Costa Rica: Tropical Dry Forest Restoration in the Guanacaste Conservation Area


The Guanacaste Conservation Area (ACG) in northwestern Costa Rica is the site of the largest forest restoration project in the tropics. The project is aimed at restoring a major tropical dry forest ecosystem that has been severely degraded as a result of anthropogenic fires associated with farming and ranching activities. These fires damage indigenous tree species that evolved in an ecosystem devoid of natural fire, and they also enable the invasive African grass jaragua (Hyparrhenia rufa) to aggressively colonize burned areas and effectively out-compete native species. Since its inception in 1986, the restoration program has sought to eliminate all fires within ACG boundaries and to facilitate the natural reforestation of denuded areas via seedling recruitment from adjacent stands of pristine forest. Although the process of recovery has been slow, former pasturelands are gradually being re-colonized by native species, and the aggressive measures taken to eliminate fires have proven successful. In 1999, the ACG was inscribed as a UNESCO World Heritage site representing the best dry forest habitats from Central America to northern Mexico.

Quick Facts

Project Location:
Guanacaste Conservation Area, 10.8393534, -85.618271

Geographic Region:
Latin America

Country or Territory:
Costa Rica

Tropical Forest

Tropical Forest - Seasonal Broadleaf

Area being restored:
70,000 + hectares

Organization Type:
University / Academic Institution


Project Stage:

Start Date:

End Date:

Primary Causes of Degradation

Agriculture & Livestock, Deforestation, Invasive Species (native or non-native pests, pathogens or plants)

Degradation Description

Five centuries ago, there were more than 200,000 square miles of tropical dry forest on the Central American Pacific coastal lowlands from Panama to Central Mexico. Today, only 0.1% of the original dry forest remaining has conservation status, making it the most threatened of all the major lowland tropical forest habitats (Janzen 1988b). The endangered condition of Mesoamerican tropical forests is primarily the result of centuries of logging, farming, and ranching (Tenenbaum 1994), although climate change, pesticide runoff from adjacent agricultural lands, and recreational user pressure have also been contributing factors.

The use of fire by local farmers and ranchers to clear new lands for agriculture and manage existing pasturelands has been especially harmful to the forest ecosystem. Because there is no lightning during the dry season in Guanacaste province, and thus no natural fires, indigenous trees lack the thick, protective bark common in other ecosystems and are vulnerable to damage from these anthropogenic fires. Moreover, native species cannot effectively compete against the invasive grass jaragua (Hyparrhenia rufa) that aggressively colonizes burned areas, and its pervasiveness results in more fuel for the next year’s fires.

Reference Ecosystem Description

The ACG is home to approximately 230,000 species (65% of the estimated number of species in Costa Rica and 2.4% of the world’s terrestrial biodiversity). By 1995, this figure was believed to include 960 species of vertebrates (including 253 species of birds and 155 species of mammals, at least half of which are bats), more than 7,000 species of plants, and more than 8,000 species of butterflies and moths, all packed into an area the size of New York City and its suburbs.

Project Goals

The project’s two primary goals are to set in place the full-scale restoration of an entire dry tropical forest ecosystem and to ensure that wildland biodiversity of the Guanacaste Conservation Area survives into the indefinite future by involving local community members and integrating the project area into local, national and international society through sustainable use practices.


The project does not have a monitoring plan.

