Home | Archives | About | Login | Submissions | Notify | Contact | Search
 E&S Home > Vol. 13, No. 1 > Art. 36

Copyright © 2008 by the author(s). Published here under license by The Resilience Alliance.
Go to the pdf version of this article

The following is the established format for referencing this article:
Killeen, T. J., A. Guerra, M. Calzada, L. Correa, V. Calderon, L. Soria, B. Quezada, and M. K. Steininger. 2008. Total historical land-use change in eastern Bolivia: Who, where, when, and how much? Ecology and Society 13(1): 36. [online] URL: http://www.ecologyandsociety.org/vol13/iss1/art36/

Research, part of Special Feature on The influence of human demography and agriculture on natural systems in the Neotropics

Total Historical Land-Use Change in Eastern Bolivia: Who, Where, When, and How Much?

Timothy J. Killeen 1, Anna Guerra 2, Miki Calzada 2, Lisette Correa 2, Veronica Calderon 2, Liliana Soria 2, Belem Quezada 2 and Marc K. Steininger 1

1Conservation International, 2Museo Noel Kempff Mercado


We documented the history of land-use change and migration in eastern Bolivia in five temporal periods: pre-1976, 1976–1986, 1986–1991, 1991–2001, and 2001–2004. We included all land-cover types located below the natural montane tree line (~3000 m), including forest, savanna, scrubland, seasonal wetland, second-growth forest, pasture, and cropland. The rate of land-use change has increased from approximately 4.7 × 104 ha/yr in the 1960s to > 2.9 × 105 ha/yr in the most recent period. Land-use change was quantified for 10 groups within which cultural traditions and production systems are shared, including three nonmechanized indigenous groups, four mechanized farming groups, two cattle ranching groups, and the forest products sector. Mechanized Cruceño farmers and Andean indigenous colonists were responsible for most land-use change in the 1960s and 1970s; deforestation by the latter group increased to twice that by all other groups during 1986–1991, declined in the 1990s, and then increased again in the most recent period. In the last 15 years, land-use change by agro-industrialists specializing in soybean has become important, and cattle ranching based on cultivated pastures has surpassed land use by all other groups. When the rates of change increased for the three indigenous nonmechanized groups, they tended to decrease for the four nonindigenous mechanized groups, and vice versa.

Key words: deforestation; drivers; land-cover change; markets; migration; social sectors; tropics


The Amazon is inhabited by various social groups that depend on its natural resources for their livelihoods. Many groups clear forest as part of the production system. An understanding of how these groups are integrated into local, national, and regional economies is essential to develop strategies to conserve natural habitat and ensure sustainability. Research has shown that deforestation is driven by multiple forces that can be identified by an analysis of spatial and temporal patterns linked to both cultural and economic phenomena (Bürgi et al. 2004). For example, forest resources are a mainstay of the local economy of traditional communities; nonetheless, most indigenous communities clear forest to provide food staples for their families. Large-scale deforestation is linked to migratory phenomena and the creation of road networks; in some instances colonies are planned by governments and international agencies, but more frequently, spontaneous settlements are created along highway corridors (Sierra 2000, Etter et al. 2005, Killeen et al. 2005, 2007).

In the Andean countries, most settlements produce rice and tropical fruits for highland urban centers; cattle production is important in Caquetá, Colombia, and Santa Cruz, Bolivia (Killeen 2007). The cultivation of coca for illicit cocaine is a driver of deforestation in Colombia, Peru, and Bolivia (United Nations Office on Drugs and Crime 2007; Bradley and Millington 2008). Soil depletion and weed infestation lead to abandonment, and the resulting anthropogenic landscapes are dominated by secondary forest (Brookfield and Padoch 1994, Steininger 2000).

In Brazil, land-use change is caused mainly by ranches that convert native vegetation to pasture to raise cattle for national markets (Margulis 2004, Fearnside 2005). Recently, the international demand for soybean has led to extensive conversion of the cerrado (a mosaic of savanna and forest) biome in Brazil, which is a biodiversity hot spot characterized by high levels of endemism (Myers et al. 2000, Machado et al. 2007). Most of the expansion in soybean has come from the conversion of pasture, but soybean is displacing the cattle industry into the Brazilian and Bolivian Amazon (Margulis 2004, Morton et al. 2006, Killeen et al. 2007).

The Andean and Brazilian economies have been separated by the geographic barrier of the Amazon wilderness. However, investments in infrastructure will soon integrate these regions; transcontinental highways will expand the agricultural frontier, and deforestation will be stimulated by technology transfer (Killeen 2007). Eastern Bolivia is a case that can be used to understand how cultural and technological interchange will influence land-use change in Andean countries following regional integration. Bolivia is already heavily influenced by Brazilian technology and has the second highest rate of deforestation in South America (Killeen et al. 2007). In addition, it is an ethnically diverse country that has indigenous peoples and longtime inhabitants of European descent, as well as more recent immigrants from Asia, North America, and Brazil. Settlements are characterized by differences in cultural tradition, access to technology, capital resources, and the security of land tenure.

