Temporal patterns of deforestation and fragmentationin lowland Bolivia: implications for climate change

Jesús N. Pinto-Ledezma & Mary Laura Rivero Mamani

Received: 2 June 2011 /Accepted: 2 June 2013# Springer Science+Business Media Dordrecht 2013

Abstract The lowlands of eastern and northeastern Bolivia are characterized by a transi-tion between the humid evergreen forests of the Amazon Basin and the deciduous thorn-scrub vegetation of the Gran Chaco. Within this landscape lies one of the world’s bestpreserved areas: the ecoregion known as the Chiquitano dry forest, where deforestationpatterns over a 30 year period were analyzed. Results indicate that the area of the naturalcover was reduced from 97.21 % before 1976 to 82.10 % in 2008, causing significantchange in the landscape, especially in the spatial configuration of forest cover. The densityof forest fragments increased from 0.073 patches per 100 ha before 1976 to 0.509 in 2008,with a mean distance between patches of 151 and 210 m over the same period, leading to aconsiderable reduction in the fragment sizes, from 1,204 ha before 1976 to a mere 54 in2008. This pattern, observed in forests, does not occur in the savannas because, on onehand the savanna area is much lower compared to that of forests, and on the other becausethe deforestation process tended to be concentrated within forested areas. Based on theobserved patterns, it is possible that in the future the natural landscapes will be substitutedprincipally by anthropic landscapes, if there is no change in the economic and landdistribution policies. If this process continues, it will stimulate the expansion of mechanizedagriculture and the colonization of new areas, which will lead to further deforestation andlandscape fragmentation.

Climatic ChangeDOI 10.1007/s10584-013-0817-1

This article is part of a Special Issue on "Climate change and adaptation in tropical basins" edited by PierreGirard, Craig Hutton, and Jean-Phillipe Boulanger.

Electronic supplementary material The online version of this article (doi:10.1007/s10584-013-0817-1)contains supplementary material, which is available to authorized users.

J. N. Pinto-Ledezma (*) :M. L. Rivero MamaniDepartamento de Ecología, Museo de Historia Natural Noel Kempff Mercado, Universidad AutónomaGabriel René Moreno, Av. Irala 565, CC. 2489, Santa Cruz de la Sierra, Boliviae-mail: jesuspintoledezma@gmail.com

J. N. Pinto-LedezmaCarreras de Biología y Ciencias Ambientales, Universidad Autónoma Gabriel René Moreno,El Vallecito Km. 9 carretera al Norte, CC. 702, Santa Cruz de la Sierra, Bolivia

1 Introduction

The global impacts of the growing human population are reflected in extensive changes in thespatial patterns of land cover and land use (O’Neill et al. 1996; Dale et al. 2000). Since the lastmillennium, the human population rise is the strongest driver of change on the planet (Dale et al.2000) including the modification of the global climatic system, reduction in the stratosphericozone, alteration of biogeochemistry cycles, changes in the distribution and abundance ofbiological resources, and decreases in water quality (Meyer and Turner 1994; Fearnside 1995;Mahlman 1997). Anthropogenic, ecological, and land-surface processes interact and occur inlandscapes at multiple scales (Southworth et al. 2006).

The rapid destruction and degradation of tropical forest is an important source of greenhousegas emissions and could play an important role in exacerbating global warming (Dale 1997;Fearnside 2000; Houghton et al. 2000; Lawton et al. 2001; Fearnside and Laurance 2004). Oneof the main drivers of the transformation of earth by humans is the effort to provide food,shelter, and other products for human consumption which deeply affect biological and physicalsystems (Kates et al. 1990). In fact, changes in land use decrease the ability of the planet tocontinue providing goods and services on which humans depend, as they result in the loss offlora, fauna and ecosystems (Spies et al. 1994; Murcia 1995; Laurance and Bierregaard 1997;Gascon et al. 1999; Pinto-Ledezma and Ruiz 2010).

At a global level, some estimates suggest that 9 million km2 of tropical humid forests havebeen lost in the last 50 years and that the current extinction rate of fauna and flora is not onlyhigh but accelerating (Pimm et al. 2001). Achard et al. (2002) estimated an annual deforestationrate of 0.38 % of humid tropical forest in Latin America. Currently in Latin America,deforestation and fragmentation are the main causes of ecosystem deterioration (Dirzo andGarcia 1992; Pacheco 1998; Steininger et al. 2001; Camacho et al. 2001; Pinto-Ledezma andRuiz 2010). Bolivia still contains around 400,000 km2 of intact tropical forests, which corre-sponds to 90 % of its original tropical forest cover (Killeen et al. 2007). Yet, deforestation israpidly advancing at an annual rate of around 0.5 %, and Bolivia is classified as a country at theforest frontier (Angelsen 2007).

