干旱区生态环境知识资源中心

The yield of dryland wheat in semi-arid and arid areas is limited by rainfall. Nitrogen application and rainfall distribution determine biomass production, soil water depletion, and grain quality. A precise base level of nitrogen fertilization is applied according to the annual rainfall, but in case of more rain, the higher biomass production would dilute the nitrogen and a low quality wheat would be harvested. On the other hand, under drought conditions, harvesting for hay or silage provides a greater income than leaving the crop for grain production. Our objective was to establish a quick and simplified decision support system (DSS) for decision making at heading. It was found that, at heading, flag leaf water concentration (FLW) and flag leaf total N concentration (FLN) data can be used to support agronomic decision making. In particular, these data can assist a decision to harvest early for hay or sila-e, since water stress exists and the test weight is expected to decline. In other cases these data can help to forecast the need for late nitrogen application to ensure sufficient protein levels. Our results show that the proposed DSS correctly forecasts wheat grain quality, test weight and protein content, in more than 80% of the 344 experimental plots, by monitoring flag leaves at heading. Therefore, application of the suggested simplified DSS would reduce the harvesting of shriveled grains, on the one hand, and would lead to improved grain protein, on the other hand, thus ensuring high-quality production. (C) 2004 Elsevier B.V. All rights reserved.