A comparison of global surface temperature variability, extremes and warming trend using reanalysis datasets and CMST‐Interim

Yang Yang,Qingxiang Li,Zhaoyang Song,Wenbin Sun,Wenjie Dong
DOI: https://doi.org/10.1002/joc.7551
2022-02-10
International Journal of Climatology
Abstract:Reanalysis data are widely used to investigate long‐term surface temperature changes due to insufficient spatial coverage of observational data. However, because of the limitations of data assimilation and model performance in the reanalysis datasets, it is essential to evaluate the quality of the reanalysis datasets. Based on the newly released version of China global Merged Surface Temperature dataset, Interim version (CMST‐Interim), the performance of five reanalysis datasets (ERA5, NCEP/NCAR R1, JRA‐55, CERA‐20C, and 20CRv3), covering more than 50 years, is compared in terms of long‐term variation bias, warming trend, and the consistency of the extreme temperature years from the period 1958–2010. The results reflect that the above reanalysis datasets have reasonable representativeness of global temperature change. ERA5 and 20CRv3 perform better than the other reanalysis datasets, with relatively small deviation, their warming trends are closer to the observation results, the appearance of extreme temperature years are also more consistent with the observed, and CERA‐20C is the next. JRA‐55 is slightly better in detecting the extremely cold years, while NCEP/NCAR R1 is slightly worse in biases but performs well in high‐temperature years reproduction. The annual mean of 1901‐2018 for surface temperature anomalies from 5 reanalysis data sets and CMST‐Interim over the globe (north of 60°S) (upper panel); Root Mean Squared Errors (RMSEs) for annual surface temperature anomalies in (a) ERA5, (b) NCEP1, (c) CERA‐20C, (d) 20CRv3 and (e) JRA‐55 over 1958‐2010 relative to CMST‐Interim (Unit: ℃). All anomalies are calculated relative to the 1961‐1990 reference period (lower panel).
meteorology & atmospheric sciences
What problem does this paper attempt to address?