenon is 159
especially pronounced in the large expanses of the basin that overlay heavily weathered, 160
nutrient-poor soils (e.g. Radtke et al. 2008), where resources such as fruits, flowers, and 161
nectar are scarce and plants are heavily defended against herbivore attack (Laurance 2001). 162
Herein lies a key implication for understanding forest fragmentation: given their rarity, many 163
species may be absent from fragments not because their populations have vanished, but 164
because they were simply not present at the time of fragment creation—a phenomenon termed 165
the ‘sample effect’ (Wilcox and Murphy 1985). Such sample effects are the hypothesized 166
explanation for the absence of many rare understory bird species from fragments (Ferraz et al. 167
2007). In addition, many beetles (Didham et al. 1998a), bats (Sampaio et al. 2003), ant-168
defended plants (Bruna et al. 2005), and trees (Bohlman et al. 2008, Laurance et al. 2010b) at 169
the BDFFP exhibit high levels of habitat specialization or patchiness. In a region where rarity 170
and patchy distributions of species are the norm, sample effects appear to play a major role in 171
structuring fragmented communities. Given these sample effects, nature reserves will have to 172
be especially large to sustain viable populations of rare species (Lovejoy and Oren 1981, 173
Laurance 2005, Peres 2005, Radtke et al. 2008). 174
2.2. Fragment size is vital 176
Although fragments range from just 1–100 ha in the BDFFP study area, understanding 177
fragment-area effects has long been a central goal of the project (Lovejoy and Oren 1981, 178
Lovejoy et al. 1984, 1986). The species richness of many organisms declines with fragment 179
area, even with constant sampling effort across all fragments. Such declines are evident in leaf 180
bryophytes (Zartman 2003), tree seedlings (Benítez-Malvido and Martinez-Ramos 2003a), 181
palms (Scariot 1999), understory insectivorous birds (Stratford and Stouffer 1999; Ferraz et 182
al. 2007), gleaning animal-eating bats (Sampaio 2000, Meyer et al. unpublished data), 183
primates (Gilbert and Setz 2001, Boyle and Smith 2010a), and larger herbivorous mammals 184
(Timo 2003), among others. For these groups, smaller fragments are often unable to support 185
viable populations and deleterious edge effectsecological changes associated with the 186
abrupt, artificial edges of forest fragmentscan also rise sharply in intensity (Didham et al. 187
1998a). A few groups, such as ant-defended plants and their ant mutualists, show no 188
significant decline in diversity with fragment area (Bruna et al. 2005). 189
Fragment size also influences the rate of species losses, with smaller fragments losing 190
species more quickly (Lovejoy et al. 1986, Stouffer et al. 2008). Assuming the surrounding 191
matrix is hostile to bird movements and precludes colonization, Ferraz et al. (2003) estimated 192
that a 1000-fold increase in fragment area would be needed to slow the rate of local species 193
extinctions by 10-fold. Even a fragment of 10,000 ha in area would be expected to lose a 194
substantial part of its bird fauna within one century (Ferraz et al. 2003). Similarly, mark-195
recapture data suggest that very large fragments will be needed to maintain fully intact 196
assemblages of some faunal groups, such as ant-following birds, which forage over large 197
areas of forest (Van Houtan et al. 2007). 198
3. Edge effects 200
3.1. Forest hydrology is disrupted 201
The hydrological regimes of fragmented landscapes differ markedly from those of intact 202
forest (Kapos 1989). Pastures or crops surrounding fragments have much lower rates of 203
evapotranspiration than do forests because they have far lower leaf area and thus less rooting 204
depth. Additionally, such clearings are hotter and drier than forests (Camargo & Kapos 1995). 205
Field observations and heat-flux simulations suggest that desiccating conditions can penetrate 206
up to 100–200 m into fragments from adjoining clearings (Malcolm 1998; Didham and 207
Lawton 1999). Further, streams in fragmented landscapes experience greater temporal 208
variation in flows than do those in forests, because clearings surrounding fragments have less 209
evapotranspiration and rainfall interception by vegetation (Trancoso 2008). Free runoff 210
promotes localized flooding in the wet season and stream failure in the dry season, with 211
potentially important impacts on aquatic invertebrates (Nessimian et al. 2008) and other 212
organisms. 213
Forest fragmentation also can alter low-level atmospheric circulation, which in turn 214
affects local cloudiness and rainfall. The warm, dry air over clearings tends to rise, creating 215
zones of low air pressure. The relatively cool, moist air over forests is drawn into this vacuum 216
(Avissar and Schmidt, 1998). As it warms it also rises and forms convectional clouds over the 217
clearing, which can lead to localized thunderstorms (Avissar and Liu 1996). In this way, 218
clearings of a few hundred hectares or more can draw moisture away from nearby forests 219
(Laurance 2004a, Cochrane and Laurance 2008). In Eastern Amazonia, satellite observations 220
of canopy-water content suggest such desiccating effects typically penetrate 1.0–2.7 km into 221
fragmented forests (Briant et al. 2010). This moisture-robbing function of clearings, in 222
concert with frequent burning in adjoining pastures, could help explain why fragmented 223
forests are so vulnerable to destructive, edge-related fires (Cochrane and Laurance 2002, 224
2008). 225
3.2. Edge effects often dominate fragment dynamics 227
Edge effects are among the most important drivers of ecological change in the BDFFP 228
fragments. The distance to which different edge effects penetrate into fragments varies 229
widely, ranging from <10 to 300 m at the BDFFP (Laurance et al. 2002) and considerably 230
further (at least 2–3 km) in areas of the Amazon where edge-related fires are common 231
(Cochrane and Laurance 2002, 2008; Briant et al. 2010). 232
Edge phenomena are remarkably diverse (Fig. 2). They include increased desiccation 233
stress, wind shear, and wind turbulence that sharply elevate rates of tree mortality and damage 234
(Laurance et al. 1997, 1998a). These in turn cause wide-ranging alterations in the community 235
composition of trees (Laurance et al. 2000, 2006a, 2006b) and lianas (Laurance et al., 2001b). 236
Such stresses may also reduce germination (Bruna 1999) and establishment (Uriarte et al., 237
2010) of shade-tolerant plant species in fragments, leading to dramatic changes in the 238
composition and abundance of tree seedlings (Benítez-Malvido 1998, Benítez-Malvido and 239
Martinez-Ramos 2003a). 240
Many animal groups, such as numerous bees, wasps, flies (Fowler et al. 1993), beetles 241
(Didham et al. 1998a, 1998b), ants (Carvalho and Vasconcelos 1999), butterflies (Brown and 242
Hutchings 1997), understory birds (Quintela 1985, Laurance 2004b), and gleaning animal-243
eating bats (Rocha et al. 2013) decline in abundance near fragment edges. Negative edge 244
effects are apparent even along forest roads (20–30 m width) in large forest tracts. Among 245
understory birds, for example, five of eight foraging guilds declined significantly in 246
abundance within 70 m of roads, whereas tree mortality increased and canopy cover declined 247
(Laurance 2004b). 248
Some groups of organisms remain stable or even increase in abundance near edges. 249
Leaf bryophytes (Zartman and Nascimento 2006), wandering spiders (Ctenus spp; Rego et al. 250
2007, Mestre and Gasnier 2008), and many frogs (Gascon 1993) show no significant response 251
to edges. Organisms that favor forest ecotones or disturbances, such as many species of gap-252
favoring and frugivorous birds (Laurance 2004b), hummingbirds (Stouffer and Bierregaard 253
1995a), frugivorous bats that exploit early successional plants (Sampaio 2000), light-loving 254
butterflies (Leidner et al. 2010), and fast-growing lianas (Laurance et al. 2001b), increase in 255
abundance near edges, sometimes dramatically. 256
3.3. Edge effects are cumulative 258
BDFFP research provides strong support for the idea that two or more nearby edges create 259
more severe edge effects than does just one (Fig. 3). This conclusion is supported by studies 260
of edge-related changes in forest microclimate (Kapos 1989, Malcolm 1998), vegetation 261
structure (Malcolm 1994), tree mortality (Laurance et al. 2006a), abundance and species 262
richness of tree seedlings (Benítez-Malvido 1998, Benítez-Malvido and Martinez-Ramos 263
2003a), liana abundance (Laurance et al. 2001b), and the density and diversity of disturbance-264
loving pioneer trees (Laurance et al. 2006a, 2006b, 2007). The additive effects of nearby 265
edges could help to explain why small (<10 ha) or irregularly shaped forest remnants are 266
often so severely altered by forest fragmentation (Zartman 2003, Laurance et al. 2006a). 267
3.4. Edge age, structure, and adjoining vegetation influence edge effects 269
When a forest edge is newly created it is open to fluxes of wind, heat, and light, creating 270
sharp edge-interior gradients in forest microclimate that stress or kill many rainforest trees 271
(Lovejoy et al. 1986, Sizer and Tanner 1999). As the edge ages, however, proliferating vines 272
and lateral branch growth tend to ‘seal’ the edge, making it less permeable to microclimatic 273
changes (Camargo and Kapos 1995, Didham and Lawton 1999). Tree death from 274
microclimatic stress is likely to decline over the first few years after edge creation (D’Angelo 275
et al. 2004) because the edge becomes less permeable, because many drought-sensitive 276
individuals die immediately, and because surviving trees may acclimate to drier, hotter 277
conditions near the edge (Laurance et al. 2006a). Tree mortality from wind turbulence, 278
however, probably increases as the edge ages and becomes more closed because, as suggested 279
by wind-tunnel models, downwind turbulence increases when edges are less permeable 280
(Laurance 2004a). 281
Regrowth forest adjoining fragment edges can also lessen edge-effect intensity. 282
Microclimatic alterations (Didham and Lawton 1999), tree mortality (Mesquita et al. 1999), 283
and edge avoidance by understory birds (Develey and Stouffer 2001, Laurance 2004b, 284
Laurance et al. 2004), and gleaning bats that feed on invertebrates and small vertebrates 285
(Meyer et al., 2013) are all reduced substantially when forest edges are buffered by adjoining 286
regrowth forest, relative to edges adjoined by cattle pastures. 287
4. Isolation and matrix effects 289
4.1. Matrix structure and composition affect fragments 290
Secondary forests have gradually overtaken most pastures in the BDFFP landscape. This re-291
growth lessens the effects of fragmentation for some taxa as the matrix becomes less hostile 292
to faunal use and movements. Several species of insectivorous birds that had formerly 293
disappeared have recolonized fragments as the surrounding secondary forest grew back 294
(Stouffer and Bierregaard 1995b). The rate of local extinctions of birds has also declined 295
(Stouffer et al. 2008). Similarly, gleaning animal-eating bats, which occurred at low 296
abundances in fragments (Sampaio 2000) and in secondary regrowth (Bobrowiec and Gribel 297
2010) 10-15 years ago, have since increased in response to matrix regeneration (Meyer et al. 298
2013). A number of other species, including certain forest spiders (Mestre and Gasnier 2008), 299
dung beetles (Quintero and Roslin 2005), euglossine bees (Becker et al. 1991), and monkeys 300
such as red howlers, bearded sakis, and brown capuchins (Boyle and Smith 2010a) have 301
recolonized some fragments. 302
The surrounding matrix also has a strong effect on plant communities in fragments by 303
mediating certain edge effects (see above), influencing the movements of pollinators (Dick 304
2001, Dick et al. 2003) and seed dispersers (Jorge 2008, Bobrowiec and Gribel 2010, Boyle 305
and Smith 2010a), and strongly affecting the seed rain that arrives in fragments. For instance, 306
pioneer trees regenerating in fragments differed strikingly in composition between fragments 307
surrounded by Cecropia-dominated regrowth and those encircled by Vismia-dominated 308
regrowth (Nascimento et al. 2006). In this way plant and animal communities in fragments 309
could come to mirror to some extent the composition of the surrounding matrix (Laurance et 310
al. 2006a, 2006b), a phenomenon observed elsewhere in the tropics (e.g. Janzen 1983, 311
Diamond et al. 1987). 312
4.2 Matrix is affected by history and forest proximity 314
Land-use history is a primary driver of secondary succession in the Central Amazon, resulting 315
in the establishment of distinct trajectories differing in structure, composition, biomass, and 316
dynamics (Mesquita et al. 1999, Williamson et al. 2014). Intensive use with prescribed fire to 317
maintain pastures compromises the regenerative potential of land which, once abandoned, is 318