Bolivia spans two major ecological transition zones. An altitudinal gradient constrains species distributions on the eastern slope of the Andes, whereas a latitudinal gradient characterized by decreased precipitation and increased seasonality broadly defines vegetation structure in the lowlands. Rain forest predominates in localized areas on the Andean piedmont (Killeen et al. 2007). Humid Amazonian species predominate on the northern plains; these transition into the seasonally dry forests of Chiquitania and eventually, the semi-arid woodlands of Gran Chaco to the south (Killeen et al. 2006). Geomorphology influences the distribution of habitat types across the latitudinal gradient, with seasonally flooded savanna wetland predominating on the alluvial plain of the Beni, and Cerrado savannas predominating on weathered upland soils or rocky landscapes on the Brazilian Shield (Fig. 1A). Differences in climate and soil fertility influence production systems and consequently, the composition of social groups that occupy the landscape.

We report patterns of land-use change in eastern Bolivia for 10 different social groups, covering half a century beginning with the Bolivian Revolution in the 1950s and the construction of the first modern highways that linked the lowlands with the Andean highlands in the 1960s. It encompasses periods when international commodity markets stimulated the cultivation of cotton and sugar (1970s) or led to the collapse of the country’s mining industry (1980s), as well as public policies targeted at the eradication of illicit drugs and structural adjustments intended to promote private enterprise (1990s). Throughout this period, the road network has been expanded and improved, which has allowed the various social groups to expand their presence and consolidate their settlements while encroaching on lands that were previously dedicated to forest exploitation. Lowland Bolivia provides a distinctly interesting landscape with which to understand how different social groups react to the stimulus of international markets and public policies that are designed to stimulate economic growth and conserve forest (Pacheco 2006, Hecht 2005).


Image classification and GIS analysis

Detailed descriptions of the satellite images, classification methodology, editing process, and validation procedures are reported by Killeen et al. (2005, 2007). In summary, we created a map of land-cover change for eastern Bolivia below the tropical montane tree line (approximately 3000 m above sea level) covering the approximate period from 1991 to 2001 using orthorectified Landsat images (Tucker et al. 2004) and a nonsupervised classification to create approximately 125 classes that were grouped into eight strata: forest, native nonforest (i.e., cerrado savanna, Chaco woodland, savanna wetland), anthropogenic nonforest, second growth forest, water, hill-slope shadow, snow, and cloud, as well as the change among classes. Anthropogenic nonforest and second growth forest were grouped into one of four change categories, i.e., forest, cerrado, Chaco, or savanna wetland, and then further stratified into five temporal periods: pre-1976, 1976–1986, 1986–1991, 1991–2001, and 2001–2004. The stratification was based in part on a previous study covering the periods prior to 1991 (Steininger et al. 2001), and this was validated by comparison with the original imagery to correct classification and registration errors. Change from the most recent temporal period (2001–2004) was added to the map from a separate land-cover classification that was compiled from 14 Landsat scenes that contained 95% of all previous land-cover change (Fig. 1A). The precision and accuracy of the land-cover map were evaluated by comparing the map with video images that were captured during validation flights conducted in late 2004 and early 2005. The probability of an error in classification for a pixel from inspected landscapes was relatively high, at ~15%, but because overestimates of deforestation cancel out underestimates, the accuracy of our summary results is relatively robust at ~1% error (see Killeen et al. 2007).

The analysis of land-use change is based on an a priori classification of groups defined by differences in social organization, cultural history, technology, and degree of integration into the national economy (Table 1, Appendices 1 and 2). Each group has a known settlement history (Stearman 1985, Bender et al. 1987, Klein 1992, Amemiya 2004), and differences in production systems are reflected in deforestation patterns. The landscape footprint for each group was obtained by digitizing polygons at the outer edge of deforestation patches in the most recent Landsat image and includes recently cleared land, agricultural fields, and second growth forest, as well as natural habitat remnants situated among forest clearings (Appendices 1 and 2). Two groups were identified by the absence of land-use change and the presence of natural vegetation types that define their production system: the forest products sector and extensive cattle ranchers; essentially, these were defined by default by their lack of incorporation into other groups. No groups overlapped, and each landscape footprint comprised the total area occupied during the most recent temporal period. We present land-use change in absolute values, rather than as proportions, to avoid the artifact of changing baselines. The restricted-use zone includes only national protected areas that have complete protection, whereas multiple-use protected areas are included within the polygons assigned to the appropriate social group. Summary statistics were generated by overlaying the polygons with the multitemporal land-use change map in GIS. Finally, the land-use intensity was determined for six farm groups based on an evaluation of video images. The GIS polygons were used to stratify the flight transects from which 50–200 video frames were randomly selected for each group; land-use within each frame (~7 ha) was summarized according to the percent cover of forest, native nonforest (e.g., Chaco, cerrado, or wetland), second growth forest, pasture, and cropland.


