基于季节气候预测的江淮流域水稻产量预测研究

Seasonal climate forecast is a useful tool that can assist in reducing the impact of anomalous climate on agriculture and improving the agricultural management and decision-making ability. However, the studies integrating seasonal climate prediction and agriculture are not common in China. In this research, by combining the seasonal climate model forecasts and the statistical downscaling method, useful forecasting information for the prediction of the climate variables in rice growing season over the Yangtze-Huaihe River valley is extracted from the climate models to construct the statistical downscaling forecast model for different station and rice growing period. In addition, the relationships between the climate anomalies in each rice growing period over the Yangtze-Huaihe River valley and the large-scale predictors including the El Niño Southern Oscillation, North Atlantic Oscillation and the North Pacific Oscillation etc. are investigated. Daily variations of the climate in Yangtze-Huaihe River valley are predicted at the station scale according to the historical climate scenarios associated with the different phases of the selected climate indices. Based on the research above, combining with the group independently developed Rice Cultivational Simulation-Optimization-Decision Making System and the introduced rice growth simulation models, accuracy of the yield forecasts based on the seasonal climate forecasts from the dynamic climate model and from the climate indices are evaluated and validated. And the dynamic and statistical combined, real-time and quantitative rice yield forecasting technology is constructed for the Yangtze-Huaihe River valley at the station scale, which can provide scientific basis for rice management decision and disaster prevention and reduction in Yangtze-Huaihe River valley.

季节气候预测是降低异常气候对农业生产影响、提高农业管理决策水平的有效工具，然而我国在季节气候预测信息与农业生产结合方面的研究并不多。本研究通过季节气候模式预测与统计降尺度结合的方法，提取气候模式对江淮流域水稻生长季气候要素预测的高技巧信息，针对不同站点及水稻生育时期建立气候要素统计降尺度预测模型。同时，深入分析江淮流域水稻各生育时期气候异常与厄尔尼诺-南方涛动、北大西洋涛动、北太平洋涛动等大尺度气候因子的关系。根据关键气候因子不同位相的历史气候情景，进行研究区域水稻生长季逐站点、逐日气候要素的统计预测。在上述基础上，结合项目组自主研制的“水稻栽培模拟优化决策系统”及国外引进水稻生长模型，评估、检验基于动力气候模式和基于气候指数的季节气候预测方法对江淮流域水稻产量的预测效果，建立统计-动力相结合的站点尺度水稻产量实时、定量预测技术，为江淮流域水稻生产管理决策及防灾减灾提供科学依据。

气候变化背景下，异常天气、气候事件频发，给农业生产带来了极大危害。本研究利用88项大气环流指标，通过相关分析、回归分析等方法，提取了江淮地区各站点最高温度、最低温度、降水和日照时数等气象要素的关键预报因子，构建了江淮地区水稻生长季气候要素的统计降尺度预测模型，并对逐月、逐站点自动输出预测文件。将构建的统计降尺度模型应用于气候模式输出的大尺度环流信息，实现了动力-统计相结合的站点尺度月气象要素预测。鉴于统计降尺度模型对温度预测效果较好，而对降水和日照时数预测效果较差，本研究采用项目组自主研制的水稻高温模型，结合月尺度温度预测结果和天气发生器，实现了江淮地区站点尺度水稻结实率预测，对江淮地区5个代表站点2004-2011年相对结实率预测的标准均方根误差为7.5%。本研究还分析了大尺度气候因子与水稻生长季气候异常的关系及其对水稻产量的影响。研究发现，厄尔尼诺年，江淮流域6月温度偏低，7月、8月温度偏高，水稻产量较常年偏低。而拉尼娜年，江淮流域6月温度偏高，7月、8月温度偏低，水稻产量较常年偏高。此外，东亚高空西风急流、北大西洋涛动、西太平洋副热带高压等其他大尺度气候因子对江淮地区水稻生长季气候异常及产量也有重要影响。根据大尺度气候因子的不同位相，可为江淮地区水稻产量波动提供有物理意义的前兆信号。综上所述，本研究在动力气候模式迅速发展、精度和分辨率不断提高的背景下，将气候模式预测信息及大尺度气候因子与水稻产量的统计关系应用于作物模拟和产量预测，对降低异常气候对农业生产的影响，保障粮食安全具有重要意义。