colonized by few species and dominated by the genus Vismia, resulting in secondary forests 319
that are depauperate in richness and stalled in succession. Where land use has been less 320
intensive, a more diverse vegetation, dominated by the genus Cecropia colonizes, fostering 321
relatively rapid plant succession. 322
Plant density and species diversity in secondary forests decrease with distance from 323
forest edge, and are significantly different between Vismia and Cecropia dominated 324
secondary forests. These differences were initially attributed to differential seed dispersal 325
limitations (Mesquita et al. 2001, Puerta, 2002). Wieland et al. (2011), however, showed that 326
the seed rain was similar for both types of second-growth and dominated by pioneer species, 327
with only the occasional presence of mature forest species, even very close to forest edges. 328
These results point to other relevant processes affecting plant establishment, such as seed 329
consumption, germination success, and seedling herbivory (Wieland et al. 2011, Massoca et 330
al. 2013). 331
4.3 Even narrow clearings are harmful 333
Many Amazonian species avoid clearings, and even a forest road can be an insurmountable 334
barrier for some. A number of understory insectivorous birds exhibit depressed abundances 335
near forest roads (20–40 m width) (Laurance 2004b) and strongly inhibited movements across 336
those roads (Laurance et al. 2004). Experimental translocations of resident adult birds reveal 337
such bird species will cross a highway (50–75 m width) but not a small pasture (250 m width) 338
to return to their territory (Laurance and Gomez 2005). Individuals of other vulnerable 339
species, however, have traversed clearings to escape from small fragments to larger forest 340
areas (Harper 1989, Van Houtan et al. 2007). Captures of understory birds declined 341
dramatically in fragments when a 100 m-wide swath of regrowth forest was cleared around 342
them, suggesting that species willing to traverse regrowth would not cross clearings (Stouffer 343
et al. 2006). 344
Aside from birds, clearings of just 100–200 m width can evidently reduce or halt the 345
movements of many forest-dependent organisms (Laurance et al. 2009b), ranging from 346
herbivorous insects (Fáveri et al. 2008), euglossine bees (Powell and Powell 1987), and dung 347
beetles (Klein 1989) to the spores of epiphyllous lichens (Zartman and Nascimento 2006, 348
Zartman and Shaw 2006). Narrow clearings can also provide invasion corridors into forests 349
for exotic and nonforest species (Gascon et al. 1999; Laurance et al. 2009b). 350
5. Landscape dynamics 352
5.1. Rare disturbances can leave lasting legacies 353
Rare events such as windstorms and droughts have strongly influenced the ecology of 354
fragments. Rates of tree mortality rose abruptly in fragmented (Laurance et al., 2001c) and 355
intact forests (Williamson et al. 2000, Laurance et al. 2009a) in the year after the intense 1997 356
El Niño drought. Such pulses of tree death help drive changes in the floristic composition and 357
carbon storage of fragments (Laurance et al. 2007). Leaf-shedding by drought-stressed trees 358
also increases markedly during droughts, especially within ~60 m of forest edges (Laurance 359
and Williamson 2001). The additional litter increases the susceptibility of fragments to 360
intrusion by surface fires (Cochrane and Laurance 2002, 2008). 361
Intense windblasts from convectional thunderstorms have occasionally strafed parts of 362
the BDFFP landscape and caused intense forest damage and tree mortality, especially in the 363
fragments. Fragments in the easternmost cattle ranch at the BDFFP have had substantially 364
lower rates of tree mortality than did those in the other two ranches, because the former have 365
so far escaped windstorms (Laurance et al. 2007). These differences have strongly influenced 366
the rate and trajectory of change in tree-community composition in fragments (Laurance et al. 367
2006b). Hence, by altering forest dynamics, composition, structure, and carbon storage, rare 368
disturbances have left an enduring imprint on the ecology of fragmented forests. 369
5.2. Fragments are hyperdynamic 371
The BDFFP fragments experience exceptionally large variability in population and 372
community dynamics, relative to intact forest, despite being largely protected from ancillary 373
human threats such as fires, logging, and overhunting. Being a small resource base, a habitat 374
fragment is inherently vulnerable to stochastic effects and external vicissitudes. Species 375
abundances can fluctuate dramatically in small communities, especially when immigration is 376
low and disturbances are frequent (Hubbell 2001). Edge effects, reduced dispersal, external 377
disturbances, and changing herbivore or predation pressure can all elevate the dynamics of 378
plant and animal populations in fragments (Laurance 2002, 2008). 379
Many examples of hyperdynamism have been observed in the BDFFP fragments. 380
Some butterfly species have experienced dramatic population irruptions in response to a 381
proliferation of their favored host plants along fragment margins (Brown and Hutchings 382
1997), and butterfly communities in general are hyperdynamic in fragments (Leidner et al. 383
2010). Bat assemblages also show pronounced species turnover, particularly in 1-ha 384
fragments (Meyer et al. 2013). Streamflows are far more variable in fragmented than forested 385
watersheds (Trancoso 2008). Rates of tree mortality and recruitment are chronically elevated 386
in fragments (Laurance et al. 1998a, b), with major pulses associated with rare disturbances 387
(see above). Further, tree species disappear and turn over far more rapidly in fragments than 388
intact forest, especially within ~100 m of forest margins (Laurance et al. 2006b). These and 389
many other instabilities plague small, dwindling populations in the BDFFP fragments. 390
5.3. Fragments in different landscapes diverge 392
An important insight is that different fragmented landscapeseven those as alike as the three 393
large cattle ranches in the BDFFP, which have very similar forests, soils, climate, fragment 394
ages, and land-use histories—can diverge to a surprising degree in species composition and 395
dynamics. Although spanning just a few dozen kilometers, the three ranches are following 396
unexpectedly different trajectories of change. 397
At the outset, small initial differences among the ranches multiplied into much bigger 398
differences. Parts of the western and eastern ranches were cleared in 1983, when an early wet 399
season prevented burning of the felled forest. Tall, floristically diverse Cecropia-dominated 400
regrowth quickly developed in these areas, whereas areas cleared in the years just before or 401
after became cattle pastures or, eventually, scrubby Vismia-dominated regrowth (Williamson 402
and Mesquita 2001). These different successional trajectories manifested, for instance, as 403
distinct differences in bat assemblages, whereby Cecropia-dominated regrowth retained a 404
considerable fraction of forest-specialist bat species found in continuous forest compared to 405
Vismia regrowth (Bobrowiec and Gribel 2010). As discussed above, the differing matrix 406
vegetation strongly affected the dynamics of plant and animal communities in the nearby 407
fragments. These differences were magnified by subsequent windstorms, which heavily 408
damaged most fragments in the central and western ranches, yet left fragments in the eastern 409
ranch unscathed. Even identically sized fragments in the three ranches have had remarkably 410
different dynamics and vectors of compositional change (Laurance et al. 2007). 411
The apparently acute sensitivity of fragments to local landscape and weather 412
dynamics—even within a study area as initially homogeneous as ours—prompted us to 413
propose a ‘landscape-divergence hypothesis’ (Laurance et al. 2007). We argue that fragments 414
within the same landscape tend to have similar dynamics and trajectories of change in species 415
composition, which will often differ from those in other landscapes. Over time, this process 416
will tend to homogenize fragments in the same landscape, and promote ecological divergence 417
among fragments in different landscapes. Evidence for this hypothesis is provided by tree 418
communities in our fragments, which appear to be diverging in composition among the three 419
cattle ranches (Fig. 4). Pioneer and weedy trees are increasing in all fragments, but the 420
composition of these generalist plants and their rate of increase differ markedly among the 421
three ranches (Scariot 2001, Laurance et al. 2006a, 2007, Nascimento et al. 2006). This same 422
pattern of landscape homogenization within ranches can also be seen for bat assemblages in 423
the secondary forest matrix (Bobrowiec and Gribel, 2010). 424
6. Broader consequences of fragmentation 426
6.1. Ecological distortions are common 427
Many ecological interactions are altered in fragmented forests. Fragmented communities can 428
pass through unstable transitional states that may not otherwise occur in nature (Terborgh et 429
al. 2001). Moreover, species at higher trophic levels, such as predators and parasites, are often 430
more vulnerable to fragmentation than are herbivores, thereby altering the structure and 431
functioning of food webs (Didham et al. 1998b, Terborgh et al. 2001). 432
BDFFP findings suggest that even forest fragments that are unhunted, unlogged, and 433
unburned have reduced densities of key mammalian seed dispersers. As a result, seed 434
dispersal for the endemic, mammal-dispersed tree Duckeodendron cestroides was far lower in 435
fragments, with just ~5% of the number of seeds being dispersed >10 m away from parent 436
trees than in intact forest (Cramer et al. 2007a). Leaf herbivory appears reduced in fragments, 437
possibly because of lower immigration of insect herbivores (Fáveri et al. 2008). Dung beetles 438
exhibit changes in biomass and guild structure in fragments (Radtke et al. 2008) that could 439
alter rates of forest nutrient cycling and secondary seed dispersal (Klein 1989, Andresen 440
2003). Exotic Africanized honeybees, a generalist pollinator, are abundant in matrix and edge 441
habitats and can alter pollination distances and gene flow for some tree species (Dick 2001, 442
Dick et al. 2003). A bewildering variety of ecological distortions can pervade fragmented 443
habitats, and a challenge for conservation biologists is to identify those of greatest importance 444
and generality. 445
6.2. Fragmentation affects much more than biodiversity 447
Habitat fragmentation affects far more than biodiversity and interactions among species; 448
many ecosystem functions, including hydrology (see above) and biochemical cycling, are also 449
being altered. Among the most important of these are fundamental changes in forest biomass 450
and carbon storage. 451
Carbon storage in fragmented forests is affected by a suite of interrelated changes. 452
Many trees die near forest edges (Laurance et al. 1997, 1998a), including an alarmingly high 453
proportion of large (60 cm dbh) canopy and emergent trees that store much forest carbon 454
(Laurance et al. 2000). Fast-growing pioneer trees and lianas that proliferate in fragments are 455
smaller and have lower wood density (Fig. 5), and thereby sequester much less carbon, than 456
do the mature-phase trees they replace (Laurance et al. 2001b, 2006a). Based on current rates 457
of forest fragmentation, the edge-related loss of forest carbon storage might produce up to 150 458
million tons of atmospheric carbon emissions annually, above and beyond that from tropical 459
deforestation per se (Laurance et al. 1998c). Such discharge would exceed the yearly carbon 460
emissions of the entire United Kingdom. Note, however, that most of this emission is already 461
counted in the existing estimates of the impact of Amazonian land-use change because the 462
deforestation emission estimates use forest biomass values for undegraded forest (Fearnside 463
2000). Because most deforestation occurs by expansion of already-existing clearings, forest 464
edges (with reduced biomass) are the first areas to be cleared. Only the annual increase in the 465
total length of forest edges represents an addition. Improved emissions estimates, accounting 466
for degradation by logging, fire and fragmentation, are a high priority. 467
In addition, biomass is being redistributed in fragmented forests. Less biomass is 468
stored in large, densely wooded old-growth trees and more in fast-growing pioneer trees, 469
disturbance-loving lianas, woody debris, and leaf litter (Sizer et al. 2000, Nascimento and 470
Laurance 2004, Vasconcelos and Luizão 2004). Finally, carbon cycling accelerates. The large, 471
mature-phase trees that predominate in intact forests can live for many centuries or even 472
millennia (Chambers et al. 1998, Laurance et al. 2004), sequestering carbon for long periods 473
