大气科学学报
大氣科學學報
대기과학학보
JOURNAL OF NANJING INSTITUTE OF METEOROLOGY
2012年
6期
641-651
,共11页
集合预报%NCEP%亚欧区域%北美区域%预报效果
集閤預報%NCEP%亞歐區域%北美區域%預報效果
집합예보%NCEP%아구구역%북미구역%예보효과
ensemble forecasting%NCEP%Eurasia region%North America region%forecast performance
使用NCEP集合预报系统(EPS)输出的500hPa位势高度场预报资料和相应的NCEP/NCAR再分析资料,针对集合平均预报和概率预报,采用多种预报效果检验评价方法,对该系统在亚欧和北美区域的预报效果进行全面的分析比较。总体而言,NCEP—EPS对亚欧区域的环流集合预报效果不亚于其对北关区域的预报效果。1)ACC检验表明,亚欧区域的集合平均预报效果在除冬季外的三个季节都明显优于北美区域,可用预报的时效相差达0.6~1d,且夏季的差别最大。RMSE检验表明,亚欧区域的预报效果在四个季节里均优于北美区域。2)集合概率预报可靠性的季节差别不明显,均为预报时效较短(长)时,北关(亚欧)区域的可靠性更好。系统对亚欧区域的事件识别范围相对较小,但其预报可靠性较高,北美区域则正好相反。3)夏季亚欧区域的集合概率预报效果明显优于北美区域,秋季和冬季北关区域的预报效果较好,春季在预报时效小于5d时北美区域占优,而其后则是亚欧区域的预报分辨能力更好。
使用NCEP集閤預報繫統(EPS)輸齣的500hPa位勢高度場預報資料和相應的NCEP/NCAR再分析資料,針對集閤平均預報和概率預報,採用多種預報效果檢驗評價方法,對該繫統在亞歐和北美區域的預報效果進行全麵的分析比較。總體而言,NCEP—EPS對亞歐區域的環流集閤預報效果不亞于其對北關區域的預報效果。1)ACC檢驗錶明,亞歐區域的集閤平均預報效果在除鼕季外的三箇季節都明顯優于北美區域,可用預報的時效相差達0.6~1d,且夏季的差彆最大。RMSE檢驗錶明,亞歐區域的預報效果在四箇季節裏均優于北美區域。2)集閤概率預報可靠性的季節差彆不明顯,均為預報時效較短(長)時,北關(亞歐)區域的可靠性更好。繫統對亞歐區域的事件識彆範圍相對較小,但其預報可靠性較高,北美區域則正好相反。3)夏季亞歐區域的集閤概率預報效果明顯優于北美區域,鞦季和鼕季北關區域的預報效果較好,春季在預報時效小于5d時北美區域佔優,而其後則是亞歐區域的預報分辨能力更好。
사용NCEP집합예보계통(EPS)수출적500hPa위세고도장예보자료화상응적NCEP/NCAR재분석자료,침대집합평균예보화개솔예보,채용다충예보효과검험평개방법,대해계통재아구화북미구역적예보효과진행전면적분석비교。총체이언,NCEP—EPS대아구구역적배류집합예보효과불아우기대북관구역적예보효과。1)ACC검험표명,아구구역적집합평균예보효과재제동계외적삼개계절도명현우우북미구역,가용예보적시효상차체0.6~1d,차하계적차별최대。RMSE검험표명,아구구역적예보효과재사개계절리균우우북미구역。2)집합개솔예보가고성적계절차별불명현,균위예보시효교단(장)시,북관(아구)구역적가고성경호。계통대아구구역적사건식별범위상대교소,단기예보가고성교고,북미구역칙정호상반。3)하계아구구역적집합개솔예보효과명현우우북미구역,추계화동계북관구역적예보효과교호,춘계재예보시효소우5d시북미구역점우,이기후칙시아구구역적예보분변능력경호。
Based on the forecast dataset of 500 hPa geopotential height fields from NCEP ensemble prediction system(EPS) and the corresponding NCEP/NCAR reanalysis dataset, the ensemble forecasting performance in Eurasia region (EA) is compared with the performance in North America region (NA). By several forecast verification and evaluation methods, the comparisons focus on the ensemble mean forecasts and ensemble probabilistic forecasts. In general, the NCEP-EPS circulation ensemble perform- ance in EA is nothing less than the prediction in NA. 1 ) ACC shows that, except in winter, the ensemble mean forecasts in EA in the other three seasons are clearly superior to those in NA, and their difference of maximal lead time of useable forecasts is 0.6-1 d, with the largest difference in summer. RMSE shows that the forecast results in EA in the four seasons are better than those in NA. 2) The seasonal differences of reliability of ensemble probability forecasting are not evident. For the shorter (longer) lead time, the reliability in NA (EA) is better. The identification range of synoptic events in EA is rela- tively smaller, but the reliability is relatively higher in the region. This conclusion is opposite in NA. 3 ) In summer, the ensemble probability predictions in EA are clearly superior to those in NA. In autumnand winter, there are better results in NA. In spring, when the lead time is less than 5 d, the results in NA are better, but when it is more than 5 d, there is better forecasting resolution in EA.