Historical land-cover change

Bolivia has experienced large amounts of land-cover change over the past 50 years. Nonetheless, > 90% of the eastern part of the country remains as forest, wood land, grassland, or wetland (Fig. 1). Deforestation has been the predominant type of habitat conversion, but other biomes also have been affected. No single social or economic actor is responsible for a majority of habitat conversion (Table 2), but four categories can be recognized: nonmechanized or partially mechanized (indigenous) farming, highly mechanized (nonindigenous) farming, cattle ranching, and forest use and conservation. Nonmechanized or partially mechanized farming occupies approximately 4.6 × 106 ha and is responsible for ~31% of converted habitat. This supra-group includes the categores of lowland indigenous communities, Yungueño indigenous farmers, and indigenous Andean colonists. Highly mechanized farming occupies approximately 3.16 × 106 ha and is responsible for ~41% of converted habitat. This supra-group includes the categories of Cruceño farmers, Japanese colonists, Mennonite colonists, and agro-industrial corporations. Cattle ranching occupies 26.9 × 106 ha and is responsible for ~20% of converted habitat. This supra-group includes the categories of intensive cattle ranchers, who clear forest for pasture establishment, and extensive cattle ranchers, who use native forage, but also convert native habitat to cultivated grasses. Forest use and conservation is the predominant activity on 43.8 × 106 ha, with ~8.3% of the total converted habitat in both the restricted-use zone and the forest products sector, which includes private companies, indigenous reserves, and state lands.

On landscapes occupied by the first two supra-groups, the amount of remnant forest was an unexpectedly large 3.1 × 106 ha, representing ~7% of the total forest cover of eastern Bolivia. Approximately one-third of this was occupied by lowland indigenous communities (Fig. 2A). After removing this group, the number is still large, at 2.1 × 106 ha (4.6%). The group with the least amount of forest remnant was Japanese colonists.

Temporal variation in land-cover change

The dynamic of land-cover change in eastern Bolivia has changed over the last half-century (Figs. 2 and 3). Historically, settlement and deforestation was limited to landscapes near the city of Santa Cruz and the Yungas region of La Paz. Agricultural expansion led to the establishment of colonization zones in the Alto Beni, Chapare, and Yapacaní. Annual rates of change in the first temporal period (pre-1976) were calculated using a variable time period reflecting the approximate date when each group initiated settlement (Table 1). The three groups responsible for the most land-use change in this first temporal period were Cruceño farmers, Yungueño indigenous farmers, and Andean indigenous colonists (Fig. 3).

The total rate of change in the second temporal period (1976–1986) remained approximately equal to that in the previous temporal period, but declined for Cruceño farmers while increasing for Mennonite colonists in the mid-1970s (Bender et al. 1987). The largest increase came from Andean indigenous colonists (Fig. 3), who were settled in four separate colonization zones by 1968 (Table 3). The newest colony, San Julian, was laid out on a grid, with communities located at intersections of secondary roads; the distinctive radial deforestation pattern reflects the land titling system within this planned colonization zone (Appendices 1 and 2). Settlement in other spontaneous colonization zones occurred along primary and secondary roads, forming a fishbone pattern characteristic of colonization projects throughout the Amazon (Skole and Tucker 1993).

The third temporal period (1986–1991) showed an increase in land-use change for all groups, with Andean indigenous colonists showing the largest increase (Fig. 3). Agro-industrial corporations became established on the alluvial plain east of the Río Grande River (Appendices 1 and 2). At this point, habitat conversion for mechanized farming approximately equaled that for nonmechanized farming (Fig. 3).

The trend in habitat conversion changed radically in the fourth temporal period (1991–2001). Agro-industrial corporations expanded exponentially, whereas Andean indigenous colonist showed a sharp decline (Fig. 4). Intensive cattle ranchers experienced a 10-fold increase in the clearing of Chiquitania cerrado and Gran Chaco woodlands (Fig. 2C). Mennonite farmers increased their annual rate of land-cover change, whereas Japanese colonists, Cruceño farmers, and lowland indigenous communities all showed decreases.

The fifth and most recent temporal period (2001–2004) demonstrates the unpredictable nature of the phenomena that drive land-cover change in Bolivia. The growth in deforestation rates leveled off for agro-industrial corporations and intensive cattle ranchers, but increased for Cruceño farmers and extensive cattle ranchers. The previous decline for indigenous groups reversed, rebounding to all-time highs for lowland indigenous communities and Andean indigenous colonists (Fig. 3).

Land-use intensity

The term land-use intensity refers to the proportion of land dedicated to crop, pasture, second-growth forest, and forest: cultivation is more intensive than pasture, and pasture is more intensive than second-growth forest. The two indigenous groups included in the video survey have a similar intensity of land use following deforestation (Fig. 5); the small amount of land dedicated to crops by the Andean indigenous colonists is due in part to climatic constraints in colonization zones where the annual rainfall exceeds 3000 mm and long fallows are necessary to maintain productivity. Fallow includes pasture, as well as second-growth forest, because indigenous groups tend to raise pigs and chickens, rather than cattle. The San Julian colony has less land in fallow compared to other colonization zones (Fig. 4), reflecting the more amenable climate (e.g., mean annual rainfall of 1500–2000 mm), fertile alluvial soils of the Río Grande plain, and an incipient process of mechanization (Cochrane et al. 2007).

