Bare Fallowing on Sandy Fields of Niger, West Africa

Abstract An experiment was conducted in three rainfall zones of Niger, West Africa, to evaluate the potential of mulched and nonmulched bare fallow to conserve water in deep, sandy soils. Water content of adjacent bare and cropped experimental plots was monitored with a neutron probe at three sites during the 1985 growing season and the subsequent dry season. Experimental sites, with soil type and 1985 rainfall, were near the villages of Chikal (Ustic Torripsamment; 233 mm) N'Dounga (Psammentic Paleustalf; 344 mm), and Kala Paté (Alfic Ustipsamment; 428 mm). Cropped plots were managed using traditional low‐input practices. Bare plots were weeded using a traditional, shallow cultivating hoe. At the end of the growing season, portions of the bare and cropped plots were mulched with millet stalks at a rate of 25 Mg ha −1 . At the end of the dry season, no significant increases in profile water storage were detected in mulched treatments. Increased soil water storage in the upper 1.65 m of bare fallowed plots toward the end of the dry season was 16.5 ± 10.5 mm at Chikal, and 40.6 ± 17.9 mm at N'Dounga. No increase in storage was observed in bare fallowed plots at Kala Paté. Mean increases in storage efficiency at Chikal and N'Dounga were 7.1% and 11.8%, respectively. Position of the zero‐flux boundary during the dry season and profile water distribution toward the end of the dry season indicate that drainage was substantial at all sites. However, large losses due to evaporation are also possible, due to high evaporative demand during the dry season. Data obtained in this study suggest that bare fallowing, with and without dry‐season mulching, is an inefficient means of water conservation for these soil systems. A possible application of bare fallow is in sandy soils underlaid by laterite in high‐risk, low‐rainfall zones of the Sahel if accompanied by adequate erosion prevention.