In Bolivia, one of the greatest impacts is in the lowlands of the Department of Santa Cruz,which comprises over a third of the total area of the country (37 million hectares) an in recentdecades has suffered from very rapid changes in the cover and structure of its forests (Steiningeret al. 2001; Killeen et al. 2007, 2008; Pinto-Ledezma and Ruiz 2010). These changes in forestcover and structure are mainly produced by the high rate of human immigration and consequentactivities, such as mechanical agriculture (Pacheco 1998, 2006; Killeen et al. 2007; Pinto-Ledezma and Ruiz 2010). The development of the agricultural frontier in Santa Cruz beganonly recently, in the late 1950s, driven by small and large scale farmers seeking mainly tosupply agricultural goods to the domestic market. Since the mid-1980’s, producers have lookedfor ways to increase the exports of some commodities, primarily soybeans to regional markets.The expansion of the agricultural frontier in Santa Cruz, with the subsequent deforestation, hasbeen accelerated since the mid-1980s as a result of shifts in the government’s macroeconomicpolicy, and efforts to open up external markets, in order to increase the contribution ofagricultural exports to national revenues (Pacheco 1998).

In this paper, we focus on the Chiquitano Ecoregion (hereafter, “Chiquitanía”), located inthe eastern of Bolivian lowlands, within the Department of Santa Cruz (Fig. 1), an area ofapproximately 20,000 km2 of deciduous forest. Here, mechanized cash crop production issteadily expanding, mainly for the cultivation of soybeans for export (Hecht 2005), such thatthe department of Santa Cruz has experienced some of the greatest rates of land-use changein all of Latin America. The region subjected to greatest deforestation is the fertile alluvial

Climatic Change

plain situated to the east of the Andes where the cultivation of soybeans and other crops hasexpanded exponentially since the early 1990s (Pacheco 1998; Steininger et al 2001; Mertenset al. 2004; Kaimowitz et al 2002).

Based on the situation described above, the central goal of this article is to quantify andreport on the long term deforestation and fragmentation of the region. We focus on the landuse practices utilized over a period of >30 years in one of the best-preserved patches ofNeotropical Dry Forest in the world—the Chiquitano dry forest of the Chiquitanía—whichcan serve as a model for the implementation of policies intended to reduce deforestation and

Fig. 1 Study area, shows the location of the Chiquitano Dry Forest in Bolivia, also shown the main cities,rivers and roads

Climatic Change

protect biodiversity. We also report on the spatial distribution of forest-savanna cover for theentire Chiquitanía.

2 Study area and methods

2.1 Lowland Bolivia and Chiquitanía

The eastern and northeastern portion of Bolivia is composed of vast lowlands with anelevation mostly below 500 m. Vegetation is composed of both forest and grasslands, andhumidity and moisture decrease from north to south. The forests of eastern lowland Boliviaare situated across a zone of climatic transition between the humid evergreen forests of theAmazon in the north and the deciduous thorn-scrub vegetation of the Gran Chaco in thesouth (Killeen et al. 2006).

The Chiquitano dry forest is a term used to describe a complex of forest communities thatoccur across the climatic transition described above (Killeen et al. 1998), representing what isprobably the largest extant patch of what is now broadly recognized as the Neotropical seasonaldry tropical forest complex (Prado and Gibbs 1993; Prado 2000; Killeen et al. 2006). TheChiquitano dry forest ranges from completely deciduous in the south to semi-deciduous in thenorth, while the degree of deciduousness in the intervening areas is highly variable dependingon the amount of precipitation that falls within any given year at any given place.

2.2 Image processing and deforestation and fragmentation analysis methodology

To analyze the spatial and temporal patterns of deforestation and fragmentation in the Chiquitanía,a set of five mosaics of Landsat scenes (pre1976, 1986, 1992, 2001, and 2008) were used. Thedetails of the image processing and the analysis of deforestation and fragmentation can be foundin the Electronic Supplementary Material.

Shortly, a supervised classification was conducted using the result of two field visits wheregeo-referenced data of the main types of coverage were collected. The classification processresulted in three thematic classes: forest, savanna and deforested (Fig. 2). The term “defores-tation” is used here when there is a replacement of natural cover (forest and savannas) bycultivated pastures, agricultural fields, urban areas and/or human settlements (Dirzo and Garcia1992; Pinto-Ledezma and Ruiz 2010). The term “fragmentation” refers to a disruption in thecontinuity of natural covers (Lord and Norton 1990), and the subdivision of landscapes intosmaller units (Laverty and Gibbs 2007).

