Advanced Search
Zhongda YU, Guangqing ZHOU, Yunjun ZHOU. Evaluation and Analysis of Soil Temperature Data over Middle and High Latitudes of East Asia[J]. Chinese Journal of Atmospheric Sciences, 2017, 41(1): 147-166. DOI: 10.3878/j.issn.1006-9895.1603.15300
Citation: Zhongda YU, Guangqing ZHOU, Yunjun ZHOU. Evaluation and Analysis of Soil Temperature Data over Middle and High Latitudes of East Asia[J]. Chinese Journal of Atmospheric Sciences, 2017, 41(1): 147-166. DOI: 10.3878/j.issn.1006-9895.1603.15300

Evaluation and Analysis of Soil Temperature Data over Middle and High Latitudes of East Asia

  • This study investigates the qualities of four soil temperature reanalysis products over middle and high latitudes of East Asia based on a set of Russian historical soil temperature data (RHSTD). The four reanalysis products are the ERA-Interim re-analysis, ERA-Interim/Land (i.e. ERA-Land), and MERRA-Land generated by running land surface models offline, and a 20th century reanalysis (i.e. the NOAA-CIRES 20CR). More emphases are put on the spring and summer, and main results are as follows. Large seasonal fluctuations of 0-2 m soil temperature around 0℃ are found in the RHSTD, but the soil temperature below 2 m stays relatively stable. Furthermore, the soil below 2 m is permanently frozen to the north of 60°N. The four reanalysis products well capture these characters. Regardless of the spring, summer or annual mean, the soil temperature climatology shown in the four reanalysis products is always "warm in the south and cold in the north", while the spatial pattern of the ERA-Land is the closest to observations. In terms of seasonal cycle, the ERA-Land at this region has a better performance in reflecting the soil freezing and thawing processes and the seasonal evolution of soil temperature. Situations are more complex regarding the interannual variability of soil temperature, i.e. standard deviation, because errors between these re-analysis products and observations vary with season and soil layer greatly. As for the interannual change, the correlation between each reanalysis product and observations is better in the summer than in the spring, and the correlation in the upper layer is better than that in the deep layer. Meanwhile, the ERA-Land is the most representative of the four re-analysis products. It better reproduces the interannual changes in observed soil temperature than the other three products.
  • loading

Catalog

    Turn off MathJax
    Article Contents

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return