海洋科学
海洋科學
해양과학
MARINE SCIENCES
2014年
10期
17-25
,共9页
Argo 观测资料%数据同化%同化系统(OVALS)%松弛逼近法%POM(Princeton Ocean Model)%OFES(OGCM for the Earth Simulator)资料
Argo 觀測資料%數據同化%同化繫統(OVALS)%鬆弛逼近法%POM(Princeton Ocean Model)%OFES(OGCM for the Earth Simulator)資料
Argo 관측자료%수거동화%동화계통(OVALS)%송이핍근법%POM(Princeton Ocean Model)%OFES(OGCM for the Earth Simulator)자료
Argo data%data assimilation%ocean variational analysis system (OVALS)%nudging%Princeton Ocean Model (POM)%OGCM for the Earth Simulator (OFES) data
使用POM(Princeton Ocean Model)海洋环流模式模拟西北太平洋海域的温盐场,之后运行中国科学院大气物理研究所同化系统(Ocean Variational Analysis System, OVALS),使用Argo观测网格化产品进行2006年整个计算域三维变分同化。无论从平面分布还是断面分布来看,除海表面温度(SST)外,温度同化的效果都比较好,尤其是800 m水深以浅,均方差值甚至能减小1.0℃以上。而且随着同化积分时间的增加,同化后误差减小越来越显著,3、4月份的同化效果明显高于1、2月份。与OFES(OGCM for the Earth Simulator)海洋模式结果相比,同化在一定程度上改进了模拟结果。在POM(Princeton Ocean Model)数值模拟基础上,利用3个西太平洋代表性浮标的观测资料,运用松弛逼近法同化2006年6月断面数据。当松弛系数取0.5时,模拟时间大约半天,模拟值就能快速逼近或完全等于观测值。虽然只对观测剖面进行了数据同化,但该同化可以产生更大范围的效果,而且这种效果可以持续更长时间。
使用POM(Princeton Ocean Model)海洋環流模式模擬西北太平洋海域的溫鹽場,之後運行中國科學院大氣物理研究所同化繫統(Ocean Variational Analysis System, OVALS),使用Argo觀測網格化產品進行2006年整箇計算域三維變分同化。無論從平麵分佈還是斷麵分佈來看,除海錶麵溫度(SST)外,溫度同化的效果都比較好,尤其是800 m水深以淺,均方差值甚至能減小1.0℃以上。而且隨著同化積分時間的增加,同化後誤差減小越來越顯著,3、4月份的同化效果明顯高于1、2月份。與OFES(OGCM for the Earth Simulator)海洋模式結果相比,同化在一定程度上改進瞭模擬結果。在POM(Princeton Ocean Model)數值模擬基礎上,利用3箇西太平洋代錶性浮標的觀測資料,運用鬆弛逼近法同化2006年6月斷麵數據。噹鬆弛繫數取0.5時,模擬時間大約半天,模擬值就能快速逼近或完全等于觀測值。雖然隻對觀測剖麵進行瞭數據同化,但該同化可以產生更大範圍的效果,而且這種效果可以持續更長時間。
사용POM(Princeton Ocean Model)해양배류모식모의서북태평양해역적온염장,지후운행중국과학원대기물리연구소동화계통(Ocean Variational Analysis System, OVALS),사용Argo관측망격화산품진행2006년정개계산역삼유변분동화。무론종평면분포환시단면분포래간,제해표면온도(SST)외,온도동화적효과도비교호,우기시800 m수심이천,균방차치심지능감소1.0℃이상。이차수착동화적분시간적증가,동화후오차감소월래월현저,3、4월빈적동화효과명현고우1、2월빈。여OFES(OGCM for the Earth Simulator)해양모식결과상비,동화재일정정도상개진료모의결과。재POM(Princeton Ocean Model)수치모의기출상,이용3개서태평양대표성부표적관측자료,운용송이핍근법동화2006년6월단면수거。당송이계수취0.5시,모의시간대약반천,모의치취능쾌속핍근혹완전등우관측치。수연지대관측부면진행료수거동화,단해동화가이산생경대범위적효과,이차저충효과가이지속경장시간。
The Ocean Variational Analysis System (OVALS) assimilation system designed by the institute of Atmospheric Physics, Chinese Academy of Sciences and nudging method were used to assimilate the Argo grid data with the resolution of 1°′1° and Argo profile data into Princeton Ocean Model(POM)numerical model, respectively. The assimilation of Argo grid data in 2006 indicated that the 3D variation data assimilation improved the simulation results of temperature, reducing about 1.0℃ in root mean square error of either horizontal or vertical distribution except for SST. Furthermore, the grid data assimilation evolved over time, which was better in March and April than that in January and February. In comparison with independent OGCM for the Earth Simulator (OFES) data, the data assimilation only partially enhanced the accuracy of simulated temperature pattern due to an inherent large difference between OFES data and Argo grid data that were used during the data assimilation. The nudging method was applied to assimilate the Argo profile data into POM numerical model in the summer of 2006. The process of nudging was very rapid, which completed within about half day assuming the relaxation coefficient was 0.5. The nudging assimilation also manifested that the temperature structure even far from the Argo profiles could be rapidly influenced, and the remote impact could remain several days.