3 Results and discussion

3.1 Deforestation patterns and land cover changes

Deforestation affects all habitats in lowland Bolivia, as well as natural disturbances (e.g. floods,wildfires. The savanna cover (Cerrado and Chaco woodland) did not show substantial loss todeforestation, with a change of only 2.66 % until 2008 (Table 1), However, the total land areadeforested increased from 2.79 % pre1976 to 17.89 % in 2008, with intermediate values of4.33 % (1986), 7.25 % (1992), and 13.96 % (2001). This pattern can also be observed by highdeforestation rates (r) in the study area (Fig. 2 and Table 1) during this period, when theAmazonand Chiquitano dry forest had the highest losses (12.34 %) within the region, until 2008.

Climatic Change

The factors that drive the alteration of natural landscapes are diverse, involving both naturaland anthropic changes (Skole and Tucker 1993; Steininger et al. 2001; Pacheco and Mertens2004; Killeen et al. 2007; Müller et al. 2010), although the latter is the most frequent and resultsin a major impact on the Bolivian lowlands (Steininger et al. 2001; Killeen et al. 2007; Pinto-Ledezma andRuiz 2010). In this sense, themain cause of the deforestation and fragmentation inthis region are demographic factors (primarily immigration), which increased due to govern-ment policies between 1950 and 1980. During this period, the importation of agricultural goodswas replaced by domestic production, mainly rice, cotton, and sugarcane (Pacheco andMertens2004; Pacheco 2006; Pinto-Ledezma andRuiz 2010), causing an increase in the region’s humanpopulation. As a result, the Chiquitanía currently has one of the highest population growth ratesin Bolivia (INE 2005). As a consequence of these high immigration rates and the resulting useof soil in the study area, there is a decrease of the original covers (especially forests), andchanges in their spatial configuration (see supplemental material for detailed information).

3.2 Forest and savanna fragmentation

During the study period, there have been changes in the size, number, distance and spatialdistribution of fragments, with different patterns in forests and savannas. As the study areabecame increasingly deforested, there were synchronous effects on the remaining forest-savanna habitat (Table 2). In this sense, we examined the behavior of some landscape metricsat different temporal scales (pre1976, 1992, and 2008), and at the class level (Table 2).

The spatial configuration of the forest changed substantially between 1976 and 2006 inthe study area, and the mean size of forest patches decreased considerably as a function of an

Table 1 Areas in hectares of natural habitats (forest and savannas) and deforestation (ha and % approxi-mately) for the five periods of study and entire Chiquitano Ecoregion. Also shown rates of total deforestationfor the five periods analyzed

Naturalhabitat

% naturalhabitat

Totalnaturalhabitat

% totalnaturalhabitat

Totaldeforestation

Totaldeforestation (%)

r r (%)

Pre1976

Forest 13082553 79.76 15836723 97.20 457025 2.79 −0.001 −0.135a

Savanna 2754170 16.79

1986

Forest 12847943 78.33 15583817 95.64 709931 4.33 −0.002 −0.161Savanna 2735874 16.68

1992

Forest 12437493 75.82 15104136 92.70 1189612 7.25 −0.005 −0.521Savanna 2666643 16.26

2001

Forest 11524984 70.26 14004458 85.95 2289290 13.96 −0.008 −0.840Savanna 2479474 15.12

2008

Forest 11059117 67.42 13377521 82.10 2917190 17.89 −0.007 −0.654Savanna 2318404 14.13

aRepresent the annual deforestation rate in the first period of study, the starting year of the first period is

considered 1950, year where performed the first national census of agriculture (Bethell 2008)

Climatic Change

increase in deforestation, causing an increase in the density of patches and in the distancebetween them, as the forest underwent substantial fragmentation. These results corroborateother analyses of change in the cover and spatial patterns of the forest (Gavier and Bucher2004; McGarigal et al. 2001. In 1976 the density was 0.073, and increased to 0.509patches/100 ha in 2008. This pattern is not repeated in savannas, since the density of patchesof savanna decreased (0.229 in pre1976 to 0.183 in 2008). This decrease may be due to thefact that savannas are naturally fragmented. This increase and decrease in patch density is

Table 2 Representation of spatial pattern at class level (forest and savannas) for three periods of study(pre1976, 1992, and 2008), based on six landscape metrics for the central Chiquitanía (San Julián district)

