Wind effects on spatial variability in pearl millet yields in the Sahel

Growth and yield of pearl millet (Pennisetum glaucum) in the West African Sahel are characterized by significant spatial variability at short distances (5–20 m). Several studies have suggested that this variability is caused by aeolian redistribution of relatively fertile topsoil material. The objectives of this study were (i) to quantify storm-based erosion/deposition patterns within a Sahelian millet field and (ii) to determine the effects of topography, erosion and deposition on millet growth and yield. An experiment was conducted at a research station in southwest Niger, on a sandy, siliceous, isohyperthermic Psammentic Paleustalf. Twenty-one sediment catchers were installed in a 40 m×60 m plot within a pearl millet field. Four wind erosion events occurred during the 1993 growing season. Maps of wind-blown mass transport were created for each storm by applying a geostatistical space–time procedure. Maps showing the spatial distribution in erosion and deposition were derived by differentiation of mass transport in the direction of the mean wind. For 21 sub-plots of 4 m×5 m the elevation, wind-blown mass balance, millet growth and yield were quantified. A correlation analysis showed neither a significant relationship between millet yield and topography, nor between millet yield and erosion/deposition. It is concluded that only wind-blown sediment transport occurring within one growing season cannot explain the spatial variability in millet growth and yield. A new hypothesis is defined that suggests aeolian erosion/deposition patterns as observed in previous studies were not the cause but more likely the result of spatial variability in millet growth.