青海夏季干旱特征及其预测模型研究

利用1961-2008年青海非干旱区(除柴达木盆地)地面气象观测资料、 74个环流特征量、 海温资料、 北半球500 hPa高度场网格点资料以及500 hPa高度场遥相关, 对夏季干旱的变化趋势和干旱发生的机理进行了研究.结果表明:1961-2008年夏季青海省非干旱区、 东部农业区分别发生干旱15 a、 18 a, 发生干旱的年几率为31.3%、 37.5%; 东部农业区发生干旱的几率较大, 中轻度干旱发生几率大于特大、 重度干旱.夏季典型干旱年500 hPa欧亚中高纬度上空高度距平分布为正距平, 极涡偏弱; 非干旱年蒙古到青藏高原上由负距平控制, 极涡偏强, 偏向东半球, 印缅低压槽十分活跃.当夏季西大西洋型、 上年秋季欧亚纬向环流指数偏弱, 而4月西太平洋型偏强, 8月青藏高原地面加热场强度距平指数偏强, 夏季容易发生干旱; 反之, 当夏季西大西洋型、 上年秋季欧亚纬向环流指数偏强, 而4月西太平洋型偏弱, 8月青藏高原地面加热场强度距平指数偏弱, 则夏季不易发生夏季干旱. 1961-2008年模拟方程的准确率为83.3%, 2009-2010年预测结果与实况接近, 趋势预测准确.

Characteristics and Prediction Model of Summer Drought in Qinghai Province

In this paper, the principle and change tendency of summer drought are studied with the data such as surface observations in Qinghai Province (except for Qaidam Basin), 74 circulation features and reseau data at 500 hPa altitude field of Northern Hemisphere. The results indicated that: 1) there were 15 drought events in the summer of Qinghai Province during 1961-2008, with the probability of 31.3% per years. The probability of moderate drought is larger than that of severe drought and extra-severe drought. The probability of drought is larger in eastern agricultural area than that in other areas. 2) In typical summer drought years, anomaly at 500 hPa altitude upon Eurasian mid-high latitude is positive, with weak polar vortex; in non-dry years, over Mongolia to the Tibetan Plateau there is negative anomaly, with stronger polar vortex towards the eastern hemisphere, and an active India-Burma trough. 3) The western Atlantic in summer and Eurasian zonal circulation index weak last autumn in April Western Pacific Partial August the strong side of the Qinghai-Tibet Plateau surface heating field intensity anomalies index prone to drought in summer; Conversely, West Atlantic type in summer, Eurasian zonal circulation index last autumn on the strong side, the West Pacific-type weak in April, August, Qinghai-Tibet Plateau surface heating field intensity anomaly index weaker, less drought in summer. 4) Back substitution of the simulation equation in 1961-2008 indicates that the accuracy is 83.3% (42/48). Severe drought events in 1977, 2000 and 2002 can be seen in the prediction. The prediction result for 2009-2010 is close to the truth, showing the trend prediction being credible.