Mennonite colonists and agro-industrial corporations have the most intensive land use, with > 60% of cleared land dedicated to annual crops and low levels of second growth. Japanese colonists maintain approximately equal proportions of land under cultivation and pasture, a level of land use that is less intensive than that of Cruceño farmers (Fig. 5).

The moderately high levels of remnant forest vegetation documented for farming groups is deceptive; these groups are expanding their land-use footprint, and forest remnants are exterminated over time. The second-growth forest observed for agro-industrial corporations is an artifact caused by a single abandoned farm. Intensive cattle ranchers maintain slightly less than 50% of the landscape as forest, even in relatively well-established settlement regions; however, this group has very little land in second-growth fallow because most cleared land is maintained as perennial pasture. The importance of livestock for mechanized farmers is evidenced by the relatively high levels of pasture for all four groups (Fig. 5).


The scientific literature and the popular press have published thousands of reports on the dimensions, extent, and impacts of deforestation in the Amazon. Understanding the social and economic actors involved in the expansion of the agricultural frontier is essential for developing policies to slow deforestation and conserve natural ecosystems.

Indigenous farmers

Subsistence agriculture continues to play an important role in the dynamic of deforestation in Bolivia. The level of habitat conversion associated with lowland indigenous communities is small, but has expanded over time. Our results show that if income from forest resources is restricted, deforestation will increase as agricultural production replaces forest resources. The increase in deforestation in the third temporal period (1986–1991) coincided with the collapse of the rubber trade when Brazil ended rubber subsidies in the 1980s. Instead of keeping workers in the forest for 12 months each year, companies contracted seasonal labor for the Brazil nut harvest. Employees migrated from remote stations to villages, and the increase in deforestation in northern Bolivia was related to an increase in subsistence farming near communities (Fig. 6).

Andean indigenous colonists initiate their settlements as subsistence farmers, but production quickly becomes focused on national markets. Wildlife resources are exhausted and timber is exterminated over the short term, obligating settlers to adopt intensive agricultural systems; cultural links to urban centers facilitate the settlers’ access to national markets (Stearman 1985). Global markets and international policy had a profound impact on the deforestation rates of Andean indigenous farmers. Coca has been legally cultivated in Bolivia for millennia by Yungueño indigenous farmers, but once cocaine consumption increased in the 1970s, illicit cultivation expanded to the Chapare colonization zone. Deforestation dramatically increased in the mid-1980s following the collapse of international tin prices and the termination of ~2.0 × 104 miners by the national mining company (Klein 1992). The subsequent increase in migration coincided with a period of weak governance that contributed to the boom in coca cultivation. In the 1990s, Bolivian governments implemented an aggressive interdiction policy with international assistance (United Nations Office on Drugs and Crime 2007), and deforestation associated with Andean indigenous colonists decreased by almost one-third; however, the downward trend was reversed in the last temporal period as coca policy changed from eradication in the 1990s to tolerance in 2002–2005, and more recently, permissiveness post 2005 (Fig. 3).

Despite its importance, it would be inaccurate to attribute most of the deforestation from Andean indigenous colonists to the cultivation of illicit coca. The Chapare colonization zone is only one of four major colonization zones: the Yapacaní colonies specialize in citrus and rice; San Julian is a major source of maize; and the Alto Beni has a diversified agricultural system (Table 3). Slash-and-burn agriculture is the major production system in all of these regions, but established settlers invest in intensive cropping systems and acquire trucks, tractors, harvesting machines, and other implements. Evidence of mechanization can be seen in San Julian, where older settlements near the major highway show less secondary growth than do younger settlements that are situated further to the northwest (Appendices 1 and 2).

The increase in land-use change in the last temporal period for Andean indigenous colonists and lowland indigenous communities demonstrates the impact of public policy on land-cover change. The dissatisfaction of the Bolivian populace with the economic reforms of the 1990s led to a fluctuation in the political dynamic, and indigenous groups have been newly empowered in the competition for land and forest resources (Movimiento al Socialismo 2007). In northern Bolivia, the former employees of rubber and Brazil nut concessionaires are being granted titles to land that was previously managed by their former employers; these settlers are supplementing Brazil nut harvest with subsistence agriculture (Ruiz 2005). Andean indigenous colonists are occupying state lands or properties with unclear titles, as well as invading forest reserves and protected areas, whereas the Bolivian government is advocating a return to planned settlement schemes (Constituyente Soberana 2006).

The cattle sector

Lowland Bolivia has a long tradition of cattle ranching that has evolved over the last three decades as new technologies increase productivity and profitability. The shift from native forage to cultivated grasses began as a response to the scarcity of fodder during the dry season, and the British Mission in Tropical Agriculture introduced many of the most successful cultivated grasses in the 1970s (Paterson 1984). Ironically, one of the stated goals of this Mission was to reduce deforestation by increasing productivity on previously deforested lands. Increased profitability led to the expansion of intensive cattle ranching, and pasture establishment showed near-exponential growth in the 1980s and 1990s (Figs. 2B and 3) and eventually impacted traditional grazing lands (Fig. 2C).