Description of Project Activities:
Restoration activities at the Guanacaste Conservation Area have focused on stopping all anthropogenic fires and all harvesting of plants and animals; purchasing adjacent farm and ranchland to create one large, solid unit; and facilitating natural re-colonization in degraded areas. The fire control plan aims to prevent the expansion of the invasive African jaragua grass and to allow "tree invasion" into pasturelands via seeds dispersed by animals and the wind (Whelan 1987). Project planners have hired a small staff dedicated solely to fire elimination and have allocated funds for vehicles and other tools that allow for the prompt extinguishment of any new fires. Park staff clears fire breaks, maintains fire-access roads, and mans an observation station around-the-clock throughout the six-month fire season. In the early years of the program, cattle were also used as biotic mowing machines to thin the jaragua, and thereby reduce the amount of available fuel. They are no longer used, however, as the other prevention strategies have proven effective. With adequate fire protection to control the jaragua, a forest can move several hundred meters into an old pasture within ten years, simply as a result of seed dispersal (Whelan 1987). The first seeds brought into an abandoned pasture are generally wind-dispersed seeds from the forest's edge. Of the 215 tree species at Santa Rosa National Park, 25% have wind-dispersed seeds and are effective pioneer species. In an area of Santa Rosa last affected by fire in 1979, twelve species of trees constituted at least 90% of the vegetative cover in 1986. The wind-dispersed species in the plot were representatives of the families: Leguminosae, Boraginaceae, Verbenaceae, Tilaceae, Meliaceae, Bignoniaceae, Cochlospermaceae, and Hippocrateaceae (Janzen 1988a). Most of the wind-dispersed initial colonizers are large trees (attaining heights of 15-25 m) and live for 50 to several hundred years (Janzen 1988a). Along with trees encroaching into pastureland from the forest's edge, an isolated tree may appear in the middle of a pasture from dung dropped by a cow or horse while grazing the field (Whelan 1987). In cases such as this, the solitary tree, called a nuclear tree, acts as an island that begins to grow into the surrounding pasture. When the tree gets large enough to provide shade for domestic grazing animals and habitat for birds, it attracts seed dispersing animals that bring the second wave of seeds. The forest island that springs up around this tree may eventually reclaim the pastureland and join the native forest (Whelan 1987). In the early years of the program, some experimental tree planting occurred, but was quickly abandoned as a waste of energy and scarce resources. However, in the moister eastern end of the ACG, it has become clear that natural rainforest invasion is extremely slow. Here, it has been shown that a single unmanaged 4-8 year rotation of a gmelina plantation is extremely effective in eliminating many kinds of deep pasture grasses and allowing a young natural rainforest to colonise the site (e.g. While this is expensive (about $300/ha) and should not be used where there is still intact forest to be purchased for inclusion in the larger conservation area, it is a very fast form of pasture elimination in sites where forest closure is important, such as narrowing the gap between two forest blocks. Besides focusing on fire suppression and vegetative succession, program planners have also purchased tracts of land adjacent to the park in order to augment the extension of protected habitat. When it became clear in the late 1990's that, in order to save the Guanacaste dry forest, a large area of intact adjacent rainforest was necessary, a fund-raising campaign was initiated to expand 5,500+ hectares to the east. This initiative took the ACG up to the developed agroscape and helped ensure the rescue of all remaining adjacent rainforest simply by purchasing it ( There are perhaps another 10,000 hectares of forest remnants remaining at other points on the park's boundaries, and it is hoped that these can also be obtained.

Ecological Outcomes Achieved

Eliminate existing threats to the ecosystem:
All fires have been eliminated in the ACG, and approximately 70,000 ha of ancient fields and pastures are now covered with young regenerating dry forest. Areas that have been protected from fire for ten years now have ten- to fifteen-foot-tall trees living where there was once just pasture. Within 200 years, the entire dry forest ecosystem and its adjacent wet areas will appear to be fully recovered. Within one half of a millennium the area should be indistinguishable from the original forest, except for those changes that are inevitable through climate change and the insular nature of the conservation area within its surrounding agroscape.

Factors limiting recovery of the ecosystem:
In a study by Gerhardt (1998), the major cause of seedling mortality in a tropical dry forest was defoliation during dry-season drought.

Socio-Economic & Community Outcomes Achieved

Key Lessons Learned

Restoration of a large tropical wildland by eliminating destructive agents and allowing natural re-invasion of the biota is not only the least expensive possible method, it allows both funds and human resources to be directed at the more complex, challenging and critical process of biodiversity development on the site. The ACG is one of several viable protocols for the conservation of large wildland areas in tropical countries (and in some circumstances, extra-tropical ones as well). The biological and sociological processes that have worked well in the ACG contain many elements that could be applied successfully in other contexts, although they would have to be tailored to the particular circumstances affecting restoration at a given site. Building the project into the framework of the local and national community, and concentrating on ecological and economic sustainability, is essential.

Long-Term Management

The ACG employs only Costa Ricans and promotes from within the ACG staff in order to foster a sense of ownership and responsibility for the project among the employees and neighbouring communities. Most personnel development is through on-the-job training. The strategy of the ACG has been to fully involve the resident community in the forest’s development and products, and to make the ACG a local source of revenue, pride and intellectual development.

Other Resources

Guanacaste Dry Forest Conservation Fund – GDFCF
Daniel H. Janzen,
Department of Biology,
415 South University Avenue
University of Pennsylvania
Philadelphia, PA 19104 USA

Área de Conservación Guanacaste

Primary Contact

Organizational Contact