of time. However, the residence time of carbon in early successional trees, vines, and 474
necromass (wood debris, litter), which proliferate in fragments, is far shorter (Nascimento and 475
Laurance 2004). Other biochemical cycles, such as those affecting key nutrients like 476
phosphorus (Sizer et al. 2000) and calcium (Vasconcelos and Luizão 2004), may also be 477
altered in fragmented forests, given the striking changes in biomass dynamics, hydrology, and 478
thermal regimes they experience. 479
7. Predicting species responses to fragmentation 481
7.1. Species losses are highly nonrandom 482
Local extincitions of species in the BDFFP fragments have occurred in a largely predictable 483
sequence, with certain species being consistently more vulnerable than others. Among birds, a 484
number of understory insectivores, including army ant-followers, solitary species, terrestrial 485
foragers, and obligate mixed-flock members, are most susceptible to fragmentation. Others, 486
including edge/ gap species, insectivores that use mixed flocks facultatively, hummingbirds, 487
and many frugivores, are far less vulnerable (Antongiovanni and Metzger 2005, Stouffer et al. 488
2006, 2008). In a similar vein, among bats, gleaning predators are consistently the most 489
vulnerable species whereas many frugivores respond positively to fragmentation and 490
disturbance (Sampaio 2000, Bobrowiec and Gribel 2010, Rocha et al. 2013). Primates exhibit 491
similarly predictable patterns of species loss, with wide-ranging frugivores, especially the 492
black spider-monkey, being most vulnerable (Boyle and Smith 2010a). Local extinctions in 493
fragments follow a foreseeable pattern, with species assemblages in smaller fragments rapidly 494
forming a nested subset of those in larger fragments (Stouffer et al. 2008). Random 495
demographic and genetic processes may help to drive tiny populations into oblivion, but the 496
species that reach this precarious threshold are far from random. 497
7.2. Fragmented communities are not neutral 500
An important corollary of nonrandom species loss is that fragmented forests are not neutral. 501
Neutral theory (Hubbell 2001) assumes that species in diverse, space-limited communities, 502
such as tropical trees, are competitively equivalent in order to make predictions about 503
phenomena such as species-area curves, the relative abundances of species in communities, 504
and the rate of species turnover in space. Hubbell (2001) emphasizes the potential relevance 505
of neutral theory for predicting community responses to habitat fragmentation: for isolated 506
communities, locally abundant species should be least extinction prone, with rare species 507
being lost more frequently from random demographic processes. Over time, fragments should 508
become dominated by initially abundant species, with rare species gradually vanishing; other 509
ecological traits of species are considered unimportant. 510
Gilbert et al. (2006) tested the efficacy of neutral theory for predicting changes in tree 511
communities at the BDFFP. Neutral theory effectively predicted the rate of local extinctions 512
of species from plots in fragmented and intact forest as a function of the local diversity and 513
the mortality rate of trees. However, in most fragments, the observed rate of change in species 514
composition was 2–6 times faster than predicted by the theory. Moreover, the theory was 515
wildly erroneous in predicting which species are most prone to local extinction. Rather than 516
becoming increasingly dominated by initially common species, fragments in the BDFFP 517
landscape have experienced striking increases in disturbance-loving pioneer species (Fig. 6) 518
(Laurance et al. 2006a), which were initially rare when the fragments were created. As a 519
model for predicting community responses to habitat fragmentation, neutral theory clearly 520
failed, demonstrating that ecological differences among species strongly influence their 521
responses to fragmentation. 522
7.3. Matrix use and area needs determine animal vulnerability 524
The responses of animal species to fragmentation appear largely governed by two key sets of 525
traits. The first is their spatial requirements for forest habitat. In birds (Van Houtan et al. 526
2007) and mammals (Timo 2003), wide-ranging forest species are more vulnerable than are 527
those with localized ranges and movements. Species with limited spatial needs, such as many 528
small mammals (Malcolm 1997), hummingbirds (Stouffer et al. 2008), frogs (Tocher et al. 529
1997), and ants (Carvalho and Vasconcelos 1999), are generally less susceptible to 530
fragmentation. 531
The second key trait for fauna is their tolerance of matrix habitats (Gascon et al. 532
1999), which comprises cattle pastures and regrowth forest in the BDFFP landscape. 533
Populations of species that avoid the matrix will be entirely isolated in fragments, and 534
therefore vulnerable to local extinction, whereas those that tolerate or exploit the matrix often 535
persist (Laurance 1991, Malcolm 1997, Antongiovanni and Metzger 2005, Ferraz et al. 2007, 536
Bobrowiec and Gribel 2010). At least among terrestrial vertebrates, matrix use is positively 537
associated with tolerance of edge habitats (Laurance 2004b, Farneda 2013), an ability to 538
traverse small clearings (Laurance et al. 2004, Laurance and Gomez 2005), and behavioral 539
flexibility (Neckel-Oliveira and Gascon 2006, Stouffer et al. 2006, Van Houtan et al. 2006, 540
Boyle and Smith 2010b). Within particular animal groups, such as beetles or small mammals, 541
traits such as body size and natural abundance are poor or inconsistent predictors of 542
vulnerability (Laurance 1991, Didham et al. 1998a, Jorge 2008, Boyle and Smith 2010a). 543
Natural abundance, however, is an important predictor of sensitivity to fragmentation for bats 544
at the BDFFP (Farneda, 2013). 545
7.4. Disturbance tolerance and mutualisms affect plant vulnerability 547
Among plants, a different suite of factors is associated with vulnerability to fragmentation. 548
Because fragments suffer chronically elevated tree mortality, faster-growing pioneer trees and 549
lianas that favor treefall gaps are favored at the expense of slower-growing mature-phase trees 550
(Laurance et al. 2006a, b). Pioneer species often flourish in the matrix and produce abundant 551
small fruits that are carried into fragments by frugivorous birds and bats that move between 552
the matrix and nearby fragments (Sampaio 2000, Nascimento et al. 2006). Especially 553
vulnerable in fragments are the diverse assemblages of smaller subcanopy trees that are 554
physiologically specialized for growth and reproduction in dark, humid, forest-interior 555
conditions (Laurance et al. 2006b). Tree species that have obligate outbreeding systems, rely 556
on animal seed dispersers, or have relatively large, mammal-dispersed seeds also appear 557
vulnerable (Laurance et al. 2006b, Cramer et al. 2007b). 558
These combinations of traits suggest that plant communities in fragmented forests are 559
structured primarily by chronic disturbances and microclimatic stresses and possibly also by 560
alterations in animal pollinator and seed-disperser communities. For long-lived plants such as 561
Heliconia species and many mature-phase trees, demographic models suggest that factors that 562
reduce adult survival and growth—such as recurring wind disturbance and edge-related 563
microclimatic stresses—exert a strong influence on population growth (Bruna 2003, Bruna 564
and Oli 2005). 565
Differential tolerance to drought also seems to play a role on secondary forests. We 566
find higher and significant mortality and lower biomass accumulation rates in Cecropia-567
dominated secondary forests, associated with drier years, while Vismia-dominated regrowth 568
showed a non-significant, but similar trend. It is likely that different species assemblages 569
account for the differential ability of these successional pathways to tolerate extreme climate 570
events (Mesquita et al. 2012). 571
8. Broad perspectives 573
8.1. Long-term research is crucial 574
Many insights from the BDFFP would have been impossible in a shorter-term study. The 575
exceptional vulnerability of large trees to fragmentation (Laurance et al. 2000) only became 576
apparent after two decades of fragment isolation. Likewise, the importance of ephemeral 577
events such as El Niño droughts (Williamson et al., 2000, Laurance et al. 2001c) and major 578
windstorms (Laurance et al. 2007) would not have been captured in a less-enduring project. 579
Many other key phenomena, such as the kinetics of species loss in fragments (Ferraz et al. 580
2003), the strong effects of matrix dynamics on fragmented bird and bat assemblages 581
(Antongiovanni and Metzger 2005, Stouffer et al. 2006, Meyer et al. 2013), the divergence of 582
fragments in different landscapes (Laurance et al. 2007), and the effects of fragmentation on 583
rare or long-lived species (Benítez-Malvido and Martinez-Ramos 2003b, Ferraz et al., 2007), 584
are only becoming understood after decades of effort. 585
Far more remains to be learned. For example, forest-simulation models parameterized 586
with BDFFP data suggest that even small (<10 ha) fragments will require a century or more to 587
stabilize in floristic composition and carbon storage (Groeneveld et al. 2009), given the long-588
lived nature of many tropical trees. Eventually, these fragments might experience a 589
fundamental reorganization of their plant communities, given striking shifts in the 590
composition of their tree, palm, liana, and herb seedlings (Scariot 2001; Benítez-Malvido and 591
Martinez-Ramos 2003a, Brum et al. 2008). If these newly recruited plants represent the future 592
of the forest, then the BDFFP fragments will eventually experience dramatic changes in 593
floristic composition—comparable to those observed in some other long-fragmented 594
ecosystems (e.g. da Silva and Tabarelli 2000, Girão et al. 2007, Santos et al. 2010). 595
8.2. The BDFFP is a best-case scenario 597
Although forest fragments in the BDFFP are experiencing a wide array of ecological changes, 598
it is important to emphasize that it is a controlled experiment. The fragments are square, not 599
irregular, in shape. They are isolated by distances of just 80–650 m from large tracts of 600
surrounding mature forest. They are embedded in a relatively benign matrix increasingly 601
dominated by regrowth forest. And they lack many of the ancillary threats, such as selective 602
logging, wildfires, and overhunting, that plague many fragmented landscapes and wildlife 603
elsewhere in the tropics (e.g. Moura et al. 2014). Such threats can interact additively or 604
synergistically with fragmentation, creating even greater perils for the rainforest biota 605
(Laurance and Cochrane 2001, Michalski and Peres 2005, Brook et al. 2008). For these 606
reasons, results from the BDFFP are clearly optimistic relative to many human-dominated 607
landscapes elsewhere in the tropics. 608
9. Conservation lessons from the BDFFP 610
9.1. Amazonian reserves should be large and numerous 611
A key conclusion from BDFFP research is that nature reserves in Amazonia should ideally be 612
very large—on the order of thousands to tens of thousands of square kilometers (Laurance 613
2005, Peres 2005). Only at this size will they be likely to maintain natural ecological 614
processes and sustain viable populations of the many rare and patchily distributed species in 615
the region (Ferraz et al. 2007, Radtke et al. 2008); provide resilience from rare calamities such 616
as droughts and intense storms (Laurance et al. 2007); facilitate persistence of terrestrial and 617
aquatic animals that migrate seasonally (Bührnheim and Fernandes 2003); buffer the reserve 618
from large-scale edge effects including fires, forest desiccation, and human encroachment 619
(Cochrane and Laurance 2002, Briant et al. 2010); maximize forest carbon storage (Laurance 620
et al. 1997, 1998c); and provide resilience from future climatic and atmospheric changes—the 621
effects of which are difficult to predict for Amazonia (Laurance and Useche 2009). Further, 622
on the ancient soils of Central and Eastern Amazonia, low plant productivity translates into 623
low population densities of many animals up the food chain, so reserves must be 624
proportionately larger to harbor viable populations (Radtke et al. 2008, Deichmann et al. 625
2011, 2012). 626
Nature reserves in Amazonia should also be numerous and stratified across major river 627
basins and climatic and edaphic gradients, in order to preserve locally endemic species 628
(Bierregaard et al. 2001, Laurance, 2007). Further, the core areas of reserves should ideally be 629
free of roads, which can promote human encroachment and hunting, internally fragment 630
wildlife populations, and facilitate invasions of exotic species and fire (Laurance et al. 631
2009b). 632