Extensive cattle ranchers remain the largest sector within the cattle industry (Table 4), but their role within the industry has shifted to providing calves and yearlings to intensive cattle ranchers and mechanized farming groups, which maintain part of their land holdings as pasture (Fig. 5). Deforestation by intensive cattle ranchers has been influenced by the ongoing reform of land tenure. During the 1970s and 1980s, large tracts of state land were distributed to individuals who had political influence; subsequent governments attempted to recover land using provisions of the Land Reform laws of 1997 and 2007. Land titles revert to the state if the holder can not demonstrate that the property fulfills an “economic and social function” (Instituto Nacional de Reforma Agraria 2007a). Many of the lands that have questionable titles are situated on forest landscapes, and most land holders choose to present legal briefs based on the intensive cattle ranching model. The area allowed to each land holder under this model is based on a formula that includes the number of cattle and the area of cultivated pasture; consequently, pasture establishment is driven partly by land holders protecting their holdings from reversion to the state.

Mechanized agriculture

The most intensive land use in Bolivia occurs on mechanized farms on the alluvial soils of the Río Grande plain. Approximately 42% of total deforestation has occurred in this area. The four mechanized farming groups show different patterns of land use that reflect access to technology, capital, cultural traditions, and political influence.

The Cruceño farmer group includes small, medium, and large properties; individual producers specialize in sugarcane, cotton, rice, soybean, beef, dairy, poultry, or some combination of these. All are highly mechanized and consume fuel, seeds, and chemicals; import machinery; and invest in infrastructure. Some farmers used disk harrows on wet soils in the 1970s, causing subsoil compaction; degraded lands were converted to pasture and are one explanation for the high cover of pasture (Fig. 5). Land change in the last temporal period occurred near the Río Grande floodplain, an area at risk from inundation. The remaining forest landscape (~30%) is situated on the rolling hills near the Andes and is suitable only for pasture and forest management (Prefectura del Departamento y Consorcio IP-CES-KWC 1996).

Japanese colonists are medium-scale farmers who arrived in the 1950s and occupied the lands with the best soil and climatic conditions. Their close-knit social organization allowed them to obtain economies of scale by forming cooperatives. Technical support from the Japanese government has played a role in introducing new crops and technology. Japanese colonists have deforested a higher percentage of their land holdings than has any other group; their land use includes the rotation of crop land and pasture (Amemiya 2004).

Mennonites are accomplished farmers who have an extensive knowledge of cropping systems; however, their cultural tradition leads them to idiosyncratically limit the use of technology (Bender et al. 1987). In Bolivia, they eschew the ownership of cars and trucks, but will rent these services as needed; they own tractors, but use steel wheels, rather than rubber tires, even though this leads to soil compaction. They use genetically modified organisms and chemical inputs, but have been slow to adopt minimum tillage. Mennonite colonists have high combined rates of land-cover change and land-use intensity, but have settled in areas that have sandy soils and are exposed to periodic drought (Appendices 1 and 2).

The expansion of agro-industrial corporations started with a World Bank project to expand agricultural production to the eastern sector of the Río Grande plain. The Eastern Lowlands Project (World Bank 1997) included investments in silos, processing facilities, highways, and technical assistance, as well as land-use planning and biodiversity conservation (Prefectura del Departamento y Consorcio IP-CES-KWC 1996). Santa Cruz is situated across a pronounced precipitation gradient, with mean annual precipitation ranging from 3000 to 800 mm across a northwest–southeast transect of 150 km (Cochrane et al. 2007). The Eastern Lowlands Project Area was situated near the southern end of this gradient; within the first decade of the project, agro-industrial corporations experienced crop failures due to drought. Subsequent investments flowed to more humid landscapes situated west of the Río Grande and north of the lands occupied by Cruceño farmers (Appendices 1 and 2). Although this area is at less risk from drought, it has greater risk of flooding because it is situated in a delta-like landscape where three rivers, i.e., the Grande, Piraí, and Yapacaní, have meandered in recent history and an upsurge in wetland conversion in the last temporal period (1224 to 14,500 ha/yr) can be attributed to settlement on this landscape.

Periodic booms fueled by international markets have played an important role in the expansion of agro-industrial corporations because producers respond to short-term opportunities by clearing new land. The existence of the agricultural frontier has impeded investment that might lead to higher levels of productivity. Investors have a number of economic incentives to clear land on the frontier to expand production: land is inexpensive in remote areas; nutrients stored in biomass ensure good harvests in the initial years of the investment cycle; and economies of scale can be easily obtained in a single land purchase, avoiding the need to consolidate small holdings.

