Study On Rice Water Demand Prediction Models Under The Wetting Irrigation

With the rapid development of economy, the imbalance between supply and demand of water resources is aggravated increasingly. Water shortage has become a global issue of common concern. Study of rice submerged irrigation is more common, while wetting irrigation condition is still less. For analyzing scientifically and comprehensively the water-saving potential paddy and realizing efficient water use of agriculture, to study rice evapotranspiration and water demand rule under the wetting irrigation is essential, which has some theory significance and application value.A field experiment in2009was conducted to study rice water demand prediction models by stepping multi-regression analysis under the wetting irrigation. The main contents include:(1) The mechanism of microclimate factors(including temperature, pressure, relative humidity and wind, etc) affecting rice transpiration, evapotranspiratior intensity was analyzed and studied;(2) The response sensitivity of rice transpiration and evapotranspiration on influencing factors was analyzed and tested to study their variation law;(3) Using the SPSS data statistical analysis software analyzed the key factors affecting rice transpiration and evapotranspiration under the wetting irrigation removing the secondary factors, and established the optimal prediction models for rice transpiration and evapotranspiration. The error analysis was conducted to verify their reliability and practicality.The results showed that the two multi-linear stepwise regression models built u(?) based on major meteorological factors were satisfying, and the maximum relative error(?) are4.39%,5.87%respectively. The research can lay a foundation for forecast of rice water requirement under the wetting irrigation, and supply reliable references fo(?) rational allocation and scientific management of water resources.