An article published by the team in Earth System Science Data on September 29, 2025, details the update to C-LSAT 2.1 and the development process of the new high-resolution Land Surface Air Temperature (LSAT) and Diurnal Temperature Range (DTR) datasets.

    The primary focus of this paper is the major update to the C-LSAT 2.0 station data. By integrating multiple data sources at global, regional, and national levels, we added data from nearly 3000 new meteorological observation stations and applied rigorous quality control and homogenization to all data. The updated C-LSAT 2.1 station data shows a significant improvement in coverage. The number of stations for the monthly mean temperature (Tavg) increased from 15936 to 25085, monthly mean maximum temperature (Tmax) increased from 13648 to 25086, and monthly mean minimum temperature (Tmin) increased from 13629 to 25083. The increase in the number of stations is most prominent after the 1970s, which greatly expands the spatial coverage of the data, thereby improving the accuracy of the gridded data and reducing uncertainty. The updated C-LSAT 2.1 station data was then used for gridding and reconstruction (Fig. 1).

Fig. 1 Updates to the C-LSAT 2.1 station data (include 1961-1990).

    Based on the updated C-LSAT 2.1 station data, we employed a technical framework combining a "thin-plate spline (for climatology fields) and adjusted inverse distance weighting method (for anomaly fields)" to develop high-resolution datasets at 0.5° × 0.5° resolution: C-LSAT HRv1 (Land Surface Air Temperature) and C-LDTR HRv1 (Diurnal Temperature Range), spanning from January 1901 to December 2023. In the temporal dimension, from 1901 to 2023, the global LSAT shows a significant warming trend at a rate of 0.132 ± 0.015 °C per decade, with this warming accelerating significantly after the 1970s. In contrast, the global DTR shows an overall decreasing trend during the same period, with a change rate of -0.031 ± 0.006 °C per decade. The change in DTR is periodic, primarily decreasing from the 1950s to the 1970s, with a slight recovery after 2000 (Fig. 2). In terms of spatial distribution, the trend in land surface air temperature shows widespread and continuous warming across the globe. The fastest warming rates are observed in regions such as northern North America, eastern South America, Eastern Europe, and East Asia. In contrast, the change in diurnal temperature range does not exhibit a clear, globally consistent spatial pattern. Although the overall trend is downward, it varies across different regions (Fig .3).

Fig. 2 The anomalies of LSAT (left) and DTR (right) from 1901 to 2023 (a: Global, b: Northern Hemisphere, c: Southern Hemisphere)..

Fig. 3 Spatial distribution of the rate of change for LSAT (top) and DTR (bottom) from 1901 to 2023.

Paper Information:

Wei,S., Li, Q., Xu, Q., Li, Z., Zhang, H., and Lin, J.: Updates to C-LSAT 2.1 andthe development of high-resolution land surface air temperature and diurnaltemperature range datasets, Earth Syst. Sci. Data, 17, 4985–5005,https://doi.org/10.5194/essd-17-4985-2025, 2025.