ASSESSING THE CURRENT EXTENT AND RECENT LOSS OF DRYLAND FOREST ECOSYSTEMS

JOSE MARIA REY BENAYAS; LUCIANA MARÍA CRISTOBAL; TOMAS KITZBERGER; ROBERT MANSON; FABIOLA LÓPEZ-BARRERA; JENNIFER SCHULZ; RAUL VACA; LUIS CAYUELA; RAUL RIVERA GARCÍA ; LUCIO MALIZIA; DUNCAN GOLICHER; CRISTIAN ECHEVERRÍA; RAFAEL DEL CASTILLO; JAVIER SALAS

Principles and Practice of Forest Landscape Restoration: Case studies from the drylands of Latin America

UICN

Año: 2011; p. 23 - 63

Land-cover change is regarded as the most important global change affecting ecological systems. Natural landscapes ? i.e. those largely unaffected or hardly affected by human activities ? are being rapidly transformed into urban and farmland landscapes throughout the world.As the characteristics of land-cover have important impacts on climate, biogeochemistry, hydrology, species diversity, and the well-being of human societies, land-cover change has been identified as a high priority for research and to inform the development of strategies for sustainable management.Dryland forests are highly prone to degradation and desertification on accountof their limited primary productivity and slow recovery following human disturbance, yet these ecosystems play a crucial role in providingservices such as climate and water regulation.We addressed these issues in selected areas in Chile, Argentina and Mexico, using standardized research protocols. The primary advantage of assessing these different areas is that they include a range of ecological, socioeconomic, and cultural characteristics. We used remote sensing data to measure and monitor land-cover change because of their ability to capture an instantaneous synoptic view of a large part of the Earth?s surface and to provide repeated measurements of the same area on a regular basis.Land-cover change detection and monitoring is especially useful in those regions where there is a lack of available cartographic information with sufficientspatial resolution to examine land-cover change. To investigate the possible causes of change in forest cover, a Geographic Information System (GIS) database incorporating satellite imagery and biophysical and socioeconomic variables was developed for each study area. This information was statistically analyzed to infer likely drivers of forest-cover change and to test a series of specific hypotheses relating to the factors responsible for deforestation. For example, we expected that the rate of forest loss would be (i) positively associated with population density and accessibility (i.e. proximity to roads and rivers), and (ii) highest on sites most suitable for agriculture such as those with gentleslopes.