PLAND NP MPS PD MNN IJI DI

pre1976

Forest 92.519 415 1204 0.073 151 61.625 0.071

Savanna 5.804 1296 24 0.229 389 25.697 0.567

1992

Forest 71.931 1622 241 0.287 158 59.106 0.288

Savanna 5.815 1411 22 0.249 367 28.381 0.582

2008

Forest 27.333 2883 54 0.509 210 14.071 0.682

Savanna 4.389 1038 24 0.183 170 24.735 0.198

PLAND Percentage of landscape, NP number of patches,MPS mean patch size, PD patch density,MNN meannearest neighbor distance, IJI interspersion and juxtaposition index, and DI dispersion index

Fig. 2 Maps derived from satellite images of medium resolution (Landsat TM and ETM). Forest (gray),savannas (clear gray), and deforested areas (black) for the five periods of study

Climatic Change

also evident in the mean nearest neighbor distance between patches (MNN), which in pre1976 had a value of 151 m for forest patches, and of 389 m for patches of savanna, whichwas increasing along with the increase in patch density (Table 2). By the year 2008, there was amean distance between forest patches of 210 m, but of 170 m for patches of savanna, which wasexpected because savanna patch density decreased (PD=0.183). Additionally, there was anincrease in themean size of savanna patches (24 ha.) relative to the previous period (22 ha.), andin the aggregated distribution of patches of savanna (Fig. 2 and Table 2). This is also explainedby a general pattern of extensive patches being divided into smaller areas. The mean size offorest patches (MPS) decreased year by year as a function of the deforested surface area, froman area of 1204 ha in pre 1976, to 54 ha in 2008. There was a slight decrease in the size ofpatches of savanna up to 1992 (from 24 ha in pre1976 to 22 ha in 1992), with the peculiarity thatthere was a recovery in the size of the patches in 2008 (Table 2).

The index of dispersion shows that for all the years analyzed, the distribution of forestpatches and savannas was aggregated, although as years go by this pattern tended to becomerandom, especially in the case of forests, where the index of dispersion (ID) increased from0.071 in pre1976 to 0.682 in 2008. Savannas practically did not change in the first two studyperiods (pre1976 and 1991), showing little increase by 1991; however, in 2008 the ID valuedecreased to 0.198 and patches tended to become more aggregated, which corroborates theresults presented above (Table 2).

The index of interspersion and juxtaposition (IJI) indicates a decrease in the mixing ofpatches over time, especially in forests, where it changed from 61.625 in pre1976 to 14.071 in2008, which helps to explain changes in landscape patterns measured with other metrics(Table 2). This is evident in 2008, when neighboring patches of forest and savanna tended tolose connections and to be distributed at random, showing a high level of landscape fragmen-tation in the study area.

Notably, a clear growth in deforestation rates from pre1976 to 2008 is evident, especiallyin the last periods (2001 and 2008), which is the same or greater than deforestation ratesreported at the national level (Camacho et al. 2001; Steininger et al. 2001; Killeen et al.2007), or than those of other tropical and subtropical regions of the world (Dirzo and Garcia1992; Skole and Tucker 1993; Achard et al. 2002; Staus et al. 2002; Hansen et al. 2009).

For savannas, patterns of deforestation and fragmentation were not observed to be substantial,possibly because of: i) the initial reduced surface of the savannas in relation to that of forest(Table 2; % of natural habitat), ii) the spatial distribution of savannas (Table 2), iii) theinaccessibility to land, and iv) an increase in deforestation, resulting on one hand in a decreaseof savanna patch density and on the other hand, in the disappearance of patches of savannah(Fig. 3 and Table 2), causing the formation of savanna relicts. Even if a substantial decrease insavanna surface (PLAND=4389) and patch size (MPS=24) by the year 2008 was not observed,substantial changes in other metrics are evident (Table 2), which explains the observed formationof savanna relicts (Fig. 2), as well as a decrease in the distance between patches of savanna(PD=0.183; MNN=170), resulting in an aggregated distribution of these patches (ID=0.198).