9.2. Protect and reconnect fragments 634
Few landscapes are as intact as those in the Amazon. Around the world, biodiversity hotspots, 635
which sustain the majority of species at risk of extinction, have, by definition, lost over 80% 636
of their natural vegetation and what remains is typically in small fragments (Myers et al. 637
2000). The BDFFP makes recommendations here, too. Reconnecting isolated fragments by 638
forest restoration will be an effective way of creating areas large enough to slow the rate of 639
local species extinctions (Lima and Gascon 1999, Pimm and Jenkins 2005). 640
In such heavily fragmented landscapes, protecting remaining forest remnants is highly 641
desirable, as they are likely to be key sources of plant propagules and animal seed dispersers 642
and pollinators (Mesquita et al. 2001, Chazdon et al. 2008). They may also act as stepping 643
stones for animal movements (Laurance and Bierregaard 1997, Dick et al. 2003). In regions 644
where forest loss is severe, forest fragments could also sustain the last surviving populations 645
of locally endemic species, thereby underscoring their potential value for nature conservation 646
(Arroyo-Rodríguez et al. 2009). 647
9.3. Fragmented landscapes can recover 649
A further lesson is that fragmented landscapes, if protected from fires and other major 650
disturbances, can begin to recover in just a decade or two. Forest edges tend to ‘seal’ 651
themselves, reducing the intensity of deleterious edge effects (Camargo and Kapos 1995, 652
Didham and Lawton 1999, Mesquita et al. 1999). Secondary forests can develop quickly in 653
the surrounding matrix (Mesquita et al. 2001), especially if soils and seedbanks are not 654
depleted by overgrazing or repeated burning (Ribeiro et al. 2009, Norden et al. 2011). 655
Secondary forests facilitate movements of many animal species (Gascon et al. 1999), allowing 656
them to recolonize fragments from which they had formerly disappeared (Becker et al. 1991, 657
Quintero and Roslin 2005, Stouffer et al. 2008, Bobrowiec and Gribel 2010, Boyle and Smith 658
2010a, Meyer et al. 2013). Species clinging to survival in fragments can also be rescued from 659
local extinction via the genetic and demographic contributions of immigrants (Zartman and 660
Nascimento 2006, Stouffer et al. 2008). 661
10. The future of the BDFFP 663
The BDFFP is one of the most enduring and influential ecological research projects in 664
existence today (Gardner et al. 2009, Peres et al. 2010). From the prism of understanding 665
habitat fragmentation, there are vital justifications for continuing it. The project, moreover, is 666
engaged in far more than fragmentation research: it plays a leading role in training 667
Amazonian scientists and decision-makers, and sustains long-term research on global-change 668
phenomena, forest regeneration, and basic ecological studies. 669
In its 36-year history, the BDFFP has faced myriad challenges. These include, among 670
others, the continuing fluctuations in currencies, challenges in obtaining research visas for 671
foreign students and scientists, inadequate core funding from its US and Brazilian sponsors, 672
and the vagaries of finding soft money for long-term research and to sustain a minimal 673
number of workers to support infrastructure and logistics. Yet today the BDFFP faces a far 674
more direct threat: encroachment from colonists and hunters. Since the late 1990s, the paving 675
of the 1100-km-long Manaus–Venezuela (BR-174) highway has greatly accelerated forest 676
colonization and logging north of the city. SUFRAMA, a Brazilian federal agency that 677
controls an expanse of land north of Manaus that includes the BDFFP, has begun settling 678
families in farming plots around the immediate periphery of the study area. At least six 679
colonization projects involving 180 families are planned for the area (Laurance and Luizão 680
2007). These settlements could be the beginning of a dramatic influx into the area, especially 681
if the proposed BR-319 highway between Manaus and Rondônia, a major deforestation 682
hotspot in southern Amazonia, is completed as planned (Fearnside and Graça 2006). 683
To date, BDFFP staff and supporters have managed to stave off most of the 684
colonization projectswhich also threaten to bisect the Central Amazonian Conservation 685
Corridor, a budding network of protected and indigenous lands that is one of the most 686
important conservation areas in the entire Amazon basin (Laurance and Luizão 2007). Yet it 687
is an uphill battle against a government bureaucracy that appears myopically determined to 688
push ahead with colonization at any cost—despite the fact that colonists can barely eke out a 689
living on the region’s infamously poor soils (Fearnside and Leal Filho, 2001). That such a 690
globally important research project and conservation area could be lost seems unthinkable. 691
That it could be lost for such a limited gain seems tragic. 692
Amazon forest is under stress from a variety of global changes that are expected to 693
increase in the coming decades. Beyond the considerable contributions of the BDFFP to date 694
in providing information relevant to understanding these changes, the project is uniquely well 695
placed to track the impacts of these changes as they occur. The BDFFP must continue its role 696
in contributing to the scientific basis for more serious global efforts to contain the current 697
human destruction of the environment at both the global and regional level. 698
Acknowledgements 701
We thank Laszlo Nagy and an anonymous referee for helpful comments on the manuscript. 702
The National Institute for Amazonian Research (INPA), Smithsonian Institution, US National 703
Science Foundation, Brazilian Science Foundation (CNPq), Amazonian State Science 704
Foundation (FAPEAM), NASA-LBA program, USAID, Mellon Foundation, Blue Moon 705
Fund, Marisla Foundation, and other organizations generously supported the BDFFP. 706
Substantial parts of this text are updated from Laurance et al. (2011). This is publication 707
number 640 in the BDFFP technical series. 708
Box 1. The Biological Dynamics of Forest Fragments Project 710
Since its inception in 1979, the Biological Dynamics of Forest Fragments Project (BDFFP) 711
has been assessing the impacts of fragmentation on the Amazon rainforest and biota (Lovejoy 712
et al. 1986, Bierregaard et al. 1992, Pimm 1998, Laurance et al. 2002, 2011). Today, it is the 713
world’s largest and longest-running experimental study of habitat fragmentation, as well as 714
one of the most highly cited ecological investigations ever conducted (Gardner et al. 2009, 715
Peres et al. 2010, Pitman et al. 2011). The BDFFP has also been a global leader in research, 716
training, and capacity development, with over 640 publications (, 717
more than 180 student theses, over 700 graduate students and conservation professionals 718
participating in sponsored courses, and over 1000 student interns to date. 719
The BDFFP is located 80 km north of Manaus, Brazil and spans ~1000 km2. The 720
topography is relatively flat (80–160 m elevation) but dissected by numerous stream gullies. 721
The heavily weathered, nutrient-poor soils of the study area are typical of large expanses of 722
the Amazon Basin. Rainfall ranges from 1900 to 3500 mm annually with a moderately strong 723
dry season from June to October. The forest canopy is 30–37 m tall, with emergent trees to 55 724
m. Species richness of trees (10 cm DBH) often exceeds 280 species ha-1 (Oliveira and 725
Mori 1999, Laurance et al. 2010b) with a comparably high level of diversity also evident in 726
many other plant and animal taxa. 727
The study area includes three large cattle ranges (~5000 ha each) containing 11 forest 728
fragments (five of 1 ha, four of 10 ha, and two of 100 ha), and expanses of nearby continuous 729
forest that serve as experimental controls. In the early 1980s, the fragments were isolated 730
from nearby intact forest by distances of 80–650 m by clearing and burning the surrounding 731
forest. A key feature was that pre-fragmentation censuses were conducted for many animal 732
and plant groups (e.g. trees, understory birds, small mammals, primates, frogs, many 733
invertebrate taxa), thereby allowing long-term changes in these groups to be assessed far more 734
confidently than in most other fragmentation studies. 735
Because of poor soils and low productivity, the ranches surrounding the BDFFP 736
fragments were largely abandoned after government fiscal incentives dried up from 1984 737
onwards. Secondary forests (initially dominated by Vismia spp in areas that were cleared and 738
burned, or by Cecropia spp in areas that were cleared without fire) proliferated in many 739
formerly forested areas (Mesquita et al. 2001). Some of the regenerating areas initially 740
dominated by Cecropia spp later developed into quite mature (>20 m tall), species-rich 741
secondary forests. Vismia-dominated regrowth, which is relatively species poor, is changing 742
far more slowly (Norden et al. 2011, Williamson et al. 2014). To help maintain isolation of 743
the experimental fragments, 100 m-wide strips of regrowth were cleared and burned around 744
each fragment on 4–5 occasions, most recently in 2013-2014. Additional human disturbances 745
that harm many fragmented landscapes in the Amazon, such as major fires and logging, are 746
largely prevented at the BDFFP. Hunting pressure has been very limited until recently, 747
following a government decision to increase colonization in the general area (Laurance and 748
Luizão 2007). Laurance and Bierregaard (1997) and Bierregaard et al. (2001) provide detailed 749
descriptions of the study area and design. 750
References 752
Andresen E (2003) Effect of forest fragmentation on dung beetle communities and functional 753
consequences for plant regeneration. Ecography 26:87–97. 754
Antongiovanni M, Metzger JP (2005) Influence of matrix habitats on the occurrence of 755
insectivorous bird species in Amazonian forest fragments. Biol Conserv 122:441–451. 756
Arroyo-Rodríguez V, Pineda E, Escobar F, Benítez-Malvido J (2009) Conservation value of 757
small patches to plant species diversity in highly fragmented landscapes. Conserv Biol 758
23:729–739. 759
Asner GP, Knapp D, Broadbent E, Oliveira P, Keller M, Silva J (2005) Selective logging in 760
the Brazilian Amazon. Science 310:480–482. 761
Avissar R, Liu Y (1996) A three-dimensional numerical study of shallow convective clouds 762
and precipitation induced by land-surface forcing. J Geophys Res 101:7499–7518. 763
Avissar R, Schmidt T (1998) An evaluation of the scale at which ground-surface heat flux 764
patchiness affects the convective boundary layer using a large-eddy simulation model. 765
J Atmos Sci 55:2666–2689. 766
Baker TR, Phillips OL, Malhi Y, Almeida S, Arroyo L, Di Fiore A, et al. (2004) Increasing 767
biomass in Amazonian forest plots. Phil Trans Roy Soc B 359:353-365. 768
Barlow J, Peres CA, Henriques L, Stouffer PC, Wunderle J (2006) The responses of 769
understorey birds to forest fragmentation, logging and wildfires: an Amazonian 770
synthesis. Biol Conserv 128:182–192. 771
Becker P, Moure JB, Peralta F (1991) More about euglossine bees in Amazonian forest 772
fragments. Biotropica 23:586–591. 773
Benítez-Malvido J (1998) Impact of forest fragmentation on seedling abundance in a tropical 774
rain forest. Conserv Biol 12:380–389. 775
Benítez-Malvido J, Martinez-Ramos M (2003a) Influence of edge exposure on tree seedling 776
species recruitment in tropical rain forest fragments. Biotropica 35:530–541. 777
Benítez-Malvido J, Martinez-Ramos M (2003b) Impact of forest fragmentation on understory 778
plant species richness in Amazonia. Conserv Biol 17:389–400. 779
Bierregaard RO, Lovejoy TE, Kapos V, dos Santos AA, Hutchings RW (1992) The biological 780
dynamics of tropical rainforest fragments. Bioscience 42:859–866. 781
Bierregaard R.O., Gascon C., Lovejoy T.E., Mesquita R. (Eds), 2001. Lessons from 782
Amazonia: Ecology and Conservation of a Fragmented Forest. Yale University Press, 783
New Haven, Connecticut. 784
Bobrowiec PED, Gribel R (2010) Effects of different secondary vegetation types on bat 785
community composition in Central Amazonia, Brazil. Anim Conserv 13:204–216. 786
Bohlman S, Laurance WF, Laurance SG, Nascimento H, Fearnside PM, Andrade A (2008) 787
Effects of soils, topography, and geographic distance in structuring central Amazonian 788
tree communities. J Veg Sci 19:863–874. 789
Boyle SA, Smith AT (2010a) Can landscape and species characteristics predict primate 790
presence in forest fragments in the Brazilian Amazon? Biol Conserv 143:1134–1143. 791
Boyle SA, Smith AT (2010b) Behavioral modifications in northern bearded saki monkeys 792
(Chiropotes satanas chiropotes) in forest fragments of central Amazonia. Primates 793
51:43–51. 794
Brazil, RADAMBRASIL (1973-1983) Levantamento de Recursos Naturais. Ministério das 795
Minas e Energia, Departamento Nacional de Produção Mineral, Rio de Janeiro, Brazil. 796
