The primary cause of deforestation worldwide is agricultural expansion, mostly by small-farm families migrating to forest frontiers (Houghton 1994; Myers 1994; Geist and Lambdin 2001). Yet in studying forest clearing and land use, scholars of human-induced environmental change have focused almost exclusively on land use and degradation on the frontier, without considering why settler families end up there in the first place (Carr, 2004). At the same time, virtually all research on migration in the developing world has focused on rural-urban migration and international migration, which are only peripherally related to deforestation. The objective of the project is to investigate the determinants of the rural out-migration that underlies the deforestation and land use/land cover change (LUCC) in Guatemala's Peten, particularly in the Maya Biosphere Reserve (MBR). While household decision-making is key to examining migration, households do not make these decisions in a vacuum; local, regional, and national contexts matter. The PIs therefore propose a research plan that involves multiple data sources and methods. Multiple scale data include proposed: 1) community and municipio-level data; 2) surveys of several thousand households in origin communities; and 3) Municipio-level data from Geographic Information System (GIS) data layers derived from remotely sensed Landsat TM images and from the Guatemalan population and agricultural censuses of 2003. Research questions will be examined through the novel application of a four-level hierarchical statistical model. The model will be used to determine the relative contributions of individual, household, community, and municipio factors to migration, and to illustrate the importance of spatial scale. The PI has enjoyed an ongoing collaboration with scholars at the National Statistics Institute and other major research institutes in Guatemala and will have full access to these data sets
The proposed research makes novel contributions to 1) human dimensions of global environmental change research and 2) the integration of multi-scale spatial and survey quantitative and qualitative methods. The intellectual merit of this project is the combination of a novel conceptual and methodological contribution to human-environment and spatial social science research. The project will enhance understanding of linkages between rural-frontier migration and tropical deforestation, the most salient form land cover change in the world. The PIs believe it is the first to link deforestation on the agricultural frontier to a major underlying cause, out-migration from areas of origin. It is also pioneering in using both survey and spatial data in areas of migrant origin, and in examining the determinants of rural-rural migration using a 4-level hierarchical random effects model. The research also makes broader impacts by fostering international collaborations - including those from underrepresented groups - that will strengthen pedagogy on demography in Guatemala, produce quality scholarly publications, and influence polity policy interventions for ameliorating rural poverty and conserving Mesoamerica's tropical forests.