4 Climate change implications

By the year 2100, global mean temperatures could increase from 1° to 4.2° C (IPCC 2001;Solomon et al. 2007). In Bolivia, major warming is expected in the Amazon region (IPCC 2001;Seiler 2009; Andersen and Mamani 2009), a situation that implies changes in temporal andspatial precipitation and evapotranspiration (Watson et al 1997;Mulligan 2000). The increase intemperature expected in the Bolivian lowlands by 2030 is of about 1.3 °C (SRES scenario A2;

Climatic Change

for more detailed information, see Seiler 2009). By 2100, temperature increases are predicted tobe about 4.7 °C, with the strongest temperature increases occurring around August andSeptember (Seiler 2009). In addition, the lowlands show a latitudinal temperature gradient,with a relative increase in temperature in a northerly direction. This tendency becomes evenmore pronounced in 2100 (Seiler 2009). Additionally, the precipitation cycle is predicted tointensify, with more precipitation during the rainy season and less precipitation during the dryseason. Most maximum relative increases of precipitation will occur from April to June andmost maximum relative decreases of precipitation will be from July to August (Seiler 2009). By2030 (SRES scenarios A2 and B2), the precipitation is predicted to decrease by about −28 % inthe Bolivian lowlands, and by 2100 the maximum reduction in precipitation is by about −36 %(Seiler 2009), similar to what happened in the 60’s and early 70’s in the past century (seesupplemental material for examples). Finally, by 2030 an annual net increase of precipitationwill occur in the southern lowlands (mainly in the dry forest).

The rapid destruction and degradation of tropical forest is considered a major source ofgreenhouse gases such as carbon dioxide, methane, and nitrous oxide, and could play animportant role in exacerbating global warming (Dale 1997; Fearnside 2000; Houghton et al.2000; Fearnside and Laurance 2004). This destruction and degradation is caused mainly byland use change and expansion of agriculture (Grau et al. 2005; Pinto-Ledezma and Ruiz2010), and is related to climate change as both a causal factor and a major way in which theeffects of climate change are expressed (Dale 1997). Recently, climate change after tropicaldeforestation has been studied using several approaches, including global climate modeling,regional climate modeling, theoretical approaches, and field observations (Berbet and Heil

0

10

20

30

40

50

60

70

80

90

100

pre1976 1986 1992 2001 2008

% o

f to

tal a

rea

Forest

Savanna

Deforestation

0

2

4

6

8

10

12

14

16

18

20

a

b

1986 1992 2001 2008

% o

f d

efo

rest

atio

n

Forest

Savanna

Fig. 3 a Dynamics offorest-savanna habitat anddeforestation for the fiveperiods of study, values inpercent show how forest andsavanna lost their surfacesand how the deforestationincrease, b Change in time(% of deforestation) offorest and savanna habitat

Climatic Change

2003). These studies explain changes in precipitation after the conversion of tropical forest(Eltahir 1996), demonstrating that precipitation change over deforested areas is not uniform(Berbet and Heil 2003), and that differences in precipitation are seasonal. However, theincrease in the surface albedo after deforestation causes a reduction in the net surface radiation,which cools the upper atmosphere over the deforested area, reducing aerodynamic roughnesslength and mechanically turbulent mixing in the boundary layer. Additionally, a reduction inevapotranspiration, and an increase in the ratio of connective sensible heat transfers latent heatfrom the surface to the atmosphere, inducing a thermally driven circulation that results insubsidence (Lawton et al. 2001; Berbet and Heil 2003). This change in surface albedo has beenconsidered the most important driver of climate change-induced deforestation (Dirmeyer andShukla 1994; Nouvellon et al. 2000; Berbet and Heil 2003).

The direct contribution of tropical deforestation to atmospheric GHG has been recog-nized (Naughton-Treves 2004), but the net role of the tropics in the global carbon cycleremains uncertain (De Jong et al. 2000; Houghton et al. 2000; Naughton-Treves 2004).Every year, between 1.6 and 2.4 Pg of carbon are released to the atmosphere from tropicalforest clearing (De Jong et al. 2000; Fearnside 2000). By these estimates, tropicaldeforestation accounts for roughly 20–29 % of global anthropogenic greenhouse gasemissions (Watson et al. 2000; Solomon et al. 2007). In addition, inter-annual variationin climate and atmospheric CO2 concentrations alter carbon uptake rates and forest-savanna covers, increasing their flammability (Nepstad et al. 2001; Naughton-Treves2004) and tree mortality, allowing for greater light penetration and lowering humiditylevels in the understory (Killeen et al. 2007). Similarly, land use change commonly alterssurface soil structure by compaction and thus reduces infiltration of rainfall and increases runoff,with the end result of reducing soil moisture and increasing the chance of wildfire (Lawton et al.2001), which is why understanding the dynamics of land use is key for more accurate modelingof global climate.

