| 摘要: |
| 建立了一个功能较为完善的四维同化系统, 解决了温度调查资料中存在的同步性问题。为了检验同化方法和模式程序的正确性, 以及讨论同步性误差的校正效果, 从同化结果与锚系调查数据、同化结果与遥感数据、比对断面均方根误差和海表平均温度 4 个方面进行了探讨, 结果表明: 1) 同化结果与锚系调查数据对比显示, 改进参数的 NMC 方法(来自美国国家气象中心)能够适用于长江口外海域中的温度四维同化, 同化温度结果与锚系调查数据在变化趋势上一致, 在数值上最大差异不超过 0.4 ℃ ; 2)同化结果与遥感数据对比显示, 同化结果能够将调查资料推演到完整的时空范围, 并具有较好的变化趋势和精度; 3) 根据比对断面均方根误差分析, 调查资料中两个比对断面温度间的均方差误差为 3.8℃, 而同化结果与实测数据的均方根误差低于 0.6℃, 说明同化方法有效地降低了调查资料中的同步性误差; 4) 海表平均温度分析显示, 同化结果能够避免调查不同步引起的各种温度斑块, 并且其温度分布和锋面结构能够更如实地反映实际情况。 |
| 关键词: 数据同化, 浅海调查, 长江口邻近海域, 同步性 |
| DOI：10.11693/hyhz201201003003 |
| 分类号: |
| 基金项目:国家重点基础研究发展计划(973)项目,2011CB409803 号, 2011CB403503 号, 2006CB400603 号; 我国近海海洋综合调查与评价(908)专项, 908-01-BC06, 908-ZC-I-01 号; 国家自然科学基金项目, 41176021 号; 浙江省自然科学基金项目, Y5110014 号; 国家海洋局第二海洋研究所基本科研业务费专项资助, JG1010 号 |
|
| APPLICATION OF DATA ASSIMILATION TO SYNOPTIC TEMPERATURE MAPPING OF THE COASTAL OCEAN SURVEY |
|
XUAN Ji-Liang1, HUANG Da-Ji1,2, ZHOU Feng1,2, ZHU Xiao-Hua1,2, FAN Xiao-Peng1, NI Xiaobo1, XING Chuan-Xi1
|
|
1.State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration;2.Department of Ocean Science and Engineering, Zhejiang University
|
| Abstract: |
| A 4-dimensional data assimilation system is improved and applied to resolve the synoptic problem in the cruise survey data usages. The validation of the assimilation method is investigated with a set of inter-comparisons between the assimilated temperature and observed temperature from the mooring data, the remote sensing satellite data and the cruise data. 1) Comparisons between the assimilated temperature and the mooring observed temperature show that
NMC (National Meteorological Center) method with modified parameters is more suitable than the IV (Innovation Vector) method in temperature data assimilation for sea areas off Changjiang estuary. The assimilated temperature is in same trend with the mooring observed temperature, with a maximum temperature difference of 0.4℃. 2) Comparisons between the assimilated sea surface temperature (SST) and the remote sensing SST show that assimilated SST are able to reproduce the spatial and temporal distribution of survey data, with better trend and accuracy. 3) The analysis of root mean square error
(RMSE) indicates RMSE is 3.8oC and lower than 0.6℃ for the observation section and assimilation section. This implies that errors caused by assimilation method are much lower than that caused by synoptic problem. 4) Analysis of surface mean temperature shows that using assimilation data can avoid unreal temperature patches, which is a mistake caused by synoptic problem, and the distributions of temperature and front got with the method are more likely to exhibit the real condition. |
| Key words: Data assimilation, Coastal ocean survey, Sea area off the Changjiang estuary, Synoptic |
