Comment on essd-2020-361

doi:10.5194/essd-2020-361-rc2

Abstract

Long-term climate data and high-quality baseline climatology surface with high resolution are highly essential to multiple fields in climatological, ecological, hydrological, and environmental sciences. Here, we created a brand-new baseline climatology surface (ChinaClim_baseline) and developed a 1 km monthly precipitation and temperatures dataset in China during 1952–2019 (ChinaClim_timeseries). Thin plate spline (TPS) algorithm in each month with different model formulations by accounting for satellite-driven products, was used to generate ChinaClim_baseline and monthly climate anomaly surface. Meanwhile, climatologically aided interpolation (CAI) was used to superimpose monthly anomaly surface with ChinaClim_baseline to generate ChinaClim_timeseries. Our results showed that ChinaClim_baseline exhibited very high performance. For precipitation estimation, the value of all R2 was over 0.860, and the values of RMSEs and MAEs were 8.149 mm~21.959 mm and 2.787~14.125 mm, respectively. Temperature elements had an average R2 of 0.967~0.992, an average MAEs of 0.321~0.785 Â°C, and an average RMSEs between 0.485 and 1.233 Â°C for all months. ChinaClim_baseline performed much better than WorldClim2 and CHELSA and there were many spatial discrepancies captured among those surfaces, especially in summer months and the regions with low-density weather stations in temperate continental and high cold Tibetan Plateau. For ChinaClim_timeseries, precipitation had an average R2 of 0.699~0.923, an average RMSE between 7.449 mm and 56.756 mm, and an average of MAE of 4.263~40.271 mm for all months. Temperature elements had an average R2 of 0.936~0.985, an average RMSE between 0.807 Â°C and 1.766 Â°C, and an average MAE of 0.548~1.236 Â°C for all months. Compared with Peng's climate surface and CHELSAcruts, R2 increased by approximately 6 %, RMSE and MAE decreased by approximately 15 % for precipitation; R2 of temperatures had no obviously changes, but RMSE and MAE decreased by 8.37~34.02 %. The results showed that the interannual variations of ChinaClim_timeseries performed much better than other datasets, thanks to the help of ChinaClim_baseline and satellite-driven products. However, ChinaClim_baseline did not significantly improve the accuracy of precipitation estimation, but it greatly improved the accuracy of temperature estimation; the satellite-driven TRMM3B43 anomaly greatly improve the accuracy of precipitation estimation after 1998, while the LST anomaly did not effectively improve the accuracy of temperature estimation. ChinaClim_baseline can be used as an excellent baseline climatology surface for obtaining high-quality and long-term climate datasets from past to future. In the meantime, ChinaClim_timeseries of 1 km spatial resolution based on ChinaClim_baseline, is very suitable for investigating the spatial-temporal climate changes and their impacts on eco-environmental systems in China. Here, ChinaClim_baseline is available at https://doi.org/10.5281/zenodo.4287824 (Gong, 2020a), ChinaClim_timeseries of precipitation is available at https://doi.org/10.5281/zenodo.4288388 (Gong, 2020b), ChinaClim_timeseries of maximum temperature is available at https://doi.org/10.5281/zenodo.4288390 (Gong, 2020c) and ChinaClim_timeseries of minimum temperature is available at https://doi.org/10.5281/zenodo.4288392 (Gong, 2020d).

Highlights

For the estimates of long-term monthly climate surface, recent studies have shown that climatologically aided interpolation (CAI) employing the temporal anomaly surface and an accurate baseline climatology surface with high resolution, is well suited for producing long-term climate datasets than direct interpolation using original weather stations
Our results showed that all R2 were very high among three datasets, but compared to Peng’s and CHELSAcruts, our root mean square error (RMSE) decreased by 10.17 % and 19.14 % for maximum temperature, and by
There are a number of baseline climatology surface products for global land surface (Hijmans et al, 2005; Karger et al, 2017; New et al, 1999; New et al, 2002; Fick et al, 2017), while few weather stations are employed to generate these surfaces in China, which might result in insufficient accuracy of these surfaces, and further affect the availability of long-term climate datasets with these surfaces as input

Summary

Introduction

Long-term information on climatic conditions is pivotal for understanding global changes including atmospheric movements, vegetation dynamics, soil moisture, and other related scientific and application fields which are conducted at a resolution of ~1 km (Chaney et al, 2014; Gao et al., 2018; Hijmans et al, 2005; Karger et al, 2017; Liu et al, 2016; New et al, 2002; Pfister et al, 2020; Wagner and Wolfgang, 2003). A variety body of work was motivated to obtain high resolution gridded climate data with spatial interpolation methods and statistical downscaling (Li and Shao, 2010; Wu and Li, 2013; Hartkamp et al, 1999; Boer et al, 2001). For the estimates of long-term monthly climate surface, recent studies have shown that climatologically aided interpolation (CAI) employing the temporal anomaly surface and an accurate baseline climatology surface with high resolution, is well suited for producing long-term climate datasets than direct interpolation using original weather stations The quality of monthly climate surface, generated by CAI method, was thought to be determined by the baseline climatology surface (Gao et al, 2018; Peng et al, 2019).

Objectives

Results

Conclusion