Briant G, Gond V, Laurance SGW (2010) Habitat fragmentation and the desiccation of forest 797
canopies: a case study from eastern Amazonia. Biol Conserv 143:2763–2769. 798
Broadbent E, Asner GP, Keller M, Knapp D, Oliveira P, Silva J (2008) Forest fragmentation 799
and edge effects from deforestation and selective logging in the Brazilian Amazon. 800
Biol Conserv 140:142–155. 801
Brook BW, Sodhi NS, Bradshaw CJA (2008) Synergisms among extinction drivers under 802
global change. Trends Ecol Evol 23:453–460. 803
Brown KS, Hutchings RW (1997) Disturbance, fragmentation, and the dynamics of diversity 804
in Amazonian forest butterflies. In: Laurance WF, Bierregaard RO (Eds), Tropical 805
forest remnants: Ecology, management, and conservation of fragmented communities. 806
University of Chicago Press, Chicago, pp 91–110. 807
Brum HD, Nascimento H, Laurance WF, Andrade A, Laurance SGW, Luizão R (2008) 808
Rainforest fragmentation and the demography of the economically important palm 809
Oenocarpus bacaba in central Amazonia. Plant Ecol 199:209– 215. 810
Bruna EM (1999) Seed germination in rainforest fragments. Nature 402:139. 811
Bruna EM (2003) Are plant populations in fragmented habitats recruitment limited? Tests 812
with an Amazonian herb. Ecology 84:932–947. 813
Bruna E, Oli M (2005) Demographic consequences of habitat fragmentation for an 814
Amazonian understory plant: analysis of life-table response experiments. Ecology 815
86:1816–1824. 816
Bruna EM, Vasconcelos HL, Heredia S (2005) The effect of habitat fragmentation on 817
communities of mutualists: a test with Amazonian ants and their host plants. Biol 818
Conserv 124:209–216. 819
Bührnheim CM, Fernandes CC (2003) Structure of fish assemblages in Amazonian rainforest 820
streams: effects of habitats and locality. Copeia 2003:255–262. 821
Butler RA, Laurance WF (2009) Is oil palm the next emerging threat to the Amazon? Trop 822
Cons Sci 2:1-10. 823
Camargo JLC, Kapos V (1995) Complex edge effects on soil moisture and microclimate in 824
central Amazonian forests. J Trop Ecol 11:205–211. 825
Carvalho KS, Vasconcelos HL (1999) Forest fragmentation in central Amazonia and its 826
effects on litter-dwelling ants. Biol Conserv 91:151–158. 827
Chambers JQ, Higuchi N, Schimel JP (1998) Ancient trees in Amazonia. Nature 391:135–828
136. 829
Chave J, Muller-Landau, HC, Baker TR, Easdale TA, ter Steege H, Webb CO (2006) 830
Regional and phylogenetic variation of wood density across 2,456 neotropical tree 831
species. Ecol Applic 16:2356-2367. 832
Chazdon RL, Harvey CA, Komar O, Griffith DM, Ferguson BG, Martinez- Ramos M, 833
Morales H, Nigh R, Soto-Pinto L, van Breugel M, Philpott SM (2008) Beyond 834
reserves: a research agenda for conserving biodiversity in human-modified tropical 835
landscapes. Biotropica 41:142–153. 836
Cochrane MA, Laurance WF (2002) Fire as a large-scale edge effect in Amazonian forests. J 837
Trop Ecol 18:311–325. 838
Cochrane MA, Laurance WF (2008) Synergisms among fire, land use, and climate change in 839
the Amazon. Ambio 37:522–527. 840
Condon MA, Sasek TW, Strain BR (1992) Allocation patterns in two tropical vines in 841
response to increased atmospheric CO2. Funct Ecol 6:680-685. 842
Cox PM, Betts RA, Collins M, Harris PP, Huntingford C, Jones CD (2004) Amazonian forest 843
dieback under climate-carbon cycle projections for the 21st century. Theoret Appl 844
Climatol 78:137-156. 845
Cox PM, Betts RA, Jones CD, Spall SA, Totterdell IJ (2000) Acceleration of global warming 846
due to carbon cycle feedbacks in a coupled climate model. Nature 408:184–187 847
Cox PM, Pearson D, Booth BB, Friedlingstein P, Huntingford C, Jones CD, Luke CMD 848
(2013) Sensitivity of tropical carbon to climate change constrained by carbon dioxide 849
variability. Nature 494:341-344. 850
Cramer JM, Mesquita R, Bentos T, Moser B, Williamson GB (2007a) Forest fragmentation 851
reduces seed dispersal of Duckeodendron cestroides, a Central Amazon endemic. 852
Biotropica 39:709–718. 853
Cramer JM, Mesquita R, Williamson GB (2007b) Forest fragmentation differentially affects 854
seed dispersal of large and small-seeded tropical trees. Biol Conserv 137:415–423. 855
Dai A (2012) Increasing drought under global warming in observations and models. Nature 856
Climate Change 3:52–58. 857
D’Angelo S, Andrade A, Laurance SG, Laurance WF, Mesquita R (2004) Inferred causes of 858
tree mortality in fragmented and intact Amazonian forests. J Trop Ecol 20:243–246. 859
da Silva JMC, Tabarelli M (2000) Tree species impoverishment and the future flora of the 860
Atlantic forest of northeast Brazil. Nature 404:72–74. 861
Develey P, Stouffer PC (2001) Roads affect movements by understory mixed-species flocks 862
in central Amazonian Brazil. Conserv Biol 15:1416–1422. 863
Deichmann JL, Lima A, Williamson GB (2011) Effects of geomorphology and primary 864
productivity on Amazonian leaf litter herpetofauna. Biotropica 43:149–156. 865
Deichmann JL, Toft CA, Deichmann PM, Lima AP, Williamson GB (2013) Neotropical 866
primary productivity affects biomass of leaf-litter herpetofaunal assemblage. J Trop 867
Ecol 28:427–435. 868
Diamond JM, Bishop KD, Balen SV (1987) Bird survival in an isolated Javan woodland: 869
island or mirror? Conserv Biol 1:132–142. 870
Dick CW (2001) Genetic rescue of remnant tropical trees by an alien pollinator. Proc Roy Soc 871
B 268:2391–2396. 872
Dick CW, Etchelecu G, Austerlitz F (2003) Pollen dispersal of tropical trees (Dinizia excelsa: 873
Fabaceae) by native insects and African honeybees in pristine and fragmented 874
Amazonian rainforest. Mol Ecol 12:753–764. 875
Didham RK, Hammond PM, Lawton JH, Eggleton P, Stork NE (1998a) Beetle species 876
responses to tropical forest fragmentation. Ecol Monogr 68:295–303. 877
Didham RK, Lawton JH (1999) Edge structure determines the magnitude of changes in 878
microclimate and vegetation structure in tropical forest fragments. Biotropica 31:17–879
30. 880
Didham RK, Lawton JH, Hammond PM, Eggleton P (1998b) Trophic structure stability and 881
extinction dynamics of beetles (Coleoptera) in tropical forest fragments. Proc Roy Soc 882
B 353:437–451. 883
Farneda FZ (2013) Bat species vulnerability to forest fragmentation in the Central Amazon. 884
M.Sc. Thesis. University of Lisbon, Portugal. 885
Fáveri SB, Vasconcelos HL, Dirzo R (2008) Effects of Amazonian forest fragmentation on 886
the interaction between plants, insect herbivores, and their natural enemies. J Trop 887
Ecol 24:57–64. 888
Fearnside PM (1997) Wood density for estimating forest biomass in Brazilian Amazonia. For 889
Ecol Manage 90:59-89. 890
Fearnside PM (2000) Global warming and tropical land-use change: Greenhouse gas 891
emissions from biomass burning, decomposition and soils in forest conversion, 892
shifting cultivation and secondary vegetation. Climatic Change 46:115-158. 893
Fearnside PM (2001a) Soybean cultivation as a threat to the environment in Brazil. Environ 894
Conserv 28:23–38. 895
Fearnside PM (2001b) Saving tropical forests as a global warming countermeasure: An issue 896
that divides the environmental movement. Ecol Econ 39:167-184. 897
Fearnside PM (2002) Avança Brasil: Environmental and social consequences of Brazil’s 898
planned infrastructure in Amazonia. Environ Manage 30:748-763. 899
Fearnside PM (2007) Brazil’s Cuiabá-Santarém (BR-163) Highway: The environmental cost 900
of paving a soybean corridor through the Amazon. Environ Manage 39:601-614. 901
Fearnside PM (2012) Brazil's Amazon forest in mitigating global warming: Unresolved 902
controversies. Climate Policy 12:70-81. 903
Fearnside PM (2013) Vines, CO2 and Amazon forest dieback. Nature Online Comment 904
( 905
Fearnside, PM (2014) The impact of land use on carbon stocks and fluxes. (this volume). 906
Fearnside PM, Graça PMLA (2006) BR-319: Brazil’s Manaus–Porto Velho highway and the 907
potential impact of linking the arc of deforestation to central Amazonia. Environ 908
Manage 38:705–716. 909
Fearnside PM, Leal Filho N (2001) Soil and development in Amazonia: Lessons from the 910
Biological Dynamics of Forest Fragments Project. In: Bierregaard RO, Gascon C, 911
Lovejoy TE, Mesquita R (Eds) Lessons from Amazonia: The ecology and 912
conservation of a fragmented forest. Yale University Press, New Haven, Connecticut, 913
USA, pp 291-312. 914
Ferraz G, Nichols JD, Hines J, Stouffer PC, Bierregaard RO, Lovejoy TE (2007) A large-915
scale deforestation experiment: effects of patch area and isolation on Amazon birds. 916
Science 315:238–241. 917
Ferraz G, Russell GJ, Stouffer PC, Bierregaard RO, Pimm SL, Lovejoy TE (2003) Rates of 918
species loss from Amazonian forest fragments. Proc Nat Acad Sci USA 100:14069–919
14073. 920
Finer M, Jenkins C, Pimm SL, Keane B, Ross C (2008) Oil and gas projects in the western 921
Amazon: threats to wilderness, biodiversity, and indigenous peoples. PLoS ONE 922
3:e2932. 923
Fowler HG, Silva CA, Ventincinque E (1993) Size, taxonomic and biomass distributions of 924
flying insects in central Amazonia: forest edge vs. understory. Rev Biol Trop 41:755–925
760. 926
Gardner TA, Barlow J, Chazdon R, Ewers R, Harvey C, Peres CA, Sodhi NS (2009) 927
Prospects for tropical forest biodiversity in a human-modified world. Ecol Lett 928
12:561–582. 929
Gascon C (1993) Breeding habitat use by Amazonian primary-forest frog species at the forest 930
edge. Biodiv. Conserv. 2:438–444. 931
Gascon C, Lovejoy TE, Bierregaard RO, Malcolm JR, Stouffer PC, Vasconcelos H, Laurance 932
WF, Zimmerman B, Tocher M, Borges S (1999) Matrix habitat and species 933
persistence in tropical forest remnants. Biol Conserv 91:223–229. 934
Gibbs HK, Reusch AS, Achard F, Clayton MK, Holmgren P, Ramankutty N, Foley JA (2010) 935
Tropical forests were the primary sources of new agricultural lands in the 1980s and 936
1990s. Proc Nat Acad SciUSA 107:16732–16737. 937
Gibson L, Lee TM, Koh LP, Brook BW, Gardner, TA, Barlow J, Peres CA, Bradshaw CJA, 938
Laurance WF, Lovejoy TE, Sodhi NS (2011) Primary forests are irreplaceable for 939
sustaining tropical biodiversity. Nature 478:378-381. 940
Gilbert B, Laurance WF, Leigh EG, Nascimento H (2006) Can neutral theory predict the 941
responses of Amazonian tree communities to forest fragmentation? Am Nat 168:304–942
317. 943
Gilbert KA, Setz EZF (2001) Primates in a fragmented landscape: six species in central 944
Amazonia. In: Laurance WF, Bierregaard RO (Eds) Tropical forest remnants: 945
Ecology, management and conservation of fragmented communities. University of 946
Chicago Press, Chicago, pp 207–221. 947
Girão LC, Lopes AV, Tabarelli M, Bruna EM (2007) Changes in tree reproductive traits 948
reduce functional diversity in a fragmented Atlantic forest landscape. PLoS ONE 949
2:e908. 950
Good P, Jones CD, Lowe JA,Betts RA, Gedney N (2013) Comparing tropical forest 951
projections from two generations of Hadley Centre Earth System models, HadGEM2-952
ES and HadCM3LC. J. Climate 26:495-511. 953
Groeneveld J, Alves L, Bernacci L, Catharino E, Knogge C, Metzger J, Pütz S, Huth A (2009) 954
The impact of fragmentation and density regulation on forest succession in the 955
Atlantic rain forest. Ecol Model 220:2450–2459. 956
Harper LH (1989) The persistence of ant-following birds in small Amazonian forest 957
fragments. Acta Amazonica 19:249–263. 958
Hubbell SP (2001) The Neutral Theory of Biodiversity and Biogeography. Princeton 959
University Press, Princeton, New Jersey. 960
Huntingford C, Zelazowski P, Galbraith D, Mercado LM, Sitch S, Fisher R, Lomas M, 961
Walker AP, Jones CD, Booth BB, Malhi Y, Hemming D, Kay G, Good P, Lewis SL, 962
Phillips OL, Atkin OK, Lloyd J, Gloor E, Zaragoza-Castells J, Meir P, Betts R, Harris 963
PP, Nobre C, Marengo J, Cox PM (2013) Simulated resilience of tropical rainforests 964
to CO2-induced climate change. Nature Geosci. 6:268–273. 965
Ingwell LL, Wright SJ, Becklund KK, Hubbell SP, Schnitzer SA (2010) The impact of lianas 966
on 10 years of tree growth and mortality on Barro Colorado Island, Panama. J Ecol 967
98:879-887. 968
IPCC (2013) Summary for policymakers. In: Stocker T F et al. (Eds) Climate change 2013: 969
The physical science basis. Intergovernmental Panel on Climate Change, Cambridge 970
University Press, Cambridge, UK. 971
Janzen DH (1983) No park is an island: Increase in interference from outside as park size 972
increases. Oikos 41:402–410. 973
Jorge ML (2008) Effects of forest fragmentation on two sister genera of Amazonian rodents 974
(Myoprocta acouchy and Dasyprocta leporina). Biol Conserv 141:617– 623. 975
Kapos V (1989) Effects of isolation on the water status of forest patches in the Brazilian 976
Amazon. J Trop Ecol 5:173–185. 977
Kapos V, Ganade G, Matusi E, Victoria RL (1993) Delta 13C as an indicator of edge effects 978
in tropical rainforest reserves. J Ecol 81:425-432. 979
Killeen TJ (2007) A Perfect Storm in the Amazon Wilderness: Development and 980
Conservation in the Context of the Initiative for the Integration of the Regional 981
Infrastructure of South America (IIRSA). Conservation International, Washington, 982
