Study on soil moisture prediction models of summer maize field in northern China | 北方夏玉米农田土壤水分预测模型的研究

chinois. 检验和比较4个夏玉米农田土壤水分预测模型的预测精度。根据农田土壤水分平衡原理，建立北方夏玉米农田土壤水分预测模型，其中潜在蒸散量子模型分别用国内彭曼修正式（简称国内法）和FAO Penman-Monteith（简称FAO法）方法，作物发育期子模型分别用多年平均发育期法和积温法。用2002年6～9月中国气象局农业气象试验基地的21个不同水分处理小区的夏玉米全生育期观测资料（包括土壤湿度、作物发育期、气象和灌溉等资料），对模型进行检验和比较。在土壤湿度中等和较低的情况下，土壤水分预测模型中潜在蒸散量子模型采用国内法对不同预测时效下夏玉米农田土壤水分的平均预测精度高于FAO 法，而在土壤湿度较高的情况下则反之；国内法和FAO法在不同预测初始日期下，对所有试验小区土壤水分的平均预测精度均随着预报时效的增加而降低。土壤水分预测模型中，作物发育期子模型采用积温法或多年平均发育期法均可对北方地区夏玉米农田土壤水分进行预测，两者对0～100 cm土层土壤水分预报的最大绝对差值小于5 mm。土壤水分预测模型中潜在蒸散量子模型采用国内法对夏玉米农田土壤水分的预测精度高于FAO法；土壤水分预测模型中作物发育期子模型采用多年平均发育期法和积温法对夏玉米农田土壤水分的预测精度基本相同。利用该模型还可对北方夏玉米农田进行灌溉预报和排渍预报。

anglais. The simulation accuracy of four soil moisture prediction models of summer maize field was tested and compared. Based on soil water balance principle of field, four soil moisture prediction models of summer maize in north china were developed. The FAO Penman-Monteith (‘FAO Model’ for short) and Chinese Penman equations (‘Chinese Model’ for short) were adopted in reference evapotranspiration sub-models. The crop sub-models at development period were established using multi-year mean growth period method and accumulated temperature method. Using examination data of summer maize during whole growth phase in 21 test plots of different soil moistures at Agrometeorological Experimental Base China Meteorological Administration from June to September in 2002 (including soil moisture data, crop development period data, meteorology data and so on), the four models were tested and compared. In the test plots of low or middle soil moisture, the average prediction accuracy of soil moisture of summer maize field in different maximum prediction time adopting ‘Chinese Model’ for reference evapotranspiration sub-models in soil moisture prediction models was higher than that adopting ‘FAO Model’. It was inverse in the test plots of higher soil moisture. For ‘Chinese Model’ and ‘FAO Model’, average prediction accuracy of all test plots based on different prediction beginning days reduced with the increase of maximum prediction time. The soil moisture of summer maize field in north of China can be predicted by adopting the multi-year mean growth period method and the accumulated temperature method for crop development period sub-models in soil moisture prediction models. The maximal absolute difference of prediction values of soil moisture in soil layer from 0 to 100 cm for the two equations was less than 5 mm. The simulation accuracy for soil moisture of summer maize field with ‘Chinese Model’ was higher than that with ‘FAO Model’. The simulation accuracy for soil moisture of summer maize field adopting the multi-year mean growth period method for crop development period sub-models in soil moisture prediction models was almost the same with adopting the accumulated temperature method. The irrigation prediction and drainage prediction for summer maize field in north of China also were made using the soil moisture prediction models.