Policy incentives could promote the use of technology to increase crop yields in older fields as an alternative to habitat conversion. Santa Cruz is situated over a large aquifer that could be exploited indefinitely because it is recharged by precipitation and high-volume Andean rivers (Cochrane et al. 2007). However, irrigation requires a capital investment that has a less attractive rate of return compared to that of buying land, clearing forest, and expanding production. Similarly, chemical fertilizers are not used in Bolivia because of the high cost; there are no fertilizer factories, despite the country’s abundant natural gas reserves, which are the primary feedstock for nitrogenous fertilizers. Economic models show that if irrigation and fertilizers were used at optimum levels, crop yields could be doubled while eliminating climate risk (Cochrane et al. 2007).

Forest management and conservation

The forest products sector showed an idiosyncratic pattern of deforestation over the five temporal periods. Initial rates of deforestation were low, but increased with time, with peaks in the third and fifth temporal periods (Fig. 3). These moderately large deforestation rates are the consequence of the large total area and numerous scattered patches of deforestation that sum to a large number in absolute terms, but are small in relative terms (Table 2). The forest products sector has a long history of sustainable exploitation that is rooted in the two rubber booms of the late 19th and 20th centuries and the ongoing exploitation of Brazil nuts. However, it also has a history of short-term exploitation and the exhaustion of renewable resources, as exemplified by the cases of palm hearts (Peña-Claros and Zuidema 2000) and mahogany (Kometter et al. 2004).

In the 1970s and 1980s, the state granted logging contracts for 40-yr periods to companies that exploited mahogany; timber royalties were based on the volume of wood harvested. This model collapsed in the late 1990s with the commercial extinction of mahogany and was replaced by a system based on multiple species exploitation, reduced-impact logging, and natural forest regeneration (Putz et al. 2004). Under the new scheme, concessionaires are required to implement management plans with rotational harvests spanning 30 yr, and royalties are based on the area of the concession. Companies had to rationalize the demand for timber with their ability to pay a tax based on spatial area; consequently, large blocks of forest were returned to the state. Under the new model, concessions were made available to communities via cooperatives (Asociaciones Sociales del Lugar), and lowland indigenous communities were granted reserves (Tierras Comunitarias de Orígen, TCOs; Instituto Nacional de Reforma Agraria 2007b).

These groups currently occupy, manage, or have claims to approximately 50% of the area defined as the forest products sector (Fig. 7) and are competing for the remainder of the forest, which is nominally held by the state. In the northern part of the country, companies that dominated the Brazil nut industry for decades are defending their concessions, arguing that they have managed the forest responsibly, but now face competition from former employees who are demanding titles to forest lands (Ruiz 2005). The forest landscape in the rest of Bolivia eventually will be divided among companies, cooperatives, and indigenous communities or will be occupied by one of the farming or ranching groups.

Forest vegetation is part of the landscape in the settlement areas of all groups; in the case of Mennonite and Japanese colonists and agro-industrial corporations, the persistence of forest is a temporal phenomenon and is likely to be lost over time. Even forest remnants that are left as windbreaks on industrial farms will be degraded due because of the well-known edge effect (Laurance et al. 2002). In contrast, the forest fragments that are associated with intensive and extensive cattle ranchers are more likely to persist over the long term. Ranchers tend to consider forest remnants as capital assets and view their properties as dual-purpose production estates.

Indigenous groups maintain a large portion of the landscape as either forest or secondary forest fallow (Fig. 4). Lowland indigenous communities are usually located near their own forest reserves (TCOs) and their farming activities complement forest resources. In contrast, Andean indigenous colonists have a much greater impact on the forest, and the fate of native forest remnants within their landscape footprint is similar to that of mechanized groups, although the abundance of secondary forest makes these landscapes more amenable for biodiversity conservation.

The absolute low rate of change reported for restricted zones is deceptive because these polygons exclude parts of protected areas that are occupied by Andean indigenous colonists and are included in the statistics for that group. A separate analysis that evaluated protected areas showed that deforestation increased within protected areas in the last temporal period (Killeen et al. 2007).


Land-cover change in eastern lowland Bolivia is caused by the implementation of distinct economic production models by various social groups. Differences among the groups are related to their integration into the national economy, degree of mechanization, and access to credit and markets. Cultural differences influence how they establish settlements, whereas their production models determine how quickly and completely the natural habitat is converted to cropland, pasture, or second-growth fallow.

The rate of deforestation has increased over time, and the relative contribution of the various social groups to this deforestation has varied according to the market and public policy. Mechanized (i.e., nonindigenous) groups tended to expand their activities when the rates of change associated with nonmechanized (i.e., indigenous) groups were in decline. All groups have been affected by market phenomena, including remote communities that rely on forest resources and subsistence agriculture. National urban and international commodity markets are the principal drivers of land-use change for both mechanized farmers and slash-and-burn agriculturalists. Markets stimulate deforestation by increased demand for commodities such as sugar, cotton, and soybean or by causing peasants to migrate when alternative production systems such as rubber and tin decline. Policy initiatives that have accelerated habitat conversion by the various groups include colonization schemes, visa permits for foreigners, infrastructure investment, and land tenure review, as well as the implementation and suspension of programs to eradicate illicit drug crops. An understanding of how various social groups react to market phenomena and policy initiatives is essential to devise effective policies. The Bolivian experience offers important lessons for devising the next round of policies aimed at reducing deforestation.