DC. 983
Klein BC (1989) Effects of forest fragmentation on dung and carrion beetle communities in 984
central Amazonia. Ecology 70:1715–1725. 985
Laurance SGW (2004b) Responses of understory rain forest birds to road edges in central 986
Amazonia. Ecol Applic 14:1344–1357. 987
Laurance SGW, Andrade A, Laurance WF (2010a) Unanticipated effects of stand dynamism 988
on Amazonian tree diversity. Biotropica 42:429-434. 989
Laurance SGW, Gomez MS (2005) Clearing width and movements of understory rainforest 990
birds. Biotropica 37:149–152. 991
Laurance SGW, Laurance WF, Andrade A, Fearnside PM, Harms K, Luizão R, (2010b) 992
Influence of soils and topography on Amazonian tree diversity: a landscape-scale 993
study. J Veg Sci 21:96–106. 994
Laurance SGW, Laurance WF, Nascimento H, Andrade A, Fearnside PM, Rebello E, Condit 995
R (2009a) Long-term variation in Amazon forest dynamics. J Veg Sci 20:323–333. 996
Laurance SGW, Stouffer PC, Laurance WF (2004) Effects of road clearings on movement 997
patterns of understory rainforest birds in central Amazonia. Conserv Biol 18:1099–998
1109. 999
Laurance WF (1991) Ecological correlates of extinction proneness in Australian tropical 1000
rainforest mammals. Conserv Biol 5:79–89. 1001
Laurance WF (2001) The hyper-diverse flora of the central Amazon: an overview. In: 1002
Bierregaard RO, Gascon C, Lovejoy TE, Mesquita R (Eds), Lessons from Amazonia: 1003
Ecology and conservation of a fragmented forest. Yale University Press, New Haven, 1004
Connecticut, pp 47–53. 1005
Laurance WF (2002) Hyperdynamism in fragmented habitats. J Veg Sci 13:595–602. 1006
Laurance WF (2004a) Forestclimate interactions in fragmented tropical landscapes. Phil. 1007
Trans. Roy. Soc. B 359:345–352. 1008
Laurance WF (2005) When bigger is better: the need for Amazonian megareserves. Trends 1009
Ecol Evol 20:645–648. 1010
Laurance WF (2007) Have we overstated the tropical biodiversity crisis? Trends Ecol Evol 1011
22:65–70. 1012
Laurance WF (2008) Theory meets reality: how habitat fragmentation research has 1013
transcended island biogeographic theory. Biol Conserv 141:1731–1744. 1014
Laurance WF, Andrade A, Magrach, A, Camargo JLC, Valsko J, Campbell M, Fearnside PM, 1015
Edwards W, Lovejoy TE, Laurance SGW (2014a) Long-term changes in liana 1016
abundance and forest dynamics in undisturbed Amazonian forests. Ecology 95:1604-1017
1611. 1018
Laurance WF, Andrade A, Magrach A, Camargo JLC, Campbell M, Fearnside PM, Edwards 1019
W, Valsko J, Lovejoy TE, Laurance SGW (2014b) Apparent environmental synergism 1020
drives the dynamics of Amazonian forest fragments. Ecology 95: 3018-3026. 1021
Laurance WF, Bierregaard RO (Eds) (1997) Tropical forest remnants: Ecology, management, 1022
and conservation of fragmented communities. University of Chicago Press, Chicago. 1023
Laurance WF, Camargo J, Luizão RCC, Laurance SGW, Pimm SL, Bruna E, Stouffer PC, 1024
Williamson GB, Benitez-Malvido J, Vasconcelos H, Van Houtan K, Zartman CE, 1025
Boyle S, Didham RK, Andrade A, Lovejoy TE (2011) The fate of Amazonian forest 1026
fragments: A 32-year investigation. Biol Conserv 144:56-67. 1027
Laurance WF, Cochrane MA (2001) Synergistic effects in fragmented landscapes. Conserv 1028
Biol 15:1488–1489. 1029
Laurance WF, Cochrane M, Bergen S, Fearnside PM, Delamonica P, Barber C, D’Angelo S, 1030
Fernandes T (2001a) The future of the Brazilian Amazon. Science 291:438–439. 1031
Laurance WF, Delamonica P, Laurance SGW, Vasconcelos HL, Lovejoy TE (2000) 1032
Rainforest fragmentation kills big trees. Nature 404:836. 1033
Laurance WF, Fearnside PM, Laurance SGW, Delamonica P, Lovejoy TE, Rankin-de Merona 1034
JM, Chambers JQ, Gascon C (1999) Relationship between soils and Amazon forest 1035
biomass: a landscape-scale study. For Ecol Manage 118:127-138. 1036
Laurance WF, Ferreira LV, Rankin-de Merona JM, Laurance SG (1998a) Rain forest 1037
fragmentation and the dynamics of Amazonian tree communities. Ecology 79:2032–1038
2040. 1039
Laurance WF, Ferreira LV, Rankin-de Merona JM, Laurance SGW, Hutchings R, Lovejoy 1040
TE (1998b) Effects of forest fragmentation on recruitment patterns in Amazonian tree 1041
communities. Conserv Biol 12:460–464. 1042
Laurance WF, Goosem M, Laurance SGW (2009b) Impacts of roads and linear clearings on 1043
tropical forests. Trends Ecol Evol 24:659–669. 1044
Laurance WF, Laurance SGW, Delamonica P (1998c) Tropical forest fragmentation and 1045
greenhouse gas emissions. For Ecol Manage 110:173–180. 1046
Laurance WF, Laurance SGW, Ferreira LV, Rankin-de Merona J, Gascon C, Lovejoy TE 1047
(1997) Biomass collapse in Amazonian forest fragments. Science 278:1117–1118. 1048
Laurance WF, Lovejoy TE, Vasconcelos H, Bruna E, Didham R, Stouffer P, Gascon C, 1049
Bierregaard R, Laurance SG, Sampaio E (2002) Ecosystem decay of Amazonian 1050
forest fragments: a 22-year investigation. Conserv Biol 16:605–618. 1051
Laurance WF, Luizão RCC (2007) Driving a wedge into the Amazon. Nature 448:409–410. 1052
Laurance WF, Nascimento H, Laurance SG, Andrade A, Ewers R, Harms K, Luizão R, 1053
Ribeiro J (2007) Habitat fragmentation, variable edge effects, and the landscape-1054
divergence hypothesis. PLoS ONE 2:e1017. 1055
Laurance WF, Nascimento H, Laurance SG, Andrade A, Fearnside PM, Ribeiro J (2006a) 1056
Rain forest fragmentation and the proliferation of successional trees. Ecology 87:469–1057
482. 1058
Laurance WF, Nascimento H, Laurance SG, Andrade A, Ribeiro J, Giraldo JP, Lovejoy TE, 1059
Condit R, Chave J, D’Angelo S (2006b) Rapid decay of tree community composition 1060
in Amazonian forest fragments. Proc Nat Acad Sci USA 103:19010–19014. 1061
Laurance WF, Nascimento H, Laurance, SG, Condit R, D’Angelo S, Andrade A (2004) 1062
Inferred longevity of Amazonian rainforest trees based on a long-term demographic 1063
study. For Ecol Manage 190:131–143. 1064
Laurance WF, Peres CA (Eds) (2006) Emerging Threats to Tropical Forests. University of 1065
Chicago Press, Chicago, USA. 1066
Laurance WF, Perez-Salicrup D, Delamonica P, Fearnside PM, D’Angelo S, Jerozolinski A, 1067
Pohl L, Lovejoy TE (2001b) Rain forest fragmentation and the structure of 1068
Amazonian liana communities. Ecology 82:105–116. 1069
Laurance WF, Useche DC (2009) Environmental synergisms and extinctions of tropical 1070
species. Conserv Biol 23:1427–1437. 1071
Laurance WF, Williamson GB (2001) Positive feedbacks among forest fragmentation, 1072
drought, and climate change in the Amazon. Conserv Biol 15:1529–1535. 1073
Laurance WF, Williamson GB, Delamonica P, Olivera A, Gascon C, Lovejoy TE, Pohl L 1074
(2001c) Effects of a strong drought on Amazonian forest fragments and edges. J Trop 1075
Ecol 17:771–785. 1076
Leidner AK, Haddad NM, Lovejoy TE (2010) Does tropical forest fragmentation increase 1077
long-term variability of butterfly communities? PLoS ONE 5:e9534. 1078
Lima M, Gascon C (1999) The conservation value of linear forest remnants in central 1079
Amazonia. Biol Conserv 91:241–247. 1080
Lovejoy TE, Bierregaard RO, Rylands AB, Malcolm JR, Quintela C, Harper L, Brown K, 1081
Powell A, Powell G, Schubart H, Hays M (1986) Edge and other effects of isolation 1082
on Amazon forest fragments. In: Soulé ME (Ed), Conservation biology: The science 1083
of scarcity and diversity. Sinauer, Sunderland, Massachusetts, pp 257–285. 1084
Lovejoy TE, Oren DC (1981) Minimum critical size of ecosystems. In: Burgess RL, Sharp 1085
DM (Eds), Forest-island dynamics in man-dominated landscapes. Springer-Verlag, 1086
New York, pp 7–12. 1087
Lovejoy TE, Rankin JM, Bierregaard RO, Brown KS, Emmons LH, Van der Voort ME 1088
(1984) Ecosystem decay of Amazon forest fragments. In: Nitecki MH (Ed), 1089
Extinctions. University of Chicago Press, Chicago, pp 295–325. 1090
Malcolm JR (1994) Edge effects in central Amazonian forest fragments. Ecology 75:2438–1091
2445. 1092
Malcolm JR (1997) Biomass and diversity of small mammals in Amazonian forest fragments. 1093
In: Laurance WF, Bierregaard RO (Eds) Tropical forest remnants: Ecology, 1094
management, and conservation of fragmented communities. University of Chicago 1095
Press, Chicago, pp 207–221. 1096
Malcolm JR (1998) A model of conductive heat flow in forest edges and fragmented 1097
landscapes. Clim Change 39:487–502. 1098
Massoca PES, Jakovac ACC, Vizcarra TB, Williamson GB, Mesquita RCG (2012) 1099
Dinâmica e trajetórias da sucessão secundária na Amazônia central. Bol Mus Para 1100
Emílio Goeldi Cienc Nat Belém 7:235-250. 1101
Mesquita R, Delamônica P, Laurance WF (1999) Effects of surrounding vegetation on edge-1102
related tree mortality in Amazonian forest fragments. Biol Conserv 91:129–134. 1103
Mesquita RCC, Bentos TV, Jakovac ACC, Massoca P, Williamson GB (2012) Amazonian 1104
secondary forests respond to climatic variation: biomass dynamics. 49th Annual 1105
Meeting of the Association for Tropical Biology and Conservation, Bonito, MS, 1106
Brazil. 1107
Mesquita R, Ickes K, Ganade G, Williamson GB (2001) Alternative successional pathways in 1108
the Amazon basin. J Ecol 89:528–537. 1109
Mestre LAM, Gasnier TR (2008) Populações de aranhas errantes do gênero Ctenus em 1110
fragmentos florestais na Amazônia Central. Acta Amazonica 38:159–164. 1111
Michalski F, Peres CA (2005) Anthropogenic determinants of primate and carnivore local 1112
extinctions in a fragmented forest landscape of Southern Amazonia. Biol Conserv 1113
124:383–396. 1114
Moura NG, Lees AC, Andretti CB, Davis BJW, Solar RRC, Aleixo A, Barlow J, Ferreira J, 1115
Gardner TA (2014) Avian biodiversity in multiple-use landscapes of the Brazilian 1116
Amazon. Biol Conserv 167:339-348. 1117
Myers N, Mittermeier RA, Mittermeier CG, Fonseca GAB, Kent J (2000) Biodiversity 1118
hotspots for conservation priorities. Nature 403:853–858. 1119
Nascimento H, Andrade A, Camargo J, Laurance WF, Laurance SG, Ribeiro J (2006) Effects 1120
of the surrounding matrix on tree recruitment in Amazonian forest fragments. Conserv 1121
Biol 20:853–860. 1122
Nascimento H, Laurance WF (2004) Biomass dynamics in Amazonian forest fragments. Ecol 1123
Applic 14:S127–S138. 1124
Neckel-Oliveira S, Gascon C (2006) Abundance, body size and movement patterns of a 1125
tropical tree frog in continuous and fragmented forests of the Brazilian Amazon. Biol 1126
Conserv 128:308–315. 1127
Nessimian JL, Venticinque EM, Zuanon J, De Marco P, Gordo M, Fidelis L, Batista J, Juen L 1128
(2008) Land use, habitat integrity, and aquatic insect assemblages in central 1129
Amazonian streams. Hydrobiol 614:117–131. 1130
Nogueira EM, Fearnside PM, Nelson BW, Barbosa RI, Keizer EWH (2008) Estimates of 1131
forest biomass in the Brazilian Amazon: New allometric equations and adjustments to 1132
biomass from wood-volume inventories. For Ecol Manage 256:1853-1857. 1133
Nogueira EM, Fearnside PM, Nelson BW, França MB (2007) Wood density in forests of 1134
Brazil’s ‘arc of deforestation’: Implications for biomass and flux of carbon from land-1135
use change in Amazonia. For Ecol Manage 248:119-135. 1136
Nogueira EM, Nelson BW, Fearnside PM (2005) Wood density in dense forest in central 1137
Amazonia, Brazil. For Ecol Manage 208:261-286. 1138
Norden N, Mesquita R, Bentos T, Chazdon R, Williamson GB (2011) Contrasting community 1139
compensatory trends in alternative successional pathways in central Amazonia. Oikos 1140
120:143-151 1141
Oliveira de AA, Mori SA (1999) A central Amazonian terra firme forest. I. High tree species 1142
richness on poor soils. Biodiv Conserv 8:1219–1244. 1143
Peres CA (2005) Why we need megareserves in Amazonia. Conserv Biol 19, 728–733. 1144
Peres CA, Gardner TA, Barlow J, Zuanon J, Michalski F, Lees A, Vieira I, Moreira F, Feeley 1145
KJ (2010) Biodiversity conservation in human-modified Amazonian forest landscapes. 1146
Biol Conserv 143:2314–2327. 1147
Phillips OL, Malhi Y, Higuchi N, Laurance WF, Nunez PV, Vasquez RM, et al. (1998) 1148
Changes in the carbon balance of tropical forests: Evidence from long-term plots. 1149
Science 282:439-442. 1150
Pimm SL (1998) The forest fragment classic. Nature 393:23–24. 1151
Pimm SL, Jenkins C (2005) Sustaining the variety of life. Sci Amer September:66–73. 1152
Powell AH, Powell GVN (1987) Population dynamics of male euglossine bees in Amazonian 1153
forest fragments. Biotropica 19:176–179. 1154
Puerta R (2002) Regeneração arbórea em pastagens abandonadas na região de Manaus em 1155
função da distância da floresta contínua. Scientia Forestalis 62:32–39. 1156
Quintela CE (1985) Forest Fragmentation and Differential Use of Natural and Man-made 1157
Edges by Understory Birds in Central Amazonia. M.Sc. Thesis, University of Illinois, 1158
Chicago. 1159
Quintero I, Roslin T (2005) Rapid recovery of dung beetle communities following habitat 1160
fragmentation in Central Amazonia. Ecology 86:3303–3311. 1161
Radtke MG, da Fonseca C, Williamson GB (2008) Forest fragment size effects on dung 1162
beetle communities. Biol Conserv 141:613–614. 1163
Rego F, Venticinque EM, Brescovit A (2007) Effects of forest fragmentation on four Ctenus 1164
