Long-term Weekly Average Field Of Sea Surface Temperature And Used In Northwest Pacific

Sea surface temperature plays an important part in the marine ecosystem. Varies the seasurface temperature is a part of the key issues people concerned. To better show the change ofsea surface temperature and improve the accuracy of marine fishery prediction, the long-termweekly average sea surface temperature field which can be a standard to characterization thevariation of SST is needed urgently to be built. Furthermore, we use the long-term weeklyaverage sea surface temperature field to indicate the change in sea surface temperature in theNorthwest Pacific. The Northwest Pacific is the one of the most productive fisheries in the wordbecause of the confluence of Kuroshio Current and Oyashio there. It is necessary formeteorology and fisheries research to analyze the change of sea surface temperature.The long-term datasets are built mainly for climatology services so far, such as ICOADS(International Comprehensive Ocean-Atmosphere Data Set)has provided the long-termmonthly average sea surface temperature field with2°×2°. Because the scale of fishinggrounds is smaller, we need the higher resolution of long-term monthly average sea surfacetemperature field in the research of fisheries.The AVHRR sensors on NOAA satellite have obtained the information of ocean for morethan30years, and have accumulated amount of data. We can be a long sequence of sea surfacetemperature data set from the satellite data.The sea surface temperature (SST) data were obtained from NOAA satellite AVHRR sensorduring the years of1982-2012in this paper. Then, the data in the same week in the31years weretreated with arithmetic average. Finally, the long-term52weeks weekly average sea surfacetemperature field was gotten, which was on a0.25o x0.25o global grid. In addition, wechoose the Northwest Pacific (125-180°E,25-55°N）as the main research region, which is animportant fishing ground for our country. The sea surface temperature anomaly value was takenfor analysis of the variation of sea surface temperature from1982to2012based on the long-termweekly average sea surface temperature field. The sea surface temperature anomaly wasanalyzed by Empirical Orthogonal Function, we got three highest change modes. The mainvariation of SST in the Northwest Pacific was analyzed in detail. The concludes as follows:1) The sea surface temperature (SST) data from NOAA satellite AVHRR sensor during the years of1982-2012were applied in this paper. The process contains the data handling, qualitycontrol, and the interpolation calculation by inverse distance weighting method and so on. IDLprogramming language was employed in all process. The weekly average sea surfacetemperature was calculated year by year. Then, the data in the same week in the31years weredealt with arithmetic average. Finally, the long-term weekly average sea surface temperaturefield was gotten, which was on a0.25o x0.25o global grid. Compared to the long-term monthlyaverage sea surface temperature field from ICOADS, the resolution of the long-term weeklyaverage sea surface temperature field we built is higher in spatial and temporal.2) There is a distinctive of the seasonable variations of sea surface temperature in thenorthwest Pacific. The temperature difference between summer and winter could be13℃in.There is a different magnitude of sea surface temperature changes in different regions. The Seaof Japan is the most violent region. The temperature difference can be17.8℃. The least violentregion is happening around of55°N,180°E.3) Variations of SST are cyclical rise in the past31years in the Northwest Pacific. Fornearly31years of the transition period in1998, the yearly average SSTA was basically under0℃, and after the year of1998, the yearly average SSTA was higher than0℃. The highestannual average SSTA happened in2008, reached0.5℃, then mean the SST reached its peak in2010.4) The Empirical Orthogonal Function (EOF) were used to analyze the variability of SSTAof the Northwest Pacific, and get11main eigenvectors explaining over50%of total variance,the first three eigenvectors are22.16%,9.59%,5.25. The first time coefficient has a periodicincreasing trend; with spatial distribution, the variation of SST in the northern region is moreviolent than the southern region. The latitudinal distribution of the second mode is coincide withthe North Pacific westerly drift region. The index time series of the third mode and the PDOindex coincide to each other well.