4.1 Sustainable land use and forest management as an alternative

In the Bolivian lowlands, deforestation, not climate change, is the major immediate threat tothe forest and its biodiversity. Deforestation is the main source of Bolivian GHG emission(10 tCO2/person) (Andersen and Mamani 2009), and if immigration, road construction andglobalization of the agricultural economy continue, we expect high rates of deforestation tocontinue. As a result, it is necessary to develop novel resource use plans, and generate newpolicies to improve conservation. Sustainable land use and forest management (Bucher andHuszar 1999) become the better alternative, because the present management system isdegrading the resource base and is therefore unsustainable. This is important because theprocess of landscape transformation observed in the study area is a risk for one of the mostimportant ecosystems in the Bolivian lowlands, from ecological, functional and economicpoints of view. Therefore, the implementation of an adequate management system with theability to protect land, enhance the resource base and provide higher net returns in asustainable manner, is an alternative to reduce the advancement of deforestation and itsinfluences on climate change. Additionally, the development of social policies is likelynecessary to overcome the resistance of campesinos (i.e., indigenous Andean colonists) andlarger producers to this management system (Huszar 1999), because management requiresan initial investment that may be uneconomical in the short-term to campesinos and evenlarger producers (Bucher and Huszar 1999).

Finally, it is important to enforce existing laws and if necessary offer economic and taxincentives to promote forest conservation, including compensation for ecological services.

Climatic Change

Acknowledgments We are grateful to the reviewers for their contributions. We thank Alex E. Jahn and DanielVillarroel for helpful comments on earlier versions of this manuscript. We are also grateful to the Museo deHistoria Natural Noel Kempff Mercado for sharing their extensive data set on deforestation. This manuscript wasfunded by the CNPq fellowship EXP-C (Projeto Sinergia-nº6 do CTHidro) to JNPL.

References

Achard F, Eva HD, Stibig HJ, Mayaux P, Gallego J, Richards T, Malingreau JP (2002) Determination ofdeforestation rates of the world’s humid tropical forests. Science 297:999–1002

Andersen LE, Mamani R (2009) Cambio climático en Bolivia hasta el 2100: síntesis de costos y oportunidades.ERECC-SA, CEPAL. 41 p

Angelsen A (2007) Forest cover change in space and time: combining the von Thünen and forest transitiontheories. World Bank Research Working Paper 4117. World Bank, Washington DC

Berbet MLC, Heil M (2003) Climate change after tropical deforestation: seasonal variability of surface albedoand its effects on precipitation change. J Clim 16:2099–2104

Bethell L (2008) The Cambridge history of Latin America: Latin America since 1930 spanish South America.Vol. 8. Cambridge University Press, Cambridge, UK, p 883

Bucher EH, Huszar PC (1999) Sustainable management of the Gran Chaco of South America: ecologicalpromise and economic constraints. J Environ Manag 57:99–108

Camacho O, Cordero W, Martínez I, Rojas D (2001) Tasa de deforestación del Departamento de Santa Cruz1993–2000. Superintendencia Forestal, BOLFOR. Santa Cruz de la Sierra, Bolivia. 43 p

Dale VH (1997) The relationship between land-use change and climate change. Ecol Appl 7:753–769Dale VH, Brown S, Haeuber RA, Hobbs NT, Huntly N, Naiman RJ, Riebsame WE, Turner MG, Valone TJ

(2000) Ecological principles and guidelines for managing the use of land. Ecol Appl 10(3):639–670De Jong B, Ochoa-Gaona S, Castillo-Santiago M, Ramirez-Marcial N, Carins A (2000) Carbon flux and

patterns of land-use/land-cover change in the Selva Lacandona, Mexico. Ambio 29(8):504–511Dirmeyer PA, Shukla J (1994) Albedo as a modulator of climate response to tropical deforestation. J Geophys

Res 99:20863–20877Dirzo R, Garcia M (1992) Rates of deforestation in Los Tuxlas, a Neotropical area in southeast México.

Conserv Biol 6(1):91–100Eltahir EAB (1996) Role of vegetation in sustaining large-scale atmospheric circulations in the tropics. J

Geophys Res 101:4255–4268Fearnside PM (1995) Potential impacts of climate change on natural forest and forestry in Brazilian

Amazonia. For Ecol Manag 78:51–70Fearnside PM (2000) Global warming and tropical land use change: greenhouse gas emissions from biomass

burning, decomposition and soils in forest conversion, shifting cultivation and secondary vegetation.Clim Chang 46:115–158

Fearnside PM, LauranceWF (2004) Tropical deforestation and greenhouse-gas emissions. Ecol Appl 14(4):982–986Gascon C, Lovejoy TE, Malcom JR (1999) Matrix habitat and species richness in tropical forest remnants.