spider populations (Araneae, Ctenidae) in central Amazonia, Brazil. Stud Neotrop 1165
Fauna Environ 42:137–144. 1166
Ribeiro MBN, Bruna EM, Mantovani W (2009) Influence of post-clearing treatment on the 1167
recovery of herbaceous plant communities in Amazonian secondary forests. Restor 1168
Ecol 18:50–58. 1169
Rocha R, López-Baucells A, Groenenberg M, Silva I, Farneda F, Bobrowiec P, Palmeirim 1170
JM, Meyer CFJ (2013) Effects of forest fragmentation on central Amazonian bats. 1171
Paper presented at the 16th International Bat Research Conference & 43rd North 1172
American Symposium on Bat Research, San José, Costa Rica. 1173
Salazar LF, Nobre CA, Oyama MD (2007) Climate change consequences on the biome 1174
distribution in South America. Geophys Res Lett 34:L09708. 1175
Sampaio EM, 2000. Effects of Forest Fragmentation on the Diversity and Abundance Patterns 1176
of Central Amazonian Bats. Ph.D. Dissertation, University of Tübingen, Germany. 1177
Sampaio EM, Kalko E, Bernard E, Rodriguez-Herrera B, Handley C, (2003) A biodiversity 1178
assessment of bats (Chiroptera) in a tropical lowland forest of central Amazonia, 1179
including methodological and conservation considerations. Stud Neotrop Fauna 1180
Environ 28:17–31. 1181
Santos BA, Arroyo-Rodríguez V, Moreno CE, Tabarelli M (2010) Edge-related loss of tree 1182
phylogenetic diversity in the severely fragmented Brazilian Atlantic forest. PLoS ONE 1183
5:e12625. 1184
Scariot A (1999) Forest fragmentation effects on diversity of the palm community in central 1185
Amazonia. J Ecol 87:66–76. 1186
Scariot A (2001) Weedy and secondary palm species in central Amazonian forest fragments. 1187
Rev Bot Brasil 15:271–280. 1188
Sizer N, Tanner EVJ (1999) Responses of woody plant seedlings to edge formation in a 1189
lowland tropical rainforest. Amazonia. Biol Conserv 91:135–142. 1190
Sizer N, Tanner EVJ, Kossman-Ferraz I (2000) Edge effects on litterfall mass and nutrient 1191
concentrations in forest fragments in central Amazonia. J Trop Ecol 16:853–863. 1192
Skole DS, Tucker CJ (1993) Tropical deforestation and habitat fragmentation in the Amazon: 1193
satellite data from 1978 to 1988. Science 260:1905–1910. 1194
Sodhi NS, Koh LP, Brook BW, Ng P (2004) Southeast Asian biodiversity: an impending 1195
disaster. Trends Ecol Evol 19:654–660. 1196
Stickler CM et al. (2009) The potential ecological costs and cobenefits of REDD: A critical 1197
review and case study from the Amazon region. Glob Change Biol 15:2803-2824. 1198
Stouffer PC, Bierregaard RO (1995a) Effects of forest fragmentation on understory 1199
hummingbirds in Amazonian Brazil. Conserv Biol 9:1085–1094. 1200
Stouffer PC, Bierregaard RO (1995b) Use of Amazonian forest fragments by understory 1201
insectivorous birds. Ecology 76:2429–2445. 1202
Stouffer PC, Bierregaard RO, Strong C, Lovejoy TE (2006) Long-term landscape change and 1203
bird abundance in Amazonian rainforest fragments. Conserv Biol 20:1212–1223. 1204
Stouffer PC, Strong C, Naka LN (2008) Twenty years of understory bird extinctions from 1205
Amazonian rain forest fragments: consistent trends and landscape-mediated dynamics. 1206
Divers Distrib 15:88–97. 1207
Stratford JA, Stouffer PC (1999) Local extinctions of terrestrial insectivorous birds in 1208
Amazonian forest fragments. Conserv Biol 13:1416–1423. 1209
Terborgh J, Lopez L, Nuñez VP, Rao M, Shahabuddin G, Orihuela G, Riveros M, Ascanio R, 1210
Adler G, Lambert T, Balbas L (2001) Ecological meltdown in predator-free forest 1211
fragments. Science 294:1923–1926. 1212
Ter Steege H, Pitman NCA, Sabatier D, Baraloto C, Salomao RP, Ernesto Guevara J, Phillips 1213
OL et al. (2013) Hyperdominance in the Amazonia flora. Science 342:326-334. 1214
Timo TPC (2003) Influência de fragmentação e matriz sobre a comunidade de mamíferos de 1215
médio e grande porte em uma floresta de terra firme na Amazônia central. M.Sc. 1216
Thesis, National Institute for Amazonian Research (INPA), Manaus, Brazil. 1217
Tocher M, Gascon C, Zimmerman BL (1997) Fragmentation effects on a central Amazonian 1218
frog community: a ten-year study. In: Laurance WF, Bierregaard RO (Eds), Tropical 1219
forest remnants: Ecology, management, and conservation of fragmented communities. 1220
University of Chicago Press, Chicago, pp 124–137. 1221
Trancoso R (2008) Hydrological Impacts of Deforestation in Small Catchments in Brazilian 1222
Amazonia. M.Sc. Thesis, National Institute for Amazonian Research (INPA), Manaus, 1223
Brazil. 1224
Uriarte M, Bruna EM, Rubim P, Anciães M, Jonckheere I (2010) Effects of forest 1225
fragmentation on the seedling recruitment of a tropical herb: assessing seed vs. safe-1226
site limitation. Ecology 91:1317–1328. 1227
Van Houtan KS, Pimm SL, Bierregaard RO, Lovejoy TE, Stouffer PC (2006) Local 1228
extinctions in Amazonian forest fragments. Evol Ecol Res 8:129– 148. 1229
Van Houtan KS, Pimm SL, Halley J, Bierregaard RO, Lovejoy TE (2007) Dispersal of 1230
Amazonian birds in continuous and fragmented forest. Ecol Lett 10:219–229. 1231
Vasconcelos HL, Luizão FJ (2004) Litter production and litter nutrient concentrations in a 1232
fragmented Amazonian landscape: edge and soil effects. Ecol Applic 14:884–892. 1233
Wilcox BA, Murphy DD (1985) Conservation strategy: the effects of fragmentation on 1234
extinction. Am Nat 125:879–887. 1235
Wieland LM, Mesquita RC, Bobrowiec PED, Bentos TV, Williamson GB (2011) Seed rain 1236
and advance regeneration in secondary succession in the Brazilian Amazon. Trop 1237
Conserv Sci 4:300-316. 1238
Williamson GB, Laurance WF, Oliveira A, Delamonica P, Gascon C, Lovejoy TE, Pohl L 1239
(2000) Amazonian wet forest resistance to the 1997–98 El Niño drought. Conserv Biol 1240
14:1538–1542. 1241
Williamson GB, Longworth JB, Bentos TV, Mesquita RC (2014) Convergence and 1242
divergence in alternative successional pathways in Central Amazonia. Plant Ecol 1243
Diversity 7:341-348. 1244
Williamson GB, Mesquita R (2001) Effects of fire on rain forest regeneration in the Amazon 1245
Basin. In: Laurance WF, Bierregaard RO (Eds), Tropical forest remnants: Ecology, 1246
management, and conservation of fragmented communities. University of Chicago 1247
Press, Chicago, pp 325–334. 1248
Zartman CE (2003) Forest fragmentation effects on epiphyllous bryophyte communities in 1249
central Amazonia. Ecology 84:948–954. 1250
Zartman CE, Nascimento HEM (2006) Are patch-tracking metacommunities dispersal 1251
limited? Inferences from abundance-occupancy patterns of epiphylls in Amazonian 1252
forest fragments. Biol Conserv 127:46–54. 1253
Zartman CE, Shaw AJ (2006) Metapopulation extinction thresholds in rainforest remnants. 1254
Am Nat 167:177–189. 1255
Figure 1. Map of the Biological Dynamics of Forest Fragments Project in central Amazonia. 1261
Figure 2. The diversity of edge-effect phenomena studied at the BDFFP and the distance to 1263
which each was found to penetrate into fragment interiors (after Laurance et al. 2002). 1264
Figure 3. Forest plots affected by two or more nearby edges (plot center <100 m from edge) 1266
suffer greater tree mortality (A) and have a higher density (B) and species richness (C) of 1267
disturbance-loving pioneer trees than do plots with just one nearby edge. Values shown are 1268
the mean ± SD (after Laurance et al. 2006a). 1269
Figure 4. Increasing divergence of tree-community composition in three fragmented 1271
Amazonian landscapes. Tree communities in forest-edge plots (<100 m from the nearest 1272
edge) are shown before forest fragmentation and 13–18 years after fragmentation, based on a 1273
single ordination of all plots and censuses in the study area. The ordination used importance 1274
values for all 267 tree genera found in the plots (after Laurance et al. 2007). 1275
Fig. 5. Tree genera that increase in forest fragments (those with positive values on the X axis) 1277
tend to have lower wood density than do those that decline in fragments (those with negative 1278
values on the X axis)(after Laurance et al. 2006b). 1279
Fig. 6. Striking increases in the density of 52 species in nine pioneer or early successional 1281
genera (Annona, Bellucia, Cecropia, Croton, Goupia, Jacaranda, Miconia, Pourouma, 1282
Vismia) near forest edges in the BDFFP study area (after Laurance et al. 2006a). 1283
... Our results have important implications for ecosystem function. Since the advent of functional trait network studies (e.g., TRY Plant Trait Database 38 ), understanding how plant species behave in terms of their physiological performance over large scales has led to important predictions of the consequences of future scenarios of climate and land-use change [39][40][41] . The importance of the Amazon region for the global climate and carbon cycle 42,43 highlights the need to devote substantial effort to investigating the taxonomy of hyperdominant plant taxa. ...
Full-text available
Recent studies have leveraged large datasets from plot-inventory networks to report a phenomenon of hyperdominance in Amazonian tree communities, concluding that few species are common and many are rare. However, taxonomic hypotheses may not be consistent across these large plot networks, potentially masking cryptic diversity and threatened rare taxa. In the current study, we have reviewed one of the most abundant putatively hyperdominant taxa, Protium heptaphyllum (Aubl.) Marchand (Burseraceae), long considered to be a taxonomically difficult species complex. Using morphological, genomic, and functional data, we present evidence that P. heptaphyllum sensu lato may represent eight separately evolving lineages, each warranting species status. Most of these lineages are geographically restricted, and few if any of them could be considered hyperdominant on their own. In addition, functional trait data are consistent with the hypothesis that trees from each lineage are adapted to distinct soil and climate conditions. Moreover, some of the newly discovered species are rare, with habitats currently experiencing rapid deforestation. We highlight an urgent need to improve sampling and methods for species discovery in order to avoid oversimplified assumptions regarding diversity and rarity in the tropics and the implications for ecosystem functioning and conservation.
... The combination of severe droughts and floods stress forests, especially if the flooding regimes of regularly inundated areas are perturbed outside of their natural range (Langerwisch et al., 2013). Amazonian droughts cause tree mortality and reduced biomass accrual (Feldpausch et al., 2016), facilitating fire (Aleixo B. et al., 2019), particularly at forest edges, where the microclimate is drier and hotter than in the continuous forest (Laurance et al., 2016). Large trees are most sensitive to these stressors, which is important because these trees regulate microclimate and forest C stocks. ...
Full-text available
The Amazon Basin is at the center of an intensifying discourse about deforestation, land-use, and global change. To date, climate research in the Basin has overwhelmingly focused on the cycling and storage of carbon (C) and its implications for global climate. Missing, however, is a more comprehensive consideration of other significant biophysical climate feedbacks [i.e., CH4, N2O, black carbon, biogenic volatile organic compounds (BVOCs), aerosols, evapotranspiration, and albedo] and their dynamic responses to both localized (fire, land-use change, infrastructure development, and storms) and global (warming, drying, and some related to El Niño or to warming in the tropical Atlantic) changes. Here, we synthesize the current understanding of (1) sources and fluxes of all major forcing agents, (2) the demonstrated or expected impact of global and local changes on each agent, and (3) the nature, extent, and drivers of anthropogenic change in the Basin. We highlight the large uncertainty in flux magnitude and responses, and their corresponding direct and indirect effects on the regional and global climate system. Despite uncertainty in their responses to change, we conclude that current warming from non-CO2 agents (especially CH4 and N2O) in the Amazon Basin largely offsets—and most likely exceeds—the climate service provided by atmospheric CO2 uptake. We also find that the majority of anthropogenic impacts act to increase the radiative forcing potential of the Basin. Given the large contribution of less-recognized agents (e.g., Amazonian trees alone emit ∼3.5% of all global CH4), a continuing focus on a single metric (i.e., C uptake and storage) is incompatible with genuine efforts to understand and manage the biogeochemistry of climate in a rapidly changing Amazon Basin.
... For example, the microclimate differs at the edge in comparison to the forest interior, on certain elements as sunlight penetrance, temperature, humidity and wind disturbance (Laurance 2004;Schwartz et al. 2017). This leads to differences in the forest structure, composition, and biomass (Laurance et al. 2011(Laurance et al. , 2016, that subsequently influence species, especially those that rely principally on plant resources. ...
