Mapping poverty from space in rural Assam, India

Abstract:

This paper investigates the relationships between welfare and geographical factors derived from remotely sensed satellite data within Assam, India. The pressure that natural resources experience from population growth is a significant barrier to sustainable human development and ecological conservation. Integrating social and geographic data offers the potential to increase our understanding of population-environment relationships. We construct a village welfare index for an extensive area of Assam in Northeast India. Classification and regression tree techniques were used to model the relationships between welfare and geographic conditions derived from remotely sensed data. Geographic metrics accounted for 61% of the variation in the lowest welfare quintile and 57% in the highest welfare quintile. Travel time to market towns, percentage of a village covered with woodland and winter crop were significantly related to welfare. These results support findings in the literature across a range of different developing countries which have used socioeconomic and geographic data derived only from household surveys. Model accuracy is unprecedented considering that the majority of information for the prediction is derived from remotely sensed data. As satellite data can provide continually updated geographic metrics, the results indicate the potential for substantially increasing our understanding of poverty-environment relationships by coupling remotely sensed and socioeconomic datasets. Further studies should be conducted using time series analysis as knowledge of population-environment inter-linkages will be required to help foster more effective policies for sustainable human development and ecological conservation.