Optimizing wheat productivity in two rain-fed environments of the West Asia-North Africa region using a simulation model

The performance of a crop simulation model (agricultural production systems simulator model, APSIM-Nwheat) was tested using data obtained from several locations in the rain-fed environments of West Asia and North Africa (WANA) in Morocco and Jordan. The model was able to simulate wheat grain yields reasonably well except at one site in one season in Morocco. The model was subsequently used to analyze the effect of soil type (soil water-holding capacity), rate and timing of nitrogen (N) fertiliser, initial soil moisture storage, cultivars (early versus late), sowing dates and density and supplemental irrigation (SI) in optimizing wheat production using 20 years of historical weather records from Morocco. The simulation indicated that yields were often limited by the amount and timing of rainfall. While the effect of N fertiliser was minimal or detrimental in dry years, it improved grain yields in wet years and when crops were sown early combined with pre-sown stored plant available water in the soil. The analysis showed that early sowing is important for achieving high yields by avoiding terminal water deficit. There is little difference between grain yields when current practice of about 300 plants/m(2) was compared with a density of 150 plants/m(2). This implies that there, is scope for reducing current planting density to save seeds without reducing yields. The simulation analysis highlighted that 40 mm of SI at sowing significantly improved average grain yields as a result of enabling early crop establishment, in particular with a N fertiliser application of 40 kg N/ha. The analysis indicated that wheat grain yields in the arid and semi-arid rain-fed regions of WANA can be improved compared to current yield levels by adjusting N management to soil type, pre-sowing soil water availability, sowing opportunity and the availability of SI. (c) 2006 Elsevier B.V. All rights reserved.