Hydro-climatic variability and agricultural production on the shores of Lake Chad
The purpose of this study is to present and analyze previously unpublished quantitative agricultural data for the area on the shores of Lake Chad in Chad, and explore its relations to hydro-climatic factors (lake levels, rainfall and temperature). This is a rural area with livelihoods based on agropastoral and fishing activities, which are directly dependent on the region’s high-varying hydro-climate. By using regression analysis on data from 1988-2012 this study was able to establish correlations between the latter and agricultural output. These correlations were used to build multivariate models to explore the predictive capacities of hydro-climatic factors with regards to the agricultural data. The selected models were able to account for considerable proportions of the agricultural dynamics. Some 5 of the 10 multivariate models tested had cross-validated R2s of 0.50 or more. Thus, there were still noteworthy unexplained variations in the agricultural data, which likely stem from technological, behavioral, economic and pest factors that were not explored in this study due do data limitations. Additional studies are called for to build on results presented here and further examine these relationships.
Erik Nilsson, Stefan Hochrainer-Stigler, Junko Mochizuki, Cintia Bertacchi Uvo
International Institute of Applied System Analysis