效率系数和一致性指数及其在干旱预测精度评价中的应用

发布时间：2018-12-11 00:48

【摘要】：基于关中平原2003—2014年Aqua-MODIS数据的条件植被温度指数的干旱监测结果,引入效率系数和一致性指数对自回归求和移动平均(ARIMA)模型和季节性ARIMA模型的预测精度进行了评价,并分析了其在干旱预测精度评价中的适用性。两种干旱预测模型的精度评价结果为ARIMA模型与季节性ARIMA模型的效率系数分别为-0.04与-4.27,一致性指数分别为0.40与0.37,表明ARIMA模型对干旱的预测精度高于季节性ARIMA模型,这与均方根误差、皮尔森相关系数及Kappa系数等的评价结果一致,且效率系数对不同干旱预测模型的预测性能的区分效果尤为显著,对干旱预测模型的精度评价比一致性指数更为合理。因此,效率系数更适合用于遥感干旱预测模型的精度评价。
[Abstract]:Based on the drought monitoring results of conditional vegetation temperature index of Aqua-MODIS data from 2003 to 2014 in Guanzhong Plain, the prediction accuracy of autoregressive summation moving average (ARIMA) model and seasonal ARIMA model was evaluated by introducing efficiency coefficient and consistency index. And its applicability in drought prediction accuracy evaluation is analyzed. The accuracy evaluation results of the two drought prediction models are as follows: the efficiency coefficients of ARIMA model and seasonal ARIMA model are -0.04 and -4.27, and the consistency index are 0.40 and 0.37, respectively. The results show that the prediction accuracy of ARIMA model for drought is higher than that of seasonal ARIMA model, which is consistent with the evaluation results of root mean square error, Pearson correlation coefficient and Kappa coefficient, etc. Moreover, the efficiency coefficient can distinguish the prediction performance of different drought prediction models, and the accuracy evaluation of drought prediction model is more reasonable than the consistency index. Therefore, the efficiency coefficient is more suitable for the precision evaluation of remote sensing drought prediction model.
【作者单位】：中国农业大学信息与电气工程学院;江苏省农业科学院农业经济与信息研究所;
【基金】：国家自然科学基金项目(41371390;41071235) 国家科技支撑计划项目(2012BAH29B03)
【分类号】：S423

【参考文献】