论文部分内容阅读
利用ASCAT海面反演风场资料对SMB—WARMS中尺度模式的海面风场预报进行了检验,结果表明:对24h、48h和72h3个时次的风速预报,4级风的准确率(TS)最大,然后向风级的两端(0和9级)TS逐渐降低;3个预报时次的漏报率(PO)随风级增大逐渐增大,而空报率(FAR)随风级增大逐渐减小。3个预报时次的风向预报评分接近,为0.6左右。在月际变化上,对4级风,各月3个预报时次的TS、PO和FAR的变化规律都比较一致,TS在12月最小,PO在12月和4月出现较低值,FAR在12月份有明显的高值;11月和4月风向预报准确度较低。对4级风的预报,模式的24h预报准确率TS在东海中部和南部较高,48h在东海北部、南部海域和台湾岛东南海域较高,72h在东海中部海域和台湾岛以东海域较高。3个预报时次的风向预报评分在低纬度海域要高于高纬度海域。风速的24h、48h和72h预报在台湾海峡都存在最大的均方根误差,为3.5~4m/s。
The sea surface wind field forecast of SMB-WARMS mesoscale model was tested by ASCAT sea surface wind field data. The results show that the accuracy of wind field wind speed (TS) is the highest at 24h, 48h and 72h , And then gradually decreased toward both ends of the wind level (0 and 9) TS; PO (3) decreased gradually with the increase of the wind level at the three forecast times, while the FAR increased with the wind level Greatly diminished. The wind forecasting scores of the three forecasting times are close to 0.6. For the monthly variation, TS, PO and FAR of the four winds and the three forecast periods of each month are all consistent. TS is the smallest in December and PO is lower in December and April. FAR In December there is a clear high value; in November and April the wind forecast accuracy is low. For the prediction of 4-level wind, the prediction accuracy 24h of pattern in the middle and southern part of the East China Sea is higher, and it is higher in the northern part of the East China Sea, the southern part of the sea and the southeast of Taiwan Island in 48h, and is higher in the middle part of the East China Sea and the east of Taiwan Island . The wind forecasting scores of the three forecasting times are higher in the low latitudes than in the high latitudes. Wind speed 24h, 48h and 72h forecast in the Taiwan Strait are the largest root mean square error of 3.5 ~ 4m / s.