A New Globally Reconstructed Sea Surface Temperature Analysis Dataset since 1900

A new globally reconstructed sea surface temperature(SST)analysis dataset developed by the China Meteorological Administration(CMA-SST),available on 2°×2°and monthly resolutions since 1900,is described and assessed in this study.The dataset has been constructed from a newly developed integrated dataset with denser and wider sampling of in situ SST observations and follows similar analysis techniques to the Extended Reconstructed SST,version 5(ERSST.v5).Assessments show that the larger observation quantity of the input data source is beneficial to making the reconstructed SSTs more realistic than those reconstructed with ICOADS 3.0+GTS(International Comprehensive Ocean-Atmosphere Dataset 3.0 and Global Telecommunication System),especially in China’s offshore sea area.Besides,a specific parameter for bias correction has been upgraded to be self-adaptive to the input data source,and serves as a mediator to improve the accuracy of the reconstructed SSTs.Generally,the reconstructed CMA-SST dataset is comparable to currently congeneric products.Its biases are similar to those of ERSST.v5,the Centennial Observation-Based Estimates of SST version 2(COBE-SST2),the Hadley Centre Sea Ice and SST dataset version 2(Had ISST2),and the Hadley Centre SST dataset version 3(Had SST3);and more specifically,they are closest to ERSST.v5 and lower than Had ISST2 and Had SST3 at high latitudes of the Southern Hemisphere where in situ observations are limited.Moreover,its temporal characteristics,such as the year-to-year variations of globally averaged SST anomalies and time series of the Nino-3.4,Atlantic multidecadal oscillation,and Pacific decadal oscillation indices are also a good match to those of congeneric products.Although the warming rates of CMA-SST are a little higher in many regions over the periods 1900-2019 and 1950-2019,they are found to be acceptable and within the quantified uncertainties of ERSST.v5.However,there are noticeable differences in the strength and stability of spatial standard deviations among the various datasets,as well as low correlations between CMA-SST and the other products around 60°S where in situ sampling is very limited.These aspects necessitate further investigation and improvement of CMA-SST.