Responses to this article are invited. If accepted for publication, your response will be hyperlinked to the article. To submit a response, follow this link. To read responses already accepted, follow this link.


Financial support for this study was provided to the Friends of the Noel Kempff Mercado Museum Foundation by the Critical Ecosystem Protection Fund, The Nature Conservancy, World Wildlife Fund, and the Vice Ministry for the Environment of the Bolivian government, as well as the Gordon and Betty Moore Foundation via its commitment to Conservation International. The Symposium on Human Demographics and Agriculture on Natural Systems in the Neotropics at the Association of Tropical Biology and Conservation 2007 in Morelia, Mexico, was supported by the national Science Foundation (NSF support #DEB-0731594).


Amemiya, K. 2004. Celebrating Okinawans in Bolivia. Japan Policy Research Institute Critique 11(4). [online] URL: http://www.jpri.org/publications/critiques/critique_XI_4.html.

Bender, H. S., M. W. Friesen, M. Ediger, I. Hiebert, and M. Gerald. 1987. Bolivia. Global Anabaptist Mennonite Encyclopedia Online. [online] URL: http://www.gameo.org/encyclopedia/contents/B665.html.

Bradley, A. V., and A. C. Millington. 2008. Coca and colonists: quantifying and explaining forest clearance under coca and anti-narcotics policy regimes. Ecology and Society 13(1): 31. [online] URL: http://www.ecologyandsociety.org/vol13/iss1/art31/.

Brookfield, H., and C. Padoch. 1994. Appreciating agrodiversity: a look at the dynamism and diversity of indigenous farming practices. Environment 36(5):8-43.

Bürgi, M., A. M. Hersperger, and N. Schneeberger. 2004. Driving forces of landscape change—current and new directions. Landscape Ecology 19(8):857-868.

Cochrane, T. A., T. J. Killeen, and O. Rosales. 2007. Agua, gas y agroindustria: gestión sostenible de agua para riego agrícola en Santa Cruz, Bolivia. Conservation International Bolivia, La Paz, Bolivia.

Constituyente Soberana. 2006. Conflicto en Yapacaní: colonos toman 500 hectáreas de ganaderos. Constituyente Soberana. [online] URL: http://constituyentesoberana.org/info/?q=yapacani-colonos-inra.

Etter, A., C. McAlpine, D. Pullar, and H. Possingham. 2005. Modeling the age of tropical moist forest fragments in heavily-cleared lowland landscapes of Colombia. Forest Ecology and Management 208(1-3):249-260.

Fearnside, P. M. 2005. Deforestation in Brazilian Amazonia: history, rates, and consequences. Conservation Biology 19(3):680-688.

Hecht, S. B. 2005. Soybeans, development and conservation on the Amazon frontier. Development and Change 36(2):375-404.

Instituto Nacional de Reforma Agraria. 2007a. Guía para la verificación de la función social y de de la función económico social. Instituto Nacional de Reforma Agraria, Ministerio de Desarrollo Agropecuaro y Medio Ambiente, Gobierno de Bolivia, La Paz, Bolivia.

Instituto Nacional de Reforma Agraria. 2007b. Saneamiento integrado al catastro legal, tierras comunitarias de origen (TCOs). Instituto Nacional de Reforma Agraria, Ministerio de Desarrollo Agropecuaro y Medio Ambiente, Gobierno de Bolivia, La Paz, Bolivia.

Killeen, T. J. 2007. A perfect storm in the Amazon wilderness, development and conservation in the context of the Initiative for Integration of the Regional Infrastructure of South America (IIRSA). Advances in Applied Biodiversity Science 7:4-99. Available online at: http://www.bioone.org/perlserv/?request=get-document&doi=10.1896%2F978-1-934151-07-5.4.

Killeen, T. J., V. Calderon, L. Soria, B. Quezada, M. K. Steininger, G. Harper, L. A. Solórzano, and C. J. Tucker. 2007. Thirty years of land-cover change in Bolivia. Ambio 36(7):600-606.

Killeen, T. J., T. M. Siles, L. Soria, and L. Correa. 2005. La estratificación de vegetación y el cambio de uso de suelo en Las Yungas y El Alto Beni de La Paz. Ecología en Bolivia 40(3):32-69. Available online at: http://editorenjefe.ecologiabolivia.googlepages.com/indice40(3).

Klein, H. S. 1992. Bolivia, the evolution of a multi-ethnic society. Second edition. Oxford University Press, New York, New York, USA.

Kometter, R. F., M. Martinez, A. G. Blundell, R. E. Gullison, M. K. Steininger, and R. E. Rice. 2004. Impacts of unsustainable mahogany logging in Bolivia and Peru. Ecology and Society 9(1): 12. [online] URL: http://www.ecologyandsociety.org/vol9/iss1/art12/.

