WATER-USE BY ANNUAL GREEN MANURE LEGUMES IN DRYLAND CROPPING SYSTEMS

Data on plant water use and soil water depletion by green manure legum es are needed to develop sustainable cropping systems in a semi-arid e nvironment. The objectives of this study were to determine: (i) season al water use by legumes, (ii) their water use efficiency (WUE), and (i ii) residual soil water contents after legume growth compared with sum merfallow and continuous spring wheat (Triticum aestivum L.). Black le ntil (Lens culinaris Medikus), Tangier flatpea (Lathyrus tingitanus L. ), chickling vetch (Lathyrus sativus L.), and feed pea (Pisum sativum L.), were seeded into wheat stubble with snow trap strips on an Orthic Brown Chernozem soil (Aridic Haploborolls) at Swift Current, SK, from 1984 to 1990. Legume water use exceeded that of fallow at 4 to 6 wk a fter seeding. When legumes were tilled into the soil at full bloom, di fferences in water content between cropped treatments and fallow were largest and most consistent in the top 0.6 m, suggesting that the legu mes extracted water primarily from this depth. In an extreme drought y ear, substantial water depletion occurred below 0.6 m. Water use effic iency of legumes was 11 to 29 kg ha(-1) mm(-1), similar to that of spr ing wheat. Feed pea and chickling vetch used water more efficiently th an the other legumes or N-fertilized wheat. Significant differences in WUE across years were related to differences in DM production. Soil w ater contents above the 0.6-m depth in late fall following green manur ing were 62 to 82% that of fallow and increased over winter to 79 to 1 03% of fallow. Subsoil water was, on average, recharged after wheat to only 68% but following green manure to 81% of fallow. Partial fallow replacement with legumes would reduce the risk of erosion and nutrient leaching and minimize the hazard of salinization and eutrophication o f downstream ecosystems.