基于ARIMA模型的云南气象干旱预测研究

Study on meteorological drought prediction in Yunnan Province based on ARIMA

近年来，云南地区的干旱引起了社会广泛关注。为了对云南省未来一个时期的干旱情况进行预测，利用1961~2013年云南省29个气象站逐月降水和气温资料计算云南各地区标准化降水蒸散指数（ SPEI），并建立了SPEI序列自回归移动平均模型（ ARIMA），对模型参数进行率定和验证后，确定了模型参数和表达式。利用所建立的模型，对云南省各地区2014~2023年年平均SPEI值进行预测。结果显示，未来一段时期内云南各地区气象干旱仍将持续。针对云南的干旱情况，提出了相应的抗旱对策建议。

The drought of Yunnan Province has caused wide public concern in recent years. To predict the drought of Yunnan Province in future period, the time series of monthly precipitation and monthly average temperature from 29 metrological stations in Yunnan Province during 1961 to 2013 were used to calculate the SPEI, and ARIMA models was built based on the SPEI. After calibration and validation, the model parameter and expression were determined. Then the annual SPEI from 2014 to 2023 in dif-ferent regions of Yunnan Province were predicted by the model. The result showed that the meteorological drought will be contin-ued in the future period, some countermeasures and suggestions for drought resistance are put forward.