Journal Article

Land Cover Change in Mixed Agroforestry: Shade Coffee in El Salvador

Feb 1, 2012 | Allen Blackman, Beatriz Ávalos-Sartorio, Jeffrey Chow


Little is known about land cover change in mixed agroforestry systems, which often supply valuable ecological services. We use a spatial regression model to analyze clearing in El Salvador’s shadecoffee–growing regions during the 1990s. Our findings buttress previous research suggesting the relationship between proximity to cities and clearing in mixed agroforestry systems is the opposite of that in natural forests. But this result, and several others, depends critically on the characteristics of the growing area, particularly the dominant cleared land use. These findings imply that policies aimed at retaining mixed agroforestry need to be carefully targeted and tailored.

In many tropical countries, mixed agroforestry systems—in which crops such as coffee, cocoa, and bananas are planted alongside trees—are important sources of ecological benefits. For example, in El Salvador, the most densely populated and heavily deforested country in Latin America, shade coffee plays a critical role in preventing soil erosion, facilitating aquifer recharge, and harboring biodiversity. Effective policy design for conserving these benefits requires understanding the drivers of land use change in mixed agroforestry systems. While clearing of natural forests tends to occur close to urban centers, mixed agroforestry becomes more profitable the closer it is to markets. RFF Senior Fellow Allen Blackman, with colleagues Beatriz Ávalos-Sartorio and Jeffrey Chow, recently completed a detailed spatial econometric analysis of land use changes in El Salvador’s shade coffee growing regions. Their results indicate that clearing in shade coffee areas tends to occur farther from markets—the opposite of patterns generally observed in natural forests. However, this result is heavily influenced by the characteristics of the growing area, in particular, the dominant use of cleared land. For example, plots farther from export markets were more likely to be cleared only if they were in primarily rural areas instead of in the urban periphery. The lesson for policy, the authors conclude, is that conservation efforts should be targeted and tailored, taking into account the institutional and socioeconomic characteristics of different regions. One-size-fits-all policies and programs aimed at expanding or retaining agroforestry systems are likely to be ineffective and inefficient when applied on broad geographical scales. Beatriz Ávalos-Sartorio is the director of academic development at the Polytechnic University of Morelos. Jeffrey Chow is a doctoral candidate at the Yale School of Forestry and Environmental Studies

Figure One: Major Coffee Growing Areas in 1993

Figure two: Clearing Between 1990 and 2000 in Major Coffee Growing Areas