利用分析Nudging方法,选取辽宁省2015年7月和10月作为夏季和秋季代表月份,开展不同分辨率嵌套网格上的分析同化试验,研究在不同分辨率的嵌套网格上使用分析Nudging对地面要素降尺度模拟的影响。结果表明,采用分析Nudging后,可以明显减小WRF模式模拟高空要素与大尺度背景场的差异,使模式在长时间的积分过程中能够与大尺度背景场保持很好的一致性,且在12 km和4 km分辨率上均采用分析Nudging相对于仅在36 km分辨率嵌套上采用分析Nudging效果更好。对于地面要素影响方面,不同分辨率嵌套网格采用分析Nudging相对于控制试验均可以提高地面2 m温度、相对湿度、10 m风速和降水的模拟准确度。总体来看,分析Nudging应用于12 km分辨率上的降尺度效果整体最优,当进一步在4 km分辨率嵌套上使用分析Nudging后,反而使得模式对温度、降水的模拟变差,说明在区域气候降尺度中,考虑大尺度背景场本身的分辨率,不宜在高分辨率嵌套网格上进一步采用分析Nudging。

Analysis Nudging in different resolution domains was adopted for dynamic downscaling, and the result of two typical months of July and October, 2015, which represented summer and autumn, was evaluated in Liaoning Province of China. Results showed that significant improvement was present in large-scale background with AN in WRF, such as geopotential height at 500 hPa and relative humidity at 925 hPa. According to the characteristics of error between simulated and observation of surface variables, including temperature and relative humidity at 2 m, wind speed at 10 m and precipitation, it also showed that analysis nudging at 12 km and 4 km resolution was more effective than at 36 km. However, after using analysis nudging at 4 km, there was no improvement of the precision of surface variables anymore, and the surface temperature and precipitation were even worse than using nudging at 12 km grid. This suggested that, according to the resolution of background filed, nudging in high resolution grid was ineffective after using analysis nudging at the relatively coarse model grid.