Laurance, W. F., T. E. Lovejoy, H. L. Vasconcelos, E. M. Bruna, R. K. Didham, P. C. Stouffer, C. Gascon, R. O. Bierregaard, S. G. Laurance, and E. Sampaio. 2002. Ecosystem decay of Amazonian forest fragments: a 22-year investigation. Conservation Biology 16(3):605-618.

Machado, R. B., M. B. R. Neto, J. M. C. Silva, and R. B. Cavalcanti. 2007. Cerrado deforestation and its effects on biodiversity conservation. Advances in Applied Biodiversity Science 8 in press.

Margulis, S. 2004. Causes of deforestation of the Brazilian Amazon. World Bank, Washington, D.C., USA.

Morton, D. C., R. S. DeFries, Y. E. Shimabukuro, L. O. Anderson, E. Arai, F. del Bon Espirito-Santo, R. Freitas, and J. Morisette. 2006. Cropland expansion changes deforestation dynamics in the southern Brazilian Amazon. Proceedings of the National Academy of Sciences 103(39):14637-14641.

Movimiento al Socialismo. 2007. Programa de gobierno MAS-IPSP. Bolivia digna, soberana y productiva para vivir bien. Título I: diagnóstico. Movimiento al Socialismo, La Paz, Bolivia. Available online at: http://www.masbolivia.org/mas/programa/programa05.htm.

Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403(6772):853-858.

Pacheco, P. 2006. Agricultural expansion and deforestation in lowland Bolivia: the import substitution versus the structural adjustment model. Land Use Policy 23(3):205-225.

Paterson, R. T. 1984. Investigación y desarrollo de pastos tropicales, Santa Cruz: informe final de la asignación técnica de la MBAT en el Programa de Pasturas del CIAT 1976–1984. Missión Británica en Agricultura Tropical, Santa Cruz, Bolivia.

Peña-Claros, M., and P. Zuidema. 2000. Limitaciones demográficas para el aprovechamiento sostenible de Euterpe precatoria para producción de palmito: resultados de dos estudios en Bolivia. Ecología en Bolivia 33:3-21.

Prefectura del Departamento y Consorcio IP-CES-KWC. 1996. Memoria del PLUS. Plan de uso del suelo del Departamento de Santa Cruz, Bolivia. Prefectura del Departamento y Consorcio IP-CES-KWC, Santa Cruz, Bolivia.

Putz, F. E., M. A. Pinard, T. S. Fredericksen, and M. Peña-Claros. 2004. Forest science and the BOLFOR experience: lessons learned about natural forest management in Bolivia. Pages 64-96 in D. J. Zarin, J. R. R. Alavalapati, F. E. Putz, and M. Schmink, editors. Working forests in the Neotropics: conservation through sustainable management? Columbia University Press, New York, New York, USA.

Ruiz, S. A. 2005. Rentismo, conflictos y bosques en el norte amazónico boliviano. Center for International Forestry Research, Bogor, Indonesia.

Sierra, R. 2000. Dynamics and patterns of deforestation in the western Amazon: the Napo deforestation front, 1986–1996. Applied Geography 20(1):1-16.

Skole, D. L., and C. J. Tucker. 1993. Tropical deforestation and habitat fragmentation in the Amazon: satellite data from 1978 to 1988. Science 260:1905-1910.

Stearman, A. M. 1985. Camba and Kolla: migration and development in Santa Cruz, Bolivia. University of Central Florida Press, Orlando, Florida, USA.

Steininger, M. K. 2000. Satellite estimation of tropical secondary forest above-ground biomass: data from Brazil and Bolivia. International Journal of Remote Sensing 21(6-7):1139-1157.

Steininger, M. K., C. J. Tucker, J. R. G. Townshend, T. J. Killeen, A. Desch, V. Bell, and P. Ersts. 2001. Tropical deforestation in the Bolivian Amazon. Environmental Conservation 28(2):127-234.

Tucker, C. J., D. M. Grant, and J. D. Dykstra. 2004. NASA’s global orthorectified Landsat data set. Photogrammetric Engineering and Remote Sensing 70(3):313-322.

United Nations Office on Drugs and Crime. 2007. Coca cultivation in the Andean region for 2006: a survey of Bolivia, Colombia, Ecuador and Peru. United Nations Office on Drugs and Crime, Vienna, Austria. Available online at: http://www.unodc.org/unodc/en/publications.html.

World Bank. 1997. Eastern lowlands natural resource management and agricultural production project. Project ID: P006152. World Bank Project database. [online] URL: http://web.worldbank.org/external/projects/main?pagePK=64283627&piPK=64290415&theSitePK=40941&menuPK=228424&Projectid=P006152.

Address of Correspondent:
Timothy J. Killeen
Center for Applied Biodiversity Science
Conservation International
2011 Crystal Drive
Suite 500
Arlington, VA 22202 USA


Home | Archives | About | Login | Submissions | Notify | Contact | Search