Biol Conserv 91:223–229Gavier GI, Bucher EH (2004) Deforestación de las Sierras Chicas de Córdoba (Argentina) en el período 1970–

1997. Miscelánea Nº 101. Academia Nacional de Ciencias (Córdoba)Grau HR, Gasparri NI, Aide TM (2005) Agriculture expansion and deforestation in seasonally dry forest of

north-west Argentina. Environ Conserv 32(2):140–148HansenMC, Stehman SV, Potapov PV, Arunarwati B, Stolle F, Pittman K (2009) Quantifying changes in the rates

of forest clearing in Indonesia from 1990 to 2005 using remotely sensed data sets. Environ Res Lett 4:1–12Hecht S (2005) Soybeans, development and conservation on the Amazon frontier. Dev Chang 36(2):375–404Houghton RA, Skole DL, Nobre CA, Hacler JL, Lawrence KT, Chomentowski WH (2000) Annual fluxes of

carbon from deforestation and regrowth in the Brazilian Amazon. Nature 403:301–304Huszar PC (1999) Justification for using soil conservation incentives. In: Sanders DW, Huszar PC,

Sombatpanit S, Enters T (eds) Incentives in soil conservation: from theory to practice. Science Publishers,Enfield, pp 57–68

INE (Instituto Nacional de Estadística) (2005) Atlas estadístico de municipios: Bolivia un mundo depotencialidades. MDSP, COSUDE, PNUD, La Paz, 697 p

IPCC (2001) Climate change 2001: the scientific basis. Cambridge University PressKaimowitz D, Mendez P, Puntodewo A, Vanclay J (2002) Spatial regression analysis of deforestation in Santa

Cruz, Bolivia. In: Wood CH, Porro R (eds) Land use and deforestation in the Amazon. University Press ofFlorida, pp. 41–65

Climatic Change

Kates RW, Clark WC, Norberg-Bohm V, Turner II BL (1990) Human sources of global change: a report onpriority research initiatives for 1990–1995. Discussion paper G-90-08. Global Environmental PolicyProject, John F. Kennedy School of Government, Harvard University, Cambridge, Massachusetts, USA

Killeen TJ, Jardim A, Mamini FM, Rojas N, Saravia P (1998) Diversity, composition, and structure ofatropical semideciduous forest in the Chiquitanía region of Santa Cruz, Bolivia. J Trop Ecol 14:803–827

Killeen TJ, Chavez E, Peña-Claros M, Toledo M, Arroyo L, Caballero J, Correa L, Guilleén R, Quebedo R,Saldías M, Soria L, Uslar Y, Vargas I, Steininger M (2006) The Chiquitano dry forest, the transitionbetween humid and dry forest in Eastern Lowland Bolivia. In: Pennington (ed) Neotropical savannas andseasonally dry forests plant diversity biogeography and conservation. CRC Press, England, pp 213–233

Killeen TJ, Calderon V, Soria L, Quezada B, Steininger MK, Harper G, Solorzano LA, Tucker CJ (2007)Thirty years of land-cover change in Bolivia. Ambio 36(7):600–606

Killeen TJ, Guerra A, Calzadilla M, Correa L, Calderón V, Soria L, Quezada B, Steininger MK (2008) Totalhistorical land-use change in eastern Bolivia: who, where, when, and how much? Ecol Soc 13(1):36

Laurance WF, Bierregaard RO Jr (1997) Tropical forest remnants: ecology, management, and conservation offragmented Communities. The University of Chicago Press, Chicago, 528 p

Laverty MF, Gibbs JP (2007) Ecosystem loss and fragmentation. Lessons Conserv 1:72–96Lawton RO, Nair US, Pielke RA, Welch RM (2001) Climatic impact of tropical lowland deforestation on

nearby montane cloud forests. Science 294:584–587Lord JM, Norton DA (1990) Scale and the spatial concept of fragmentation. Conserv Biol 4:197–202Mahlman JD (1997) Uncertainties in projections of human caused climate warming. Science 278:1416–1417McGarigal K, Romme WH, Crist M, Roworth E (2001) Cumulative effects of roads and logging on landscape

structure in the San Juan Mountains, Colorado (USA). Landsc Ecol 16:327–349Mertens B, Kaimowitz D, Puntodewo A, Vanclay J, Mendez P (2004) Modeling deforestation at distinct

geographic scales and time periods in Santa Cruz, Bolivia. Int Reg Sci Rev 27(3):271–296Meyer WB, Turner BL II (1994) Changes in land use and land cover: a global perspective. Cambridge

University Press, CambridgeMüller R, Müller D, Scheirhorn F, Gerold G (2010) Spatiotemporal modeling of the expansion of mechanized

agriculture in Bolivia. Appl Geogr 31(2):631–640Mulligan M (2000) Downscaled climate change scenarios for Colombia and their hydrological consequences.

