摘要
基于自然资源部浮标数据,通过分析均值差、均方根误差、相关系数和标准差偏差4个统计量,检验了2018年7月1日至8月6日全时段及该时段内3个台风(1808号台风"玛利亚"、1810号台风"安比"、1812号台风"云雀")过境期间,3套海表面温度(Sea Surface Temperature,SST)再分析资料(OISST、OSTIA SST、RTG SST)在中国近海区域的可靠性。对比结果表明,在全时段内,3套SST再分析资料都能在一定程度上反映中国近海SST的基本状况,其中OSTIA SST资料同浮标实测SST数据的均值差为0.12℃、相关系数为0.94,均优于OISST资料(均值差为–0.85℃、相关系数为0.90)和RTG SST资料(均值差为–0.17℃、相关系数为0.86)。通过对比单个浮标数据发现,相较约80%的MF浮标实测SST数据,OSTIA SST资料都显著优于RTG SST资料和OISST资料,具有较高的可信度。在台风过境期间,较之RTG SST资料和OISST资料,OSTIA SST资料同大部分浮标实测数据的均值差绝对值及均方根误差更小、相关系数更大,表明在高海况条件下,OSTIA SST资料能更真实地反映中国近海SST的基本状况。
Based on the buoys data over the China offshore area during July 1, 2018 to August 6, 2018 collected by the Ministry of Natural Resources, three SST reanalysis products(OISST, OSTIA SST, RTG SST) were validated and compared through analyzing the mean bias, root mean square error(RMSE), correlation coefficient and standard deviation bias. The comparison results show that in the whole time, three SST reanalysis products we used arecomparable to the buoys data. The mean bias and correlation coefficient between OSTIA SST dataset and buoy SST data is 0.12℃ and 0.94, better than OISST dataset(-0.85℃, 0.90) and RTG SST dataset(-0.17℃, 0.86). Compared with 80% buoys, the reliability of OSTIA SST dataset is higher than OISST dataset and RTG SST dataset significantly. During typhoon transit periods, absolute values of mean bias and RMSE(correlation coefficient)between OSTIA SST dataset and buoy SST data are smaller(larger) than RTG SST dataset and OISST dataset,which means that OSTIA SST dataset can capture the basic characteristics of SST over the China offshore area more accurately in severe sea conditions.
作者
王晨琦
李响
张蕴斐
祖子清
张润宇
Wang Chenqi;Li Xiang;Zhang Yunfei;Zu Ziqing;Zhang Runyu(Key Laboratory of Marine Hazards Forecasting,Ministry of Natural Resources,National Marine Environmental Forecasting Center,Beijing 100081,China;National Marine Environmental Forecasting Center,Beijing 100081,China)
出处
《海洋学报》
CAS
CSCD
北大核心
2020年第3期118-128,共11页
基金
国家重点研发计划(2017YFA0604203)
全球变化与海气相互作用专项(GASI-IPOVAI-06)
国家自然科学基金(41806003)。
