Explaining how and why landscapes are shaped through human--environment interactions is a long-standing tradition in geography (Turner 1997, 2002). Researchers have recently sought to understand how changes in land use and land cover interact with numerous environmental and social issues, such as biodiversity; environmental hazards, global climate change, and the economic well-being of social groups that rely on an area's resource base (Turner and others 1995; Klepeis and Turner 2001). The Land-Use and Land-Cover Change Project, created under the auspices of the International Human Dimensions Programme on Global Environmental Change and the International Geosphere-Biosphere Programme, calls for documenting the magnitude and pace of land-use and land-cover change in critical regions throughout the world (Turner and others 1995). Research on land-use and land-cover change has concentrated on such hot spots of biodiversity as the Yucatdn (Kiepeis and Turner 2ool; Turner and others 2001) and Amazonia (Moran and B rondizio 1998), but little research has focused explicitly on land-use and land-cover change in tropical Andean forests, even though the tropical Andes are a biodiversity hotspot and home to more endemic plants and vertebrates than any other hotspot in the world (Myers and others 2ooo; see also Henderson, Churchill, and Luteyn 1991; Jorgensen and Valencia 1992; Jorgensen and Ulloa 1994; Madsen and Ollgaard 1994; Gentry 1995).
Andean environments have been modified by humans for at least 7,000 years (Bruhns 1994), but the magnitude of land-use and land-cover change has grown considerably in the past fifty years (Ellenberg 1979; White and Maldonado 1991; Luteyn 1992; Churchill and others 1995). In Peru and Ecuador, most of the montane forest has been cleared, and much of what remains is in protected areas and parks (White and Maldonado 1991; Young and Leon 1995). With the notable exceptions of Philip Keating (1995, 1997) and Fernando Echavarria (1993, 1998), few researchers have examined land-use and land-cover change in Ecuador's remaining native montane forests. Our understanding of the magnitude, pace, and causes of land-use and land-cover change in this critical ecological region is left wanting.
In this article we quantify land-use and land-cover change from 1987 to 1998 in two watersheds on the eastern cordillera of Ecuador (Figures 1 and 2). The Mazar and Llavircay watersheds in eastern Canar Province contain contiguous old-growth Andean forests and are experiencing deforestation pressures from the region's small-holder population. (1) By combining remotely sensed satellite images from Landsat Thematic Mapper (TM) and SPOT with fieldwork, we show that land-use and land-cover change in tropical montane forests is more dynamic than previously reported. Deforestation, driven largely by demand for cattle products, persisted during the 1990s; but, contrary to other reports of land-use and land-cover change, we document that forest regeneration and even afforestation of the paramo (grassland) is occurring as well. We discuss these two main findings in terms of trends inland-use and land-cover change in the Andes and speculate about how international migration from the region to the United States (and Spa in) will influence the livelihoods of Andean farmers and the land-use and land-cover change of the area.
THE TROPICAL MONTANE FORESTS OF ECUADOR
The Ecuadorian sierra consists of two parallel cordilleras that trend north-south and are joined in several places by transverse ranges. The cordilleras include more than twenty peaks that exceed 4,500 meters above sea level. Fausto Sarmiento divides Andean forests into two domains: Transandean forests, which occupy the outer slopes of both the western cordilera, facing the Andean piedmont of the coastal region, and the eastern cordillera, facing the Amazon Basin; and Interandean (intermontane) forests, which occupy small watersheds between the cordilleras (1995, 640). …