Adv Environ Monit Model 1(1):3–35Murcia C (1995) Edge effects in fragmented forests: implications for conservation. Trends Ecol Evol 10(2):58–62Naughton-Treves L (2004) Deforestation and carbon emissions at tropical frontiers: a case study from the

Peruvian amazon. World Dev 32(1):173–190Nepstad D, Carvalho G, Barros A, Alencar A, Capobianco J, Bishop J, Moutinho P, Lefebvre P, Lopes SU,

Prins E (2001) Road paving, fire regime feedbacks, and the future of Amazon forests. For Ecol Manag154:395–407

Nouvellon Y, Bégué A, Moran MS, Seen DL, Rambal S, Luquet D, Chehbouni G, Inoue Y (2000) PAR extinctionin shortgrass ecosystems: effects of clumping, sky conditions and soil albedo. Agric For Meteoroly 105:21–41

O’Neill RV, Hunsaker CT, Timmins SP, Jackson BL, Jones KB, Riitters KH, Wickham JD (1996) Scaleproblems in reporting landscape pattern at the regional scale. Landsc Ecol 11:169–180

Pacheco P (1998) Estilos de desarrollo, deforestación y degradación de los bosques en las tierras bajas deBolivia. Centro para los estudios laboral y agrario. La Paz. 376 pp

Pacheco P (2006) Agricultural expansion and deforestation in the lowlands Bolivia: the import substitutionversus the structural adjustment model. Land Use Policy 23:205–225

Pacheco P, Mertens B (2004) Land use change and agricultural development in Santa Cruz, Bolivia. Boit etForests des Tropiques 280(2):29–40

Pimm SL, Ayres M, Balmford A, Branch G, Brandon K, Brooks T, Bustamante R, Costanza R, Cowling R,Curran LM, Dobson A, Farber S, da Fonseca GAB, Gascon C, Kitching R, McNeely J, Lovejoy T,Mittermeier RA, Myers N, Patz JA, Raffle B, Rapport D, Raven P, Roberts C, Rodríguez JP, Rylands AB,Tucker C, Safina C, Samper C, Stiassny MLJ, Supriatna J, Wall DH, Wilcove D (2001) Can we defynature’s end? Science 293:2207–2208

Pinto-Ledezma JN, Ruiz T (2010) Deforestación y fragmentación 1976–2006 en el municipio de San Julián(Santa Cruz, Bolivia). Ecología en Bolivia 45(2):101–115

Prado DE (2000) Seasonally dry forests of tropical South America: from forgotten ecosystems to a newphytogeographic unit. Edinb J Bot 57:437–461

Prado DE, Gibbs PE (1993) Patterns of species distributions in the dry seasonal forests of South America. AnnMo Bot Gard 80:902–927

Seiler Ch (2009) Implementation and validation of a regional climate model for Bolivia. Friends of NatureFoundation, Santa Cruz de la Sierra, Bolivia. 38 p

Skole D, Tucker CJ (1993) Tropical deforestation and habitat fragmentation in the Amazon: satellite data from1978 to 1988. Science 260:1905–1910

Climatic Change

Solomon S, Qin D, Manning M, Chen Z, Marquis M, Avery KB, Tignor M, Miller HL (eds) (2007) Climatechange 2007: the physical science basis. Working Group I Contribution to the Fourth assessment report ofthe Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge

Southworth J, Cumming GS, Marsik M, Binford MW (2006) Linking spatial and temporal variation atmultiple scales in a heterogeneous landscape. Prof Geogr 58(4):406–420

Spies TA, Ripple WJ, Bradshaw GA (1994) Dynamics and pattern in management coniferous forest landscapein Oregon. Ecol Appl 4:555–568

Staus N, Strittholt J, Dellasala D, Robinson R (2002) Rate and pattens of forest disturbance in the Klamath-Siskiyou ecorregión, USA, between 1972 and 1992. Landsc Ecol 17: 455–470

Steininger MK, Tucker CJ, Ersts P, Killeen TJ, Villegas Z, Hetcht SB (2001) Clearence and fragmentation oftropical deciduous forest in the Tierras Bajas, Santa Cruz, Bolivia. Conserv Biol 15:856–866

Watson RT, Zinyowera MC, Moss RH (eds) (1997) The regional impacts of climate change: an assessment ofvulnerability. Cambridge University Press, Cambridge

Watson RT, Noble IR, Bolin B, Ravindranath NH, Verardo DJ, Dokken DJ (2000) Land use, land-use change,and forestry. Cambridge University Press, Cambridge

Climatic Change