Full-text available
From the forest edge to the forest interior, a small-scale gradient in the microclimate exists. Little is known about its influence on the abundance, diversity and morphological traits of insects in Amazonian forests, a major component of global terrestrial diversity. Our study investigates these traits in Arctiinae and Geometridae moths at the interior and the edge of a Peruvian lowland rainforest (Panguana field station, Puerto Inca Province). A total of 1286 Arctiinae and 2012 Geometridae specimens were collected, sorted according to DNA barcodes and identified using relevant type material. Moths’ assemblages at the forest edge differed significantly in their composition. At the forest edge, small-sized taxa (Lithosiini, Sterrhinae, Geometrinae) were less abundant whereas larger-sized Arctiini were more abundant. Moths were significantly larger at the forest edge than inside the forest, and these differences hold at subfamily and tribal level, possibly reflecting moth mobility, and abiotic conditions of habitats: larger moths might better tolerate desiccating conditions than smaller moths. A larger proportion of females was found at the forest edge, probably due to differences in the dispersal activity among sexes and/or in the tolerance to desiccation due to size. Our results revealed the edge effect on two rich herbivorous taxa in the Amazon basin. We provide a fully illustrated catalogue of all species as a baseline for further study and conservation purposes.
... Two of the most widespread forms of forest conversion are fragmentation, which results in isolated forest patches, and selective logging, which involved the extraction of a subset of forest trees. The impacts of fragmentation and logging on the structure and function of ecological communities has received attention since the dawn of conservation biology (reviews in Gibson et al., 2011;Laurance et al., 2016). Further, there is a broad consensus on traits correlated with vulnerability to forest modification (Burivalova et al., 2015). ...
Full-text available
The response of biodiversity to land-use change has been a central focus in applied ecological research for close to half a century. However, despite a vast body of literature, our understanding of how species' traits influence demographic vital rates in anthropogenically-modified habitats is remarkably scant. Such an understanding is crucial because vital rates determine population viability in modified habitats, and underlie emergent occupancy, abundance and community-level patterns. I used capture-recapture analyses to estimate variation in survival of birds in intact and logged tropical montane forest in the eastern Himalayas. In general, variation in body mass and alternative behavioral strategies (e.g., mixed-species flocking vs. solitary behavior) were not associated with survival differences in intact forest. However, year-round residents, and species that did not participate in mixed-species flocks had appreciably lower survival in logged forest compared with intact forest. Solitary foragers, for instance, faced a 30% decline in survival in logged forest compared with intact forest. Non-migratory habit and solitary foraging behavior might make species vulnerable to extinction in logged forest through reduced survival, an especially important process in influencing population viability. Identifying how species' traits modulate their response to land-use change is crucial to predict population responses to forest modification, and to better plan and manage biodiversity-friendly forest use.
... The role of the biosphere in the dynamics of the atmosphere has been widely studied, and the Amazonian forest is commonly recog-nized as an important factor in many aspects of CC, one that can contribute to either mitigate or increase the greenhouse effect (Bernstein et al., 2007;Dai, 2013;Laurence et al., 2016). The 2007 report of the IPCC expressly states that keeping the forest intact can significantly neutralize and mitigate greenhouse gases, particularly CO 2 . ...
Full-text available
2018 - Higuchi, M. I. G., Paz, D. T., Roazzi, A., & Souza B. C. (2018). Knowledge and Beliefs about Climate Change and the Role of the Amazonian Forest among University and High School Students. Ecopsychology, 10(2), 106-116. Doi: 10.1089/eco.2017.0050 // Abstract: Although the recognition of the role of forests in the reduction of polluting gases that produce climate change (CC) is widespread amongst scientists, society is far from adopting effective action to protect the forests, or from coping with the phenomenon. To help explain this impasse this article investigated the underlying psychosocial structure in the relation between knowledge, environmental beliefs, attitude towards climate change and the role of the Amazon forest in this phenomenon. Four hundred structured questionnaires were given to students from universities and high school in a city in Amazonas. The data was analysed using the Similarity Structure Analysis technique, it showed that knowledge about CC and the role of the forest in this phenomenon played a central modulating role in relation to the beliefs. It was possible to verify that each set of knowledge and beliefs end up inferring distinct attitudes that affect environmental behaviour towards CC. Keywords: climate change; Amazon forest; environmental beliefs; environmental attitudes; environmental behaviour.
This chapter delineates the characteristics of impacts and changes brought by the expansion of oil palm plantations to the biomass-rich interior region of Sarawak, Malaysia. It argues that many of the changes, both social and ecological, and the combination of the two, are derived from interfaces being formed when different and often distant landscapes, peoples, institutions and networks come into contact and are abruptly juxtaposed. Such new encounters have led to the temporal compression of succession—the transplanting, mobilisation, proliferation, reduction and extirpation of fauna, flora and human communities in a relatively short time. What emerges is a mixed landscape consisting of first nature and capitalist nature, where habitat fragmentation, biodiversity loss and multifaceted displacements proceed. Spatial compression brought by infrastructure development also connects the local community, both human and non-human, with distant people and markets, leading to a new kind of rural–urban continuum as well as the commodification of nature and labour. Along newly created commodity chains, there emerge numerous cultural encounters of individuals and social groups, adding a new social amalgam to the local community.
Full-text available
Background: Pathways (footpaths and roads) in forests are associated with edge effects, affecting forest structure and composition and associated wildlife. However, little is known about how edge effects along pathways may impact the dynamics of fruit production and their availability for frugivores. Aim: We related pathway width as a proxy for edge effects to fruit production. Our underlying hypothesis was that pathway width would be positively related to fruit production. Methods: We observed fruit production along three pathways of different widths – 2, 10 and 20 m wide – and in a control area of undisturbed forest in an Atlantic rain forest stand monthly over a 2-year period. Results: The number of species and individuals-bearing fruit was higher along the wider pathways than along the narrowest pathway and in the control area. The amount of zoochorous fruits was higher in the control area than along pathways, and the widest pathway had higher non-zoochorous fruits production. Fruiting peaks occurred along pathways, while fruiting in the control area was aseasonal. Conclusions: Pathway width is related to fruit type and its quantity and temporal availability. These effects extend towards the forest interior beyond 35 m. The presence of paths affects food resources for frugivores and thus can contribute to reconfiguring the spatio-temporal distribution of the fauna.
Full-text available
Background: Fragmentation can fundamentally alter the structure of tropical forests. However, the impacts of fragmentation may vary significantly among regions and lead to different outcomes. Aims: We examined the structure, composition and dynamics of a forest fragment in Singapore to investigate reasons for the apparent resilience of this forest to long-term isolation. Methods: We conducted 5 censuses of 12,688 trees ≥1-cm dbh in a 2-ha plot on the edge of the fragment between 1993 and 2012. Results: Stem density and basal area were not significantly different between 1993 and 2012 and were typical of other south-east Asian forests. However, there were short-term decreases in both variables after droughts in 1997 and 2009, both followed by recovery. Total mortality rate over the 19 years was 3.3% year⁻¹, considerably higher than other tropical forests in Asia, but it was balanced by high recruitment. The 10 most abundant species were primary forest species, pioneer species comprised <5% of all stems, and none of the 338 species in the plot was exotic. However, species abundances changed more than expected by chance for 86 species, and the rank order of the commonest species changed significantly. Species abundance changes were not related to known species traits. Conclusions: Despite the long period of isolation, we found a surprising level of resilience of the Bukit Timah forest. While the forest may be more sensitive to the effects of climatic fluctuations at decadal time scales, there were very few signs of forest degradation in this diverse fragment of tropical forest.
Full-text available
The prospect of accelerated global warming and the large-scale conversion of natural vegetation to agricultural land in the Amazon basin was the driving motive for the establishment of the Large-scale Biosphere–Atmosphere (LBA) experiment in 1998. The aim of this research programme, in collaboration with many others, has been to tackle the uncertainties with regard to the integrated effects of natural and man-made phenomena on the biophysical environment and, in turn, its feedback for land use/policy decisions. We provide an overview of the main issues treated in this book.
Full-text available
The western Amazon is the most biologically rich part of the Amazon basin and is home to a great diversity of indigenous ethnic groups, including some of the world’s last uncontacted peoples living in voluntary isolation. Unlike the eastern Brazilian Amazon, it is still a largely intact ecosystem. Underlying this landscape are large reserves of oil and gas, many yet untapped. The growing global demand is leading to unprecedented exploration and development in the region.
The vast extent of the Amazon Basin has historically restricted the study of its tree communities to the local and regional scales. Here, we provide empirical data on the commonness, rarity, and richness of lowland tree species across the entire Amazon Basin and Guiana Shield (Amazonia), collected in 1170 tree plots in all major forest types. Extrapolations suggest that Amazonia harbors roughly 16,000 tree species, of which just 227 (1.4%) account for half of all trees. Most of these are habitat specialists and only dominant in one or two regions of the basin. We discuss some implications of the finding that a small group of species—less diverse than the North American tree flora—accounts for half of the world’s most diverse tree community.
Although habitat fragmentation is a major threat to global biodiversity, the demographic mechanisms underlying species loss from tropical forest remnants remain largely unexplored. In particular, no studies at the landscape scale have quantified fragmentation’s impacts on colonization, extinction, and local population growth simultaneously. In central Amazonia, we conducted a multiyear demographic census of 292 populations of two leaf‐inhabiting (i.e., epiphyllous) bryophyte species transplanted from continuous forest into a network of 10 study sites ranging from 1, 10, and 100 to >10,000 ha in size. All populations experienced significantly positive local growth ( \documentclass{aastex} \usepackage{amsbsy} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{bm} \usepackage{mathrsfs} \usepackage{pifont} \usepackage{stmaryrd} \usepackage{textcomp} \usepackage{portland,xspace} \usepackage{amsmath,amsxtra} \usepackage[OT2,OT1]{fontenc} \newcommand\cyr{ \renewcommand\rmdefault{wncyr} \renewcommand\sfdefault{wncyss} \renewcommand\encodingdefault{OT2} \normalfont \selectfont} \DeclareTextFontCommand{\textcyr}{\cyr} \pagestyle{empty} \DeclareMathSizes{10}{9}{7}{6} \begin{document} \landscape $\lambda > 1$ \end{document} ) and a nearly constant per‐generational extinction probability (15%). However, experimental leaf patches in reserves of ≥100 ha experienced nearly double (48%) the colonization probability observed in small reserves (27%), suggesting that the proximate cause of epiphyll species loss in small fragments (≤10 ha) is reduced colonization. Nonetheless, populations of small fragments exhibit rates of colonization above patch extinction, positive local growth, and low temporal variation, which are features that should theoretically reduce the probability of extinction. This result suggests that for habitat‐tracking metapopulations subject to frequent and stochastic turnover events, including epiphylls, colonization/extinction ratios must be maintained well above unity to ensure metapopulation persistence.
We use Hubbell’s neutral theory to predict the impact of habitat fragmentation on Amazonian tree communities. For forest fragments isolated for about two decades, we generate neutral predictions for local species extinction, changes in species composition within fragments, and increases in the probability that any two trees within a fragment are conspecific. We tested these predictions using fragment and intact forest data from the Biological Dynamics of Forest Fragments Project in central Amazonia. To simulate complete demographic isolation, we excluded immigrants—species absent from a fragment or intact forest plot in the initial census but present in its last census—from our tests. The neutral theory correctly predicted the rate of species extinction from different plots as a function of the diversity and mortality rate of trees in each plot. However, the rate of change in species composition was much faster than predicted in fragments, indicating that different tree species respond differently to environmental changes. This violates the key assumption of neutral theory. When immigrants were included in our calculations, they increased the disparity between predicted and observed changes in fragments. Overall, neutral theory accurately predicted the pace of local extinctions in fragments but consistently underestimated changes in species composition.
A previous study by Phillips et al. of changes in the biomass of permanent sample plots in Amazonian forests was used to infer the presence of a regional carbon sink. However, these results generated a vigorous debate about sampling and methodological issues. Therefore we present a new analysis of biomass change in old-growth Amazonian forest plots using updated inventory data. We find that across 59 sites, the above-ground dry biomass in trees that are more than 10 cm in diameter (AGB) has increased since plot establishment by 1.22 ± 0.43 Mg per hectare per year (ha-1 yr-1), where 1 ha = 104 m2), or 0.98 ± 0.38 Mg ha-1 yr-1 if individual plot values are weighted by the number of hectare years of monitoring. This significant increase is neither confounded by spatial or temporal variation in wood specific gravity, nor dependent on the allometric equation used to estimate AGB. The conclusion is also robust to uncertainty about diameter measurements for problematic trees: for 34 plots in western Amazon forests a significant increase in AGB is found even with a conservative assumption of zero growth for all trees where diameter measurements were made using optical methods and/or growth rates needed to be estimated following fieldwork. Overall, our results suggest a slightly greater rate of net stand-level change than was reported by Phillips et al. Considering the spatial and temporal scale of sampling and associated studies showing increases in forest growth and stem turnover, the results presented here suggest that the total biomass of these plots has on average increased and that there has been a regional-scale carbon sink in old-growth Amazonian forests during the previous two decades.