安徽省稻曲病气象等级预报方法研究—以池州为例

为进一步明确安徽省稻曲病发生关键期, 探索稻曲病气象等级预报方法, 以满足对该病害气象等级预报的服务需求?本文以安徽省池州市为例, 利用1995年-2018年一季稻稻曲病观测数据和同期气象资料, 通过相关分析确定稻曲病发生关键期?根据稻曲病大发生对适温高湿环境的需求及不同降水等级和温度对稻曲病发生程度的影响不同, 以稻曲病发生关键期降水日数为基础, 引入雨量系数和温度系数, 形成稻曲病发生综合气象条件指数, 通过最优曲线回归分析, 建立稻曲病预报模型?结果表明, 7月下旬至8月中旬是池州市一季稻稻曲病发生关键期; 综合气象条件指数与稻曲病病穗率相关性明显高于降水日数与病穗率相关性; 预报模型回代检验准确率为81.0%, 2016年-2018年模型预测结果均与实际情况相吻合, 由于样本中轻发生和大发生年份较少, 对轻发生和大发生预报的准确性需在样本丰富条件下进一步验证?模型在业务应用中, 可结合CFSv2模式逐日降水和气温预报产品, 提前10~30 d开展稻曲病气象等级预报, 对做好稻曲病的防治工作具有重要参考价值?

Forecasting methods for meteorological grading of the rice false smut in Anhui province: Chizhou city as a case study

In order to define the critical period of occurrence and explore the forecasting methods for meteorological grading of the rice false smut in Anhui province so as to meet the demands of forecasting the meteorological grades of this disease, this study took Chizhou city as a case study to define the critical period of the occurrence of the rice false smut through correlation analysis, based on the observation data of rice false smut in single-season rice and meteorological data from 1995 to 2018. The composite meteorological conditions index of rice false smut was formed from the number of precipitation days during the critical occurrence period of rice false smut, corrected by the precipitation coefficient and temperature coefficient. Based on the index, the forecast model was created by means of optimal curve regression analysis. The results showed that the critical occurrence period of rice false smut was from late July to mid-September in Chizhou city. The relevance between the composite meteorological conditions index and the percentage of diseased panicles for rice false smut was obviously higher than that between the precipitation days and the percentage of diseased panicles. The accuracy of validation for the forecast model was 81.0%, and the predicted result of the model was corresponding to the observed data from 2016 to 2018. Because there were fewer light occurrence and heavy occurrence years in the samples, the prediction accuracy was further verified using more samples. In the application of the model, the meteorological grade forecast of rice false smut is suggested to be carried out 10-30 days ahead by combining the daily precipitation and temperature forecast products from CFSv2 model, which provided an important reference value for the prevention and control of rice false smut.