边界条件对中尺度数值预报模式影响的试验研究
|
摘要
利用WRF非静力中尺度数值预报模式,对2006年1月13—14日华北地区及山东半岛一次大雾天气过程进行不同大小模拟区域、不同边界条件(分别采用NCEP和T213资料)的模拟对比试验,对模拟预报场与实际分析场进行对比分析.结果表明:中尺度模式初值显著影响模式积分开始后的一段时间,随着时间的延长,边值对模式预报的影响愈加显著;初值与边值影响的时间长短与模式区域有关,模式区域越大,初值影响时间越长,边值影响整个区域的时间越晚;反之,模式区域越小,则模式初值影响时间越短,边值起决定作用的时间越早;模式边界值信息影响模式内区的地面信息的速度大约与中低层天气尺度波动移动速度相当;模式初值影响地面及低层要素的时间比高层要素长,相反边值影响高层要素的时间比低层快得多,这与稳定形势和高低空气流速度的差异有关;模式初值影响热力要素及水汽要素的时间比动力要素要长.这些结论为区域中尺度模式的建立起到一定的指导意义.
Non-hydrostatic model WRF is used to simulate a dense fog case occurred in Huabei area and Shandong Peninsula of China from January 13 to 14,2006,and the WRF simulated in different sizes of model domain,initial fields and lateral boundaries are tested.Results show that the initial-field obviously affects the forecast fields in the beginning of the whole forecast period,and the lateral boundary affects the forecast fields more in the later period.The influence time of initial field and lateral boundary is related to the size of model domain.With the larger size of the model domain,the influence time of initial value is longer,and the influence time of the lateral boundary value on the whole region start more latter.On the contrary,with the smaller size of the model domain,the influence time of initial value is shorter,and the influence time of the lateral boundary value start more early.The speed which affected surface elements of the model inner area because of the lateral boundary condition is the same as the speed of the propagation and movement of weather signals(waves) in the middle and low level.There are more longer time that affected the elements of surface and low level than the high level because of the initial value,otherwise there are faster speed that affects the high level elements than low level,which is related to the stable situation and the difference of the air flow speed between the high level and the low level.The thermal elements and moisture elements are affected in the much longer time by the initial value than move elements.All the results will be the important guidance to the construction of mesoscale numerical weather prediction.
Non-hydrostatic model WRF is used to simulate a dense fog case occurred in Huabei area and Shandong Peninsula of China from January 13 to 14,2006,and the WRF simulated in different sizes of model domain,initial fields and lateral boundaries are tested.Results show that the initial-field obviously affects the forecast fields in the beginning of the whole forecast period,and the lateral boundary affects the forecast fields more in the later period.The influence time of initial field and lateral boundary is related to the size of model domain.With the larger size of the model domain,the influence time of initial value is longer,and the influence time of the lateral boundary value on the whole region start more latter.On the contrary,with the smaller size of the model domain,the influence time of initial value is shorter,and the influence time of the lateral boundary value start more early.The speed which affected surface elements of the model inner area because of the lateral boundary condition is the same as the speed of the propagation and movement of weather signals(waves) in the middle and low level.There are more longer time that affected the elements of surface and low level than the high level because of the initial value,otherwise there are faster speed that affects the high level elements than low level,which is related to the stable situation and the difference of the air flow speed between the high level and the low level.The thermal elements and moisture elements are affected in the much longer time by the initial value than move elements.All the results will be the important guidance to the construction of mesoscale numerical weather prediction.
引文
[1] 陈德辉,薛纪善.数值天气预报业务模式现状与展望[J].气象学报,2004,62(5):623-633.
[2] 王晓君,马浩.新一代中尺度预报模式(WRF)国内应用进展[J].地球科学进展,2011,26(11):1 191-1 199.
[3] 曹巧莲,马旭林,宋方超,等.“7.09”山西暴雨数值模拟与诊断分析[J].云南大学学报:自然科学版,2016,38(4):575-585.
[4] 王将,刘亚玲.不同侧边界方案对正压原始方程组预报效果的比较研究[J].云南大学学报:自然科学版,2013,35(S1):244-252.
[5] 龚建东,邱崇践,王强,等.区域四维变分资料同化的数值试验[J].气象学报,1999,57(2):131-142.
[6] 朱红芳,王东勇,管兆勇,等.不同初始场条件对GRAPES模式数值预报的影响[J].气象学报,2007,65(4):493-502.
[7] 闫敬华,MAJEWSKI D.中尺度数值预报模式初边值作用的试验研究[J].热带气象学报,2003,19(4):337-344.
[8] 李元平,梁爱民,张中锋,等.北京地区一次冬季平流雾过程数值模拟分析[J].云南大学学报:自然科学版,2007,29(2):167-172.
[9] 梁爱民,张庆红,刘开宇,等.华北地区一次大雾过程的三维变分同化试验[J].气象学报,2007,65(5):792-804.
[2] 王晓君,马浩.新一代中尺度预报模式(WRF)国内应用进展[J].地球科学进展,2011,26(11):1 191-1 199.
[3] 曹巧莲,马旭林,宋方超,等.“7.09”山西暴雨数值模拟与诊断分析[J].云南大学学报:自然科学版,2016,38(4):575-585.
[4] 王将,刘亚玲.不同侧边界方案对正压原始方程组预报效果的比较研究[J].云南大学学报:自然科学版,2013,35(S1):244-252.
[5] 龚建东,邱崇践,王强,等.区域四维变分资料同化的数值试验[J].气象学报,1999,57(2):131-142.
[6] 朱红芳,王东勇,管兆勇,等.不同初始场条件对GRAPES模式数值预报的影响[J].气象学报,2007,65(4):493-502.
[7] 闫敬华,MAJEWSKI D.中尺度数值预报模式初边值作用的试验研究[J].热带气象学报,2003,19(4):337-344.
[8] 李元平,梁爱民,张中锋,等.北京地区一次冬季平流雾过程数值模拟分析[J].云南大学学报:自然科学版,2007,29(2):167-172.
[9] 梁爱民,张庆红,刘开宇,等.华北地区一次大雾过程的三维变分同化试验[J].气象学报,2007,65(5):792-804.