Linking Socioeconomic and Remote Sensing Data at the Community or at the Household Level

Abstract

Listen

Translate article

One of the major challenges of linking remote sensing and socioeconomic data concerns the definition of the appropriate spatial observation units. This chapter discusses whether it is always beneficial and desirable to link socioeconomic data at the level of households rather than at the level of villages or communities. Some of the pros and cons of linking socioeconomic data at the level of households and at the level of the villages or communities are presented. This is illustrated by two case studies where socioeconomic data were linked with remote sensing data: in southern Cameroon and in the buffer zone of the Masai Mara Natural Reserve in Kenya. In the first case, macroeconomic changes affecting Cameroon have played a fundamental role in the way land-use practices influence the forest cover. In the second case, conversion of rangelands has led to a major decline in nonmigratory wildlife due to the spread of mechanized agriculture on critical spatial locations. This was driven by changes in markets and national land-tenure policies. Political economic theory was used as a framework for both studies. The two studies demonstrated the complementarity of remote sensing and socioeconomic survey data for understanding the causes, processes, and impacts of land-use/land-cover changes. Even though socioeconomic data were always collected at the level of households, the two studies have linked remote sensing with household survey data at the level of villages or groups of households following a similar land-use strategy rather than at the household level. Reasons for this include logistical constraints in the organization of the field survey and land-tenure system of pastoralist societies. The optimal level of analysis of a joint remote sensing-socioeconomic study depends on the research question and involves a trade-off between the information that can be extracted with a reasonable level of accuracy and the cost of field data collection. Working at the finest possible level is not always the optimal strategy. However, aggregated- level analyses obscure the role of the heterogeneity between actors (e.g., social networks, lcadership status, role of education, and differing access to resources, knowledge, and land) in driving land-use changes.