耦合作物模型与水文模型的农业干旱评估方法研究

Agricultural drought has the essential feature of crop water shortage. How to obtain the crop water variation datasets with high spatial-temporal resolution, and how to construct the crop water stress index, are the key issues for the efficient assessment of agricultural drought. In this project, crops water demand under the suitable condition and water consumption under the actual condition will be simulated through coupling crop model with hydrologic model. A drought index will be built and the quantitative assessment method of regional agricultural drought will be studied. Specifically, the optimized crops parameters will be determined using the remote sensing data. A coupling method of crop model with hydrologic model will be conducted. Based on the assimilation datasets of the multi-source soil moisture, the remote sensing of evapotranspiration and leaf area index, the dynamic process of regional crops water demand and consumption with high spatial-temporal resolution will be simulated. Finally, quantitative assessment method of regional agricultural drought will be presented with the constructed improved crop water stress index (ICWSI). The research will improve the comprehensive understanding of drought spatial-temporal evolution characteristics. It is critical for the precise assessment of agricultural drought and can provide theoretical principle and technical support for effective prevention and mitigation of drought disaster.

农作物水分亏缺是农业干旱的本质特征，获取高时空分辨率作物水分变化过程数据，构建作物水分亏缺指数是有效评估农业干旱的关键。本项目将通过耦合作物模型与水文模型，模拟适宜条件下作物需水量与实际条件下作物耗水量，构建干旱指数，研究区域农业干旱定量评估方法。项目将利用地面遥感反演数据优化作物模型参数，并对作物模型与水文模型进行耦合，在同化多源土壤含水量数据、遥感反演蒸散发与叶面积指数的基础上，模拟区域尺度高时空分辨率的农作物生长需水和耗水变化过程。构建改进的作物缺水指数（ICWSI），据此提出区域农业干旱定量评估方法。研究成果对认识农业干旱时空演变规律，准确评估农业旱情意义重大，可为抗旱减灾提供重要的理论基础和技术支撑。

基于模拟土壤含水量构建的干旱指数已被广泛用于大范围农业干旱评估中。当前模拟用于干旱评估的土壤含水量时，常采用水文模型进行模拟。然而，水文模型常过于简化甚至缺少作物模拟方案，无法模拟作物需水量，仅仅基于土壤含水量构建的土壤干旱指数因忽略作物需水而难以准确评估实际的农业干旱情况。为此，项目构建了水文-作物耦合模型VIC-EPIC，提出了基于VIC-EPIC模拟作物需耗水过程的作物缺水距平指数CWAPI（Crop Water Anomaly Percentage Index），开展了区域农业干旱评估方法研究。项目主要内容和结论如下：.在提出灌溉过程对作物蒸散发影响的参数化方案的基础上，构建了VIC-EPIC模型，基于水分胁迫方法改进了耦合模型中的蒸散发模块，模拟区域流量、土壤含水量和蒸散量变化过程。在江苏北部青口河流域的验证结果表明，验证期VIC-EPIC模型的日流量纳什效率系数为0.76，高于VIC模型的0.64；VIC-EPIC模型的蒸散发模块模拟效果好于VIC和EPIC模型。可见，模型适用于流域水文过程和作物耗水量的模拟。.基于VIC-EPIC模拟作物需耗水过程构建了CWAPI并提出相应的农业干旱评估方法。在青口河流域验证结果表明，VIC-EPIC模拟得到的作物需耗水量可反映不同轮作方式下作物水量的变化过程，较大的缺水量与当年因旱成灾面积的较大统计值相吻合。在此基础上，基于因旱成灾面积统计数据验证分析表明，构建的CWAPI能够直接反映作物的缺水状态和缺水对作物生长的累积影响，CWAPI能够较SMAPI（Soil Moisture Anomaly Percentage Index）更客观地反映区域作物旱情。.基于CWAPI干旱指数研究了江苏省农业干旱时空变化特征，揭示了土壤水分亏缺与作物水分亏缺对农业干旱形成的影响规律。结合实际旱情记录和SMAPI对CWAPI结果进行验证：发现受旱面积统计数据与CWAPI模拟的轻旱面积的相关系数为0.79，CWAPI表征的干旱特征值对记载的较严重干旱事件能够较好地吻合；因为CWAPI直接考虑了作物休耕、作物需水、灌溉条件等因素的影响，CWAPI在评估农业干旱方面相比SMAPI更具合理性。