Preprints
https://doi.org/10.5194/hess-2020-633
https://doi.org/10.5194/hess-2020-633

  10 Dec 2020

10 Dec 2020

Review status: this preprint was under review for the journal HESS. A revision for further review has not been submitted.

Rainfall erosivity estimation using gridded daily precipitation datasets

Maoqing Wang1, Shuiqing Yin1, Tianyu Yue1, Bofu Yu2, and Wenting Wang3,1 Maoqing Wang et al.
  • 1State Key Laboratory of Earth Surface Processes and Resource Ecology, Faculty of Geographical Science, Beijing Normal University, Beijing, 100875, China
  • 2Australian Rivers Institute, School of Engineering and Built Environment, Griffith University, Nathan, Qld 4111, Australia
  • 3Zhuhai Branch of State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University at Zhuhai, 519087, China

Abstract. Rainfall erosivity is one of the most important factors incorporated into the empirical soil erosion models USLE (Universal Soil Loss Equation) and RUSLE (Revised Universal Soil Loss Equation). Gridded precipitation datasets have been widely used in the estimation of rainfall erosivity, whereas biases due to scale differences between gridded data and gauge data have been ignored. Based on daily precipitation observations from over 2000 stations in China, as well as four widely used gauge-based gridded daily precipitation datasets, CPC, GPCC, CN05.1 and NMIC, this study compared the probability density functions (PDFs) of the gridded and gauge datasets using the skill score method, quantified the bias of rainfall erosivity (including the R-factor and 1-in-10-year event rainfall erosivity) estimated using the gridded daily precipitation datasets from that estimated using the gauge daily precipitation dataset based on the area reduction factor (ARF) method, and established correction factors for rainfall erosivity maps generated from gridded datasets. The results showed that the gridded daily data reduced the frequency of no-rain days and the intensity of heavy precipitation. In the eastern part of China, the grid-estimated R-factor values were underestimated by 15–40 % compared with the gauge-estimated values, and the grid-estimated 1-in-10-year event rainfall erosivity values were underestimated by 25–50 %, whereas in the western part of China, noticeable random errors were introduced. The lower probability and intensity of the daily precipitation larger than the 90th percentile in the gridded datasets were mainly responsible for the underestimation. CN05.1 was the most-recommended among the four datasets, as it had the lowest mean relative error (MRE), and the accuracy was higher for the eastern part of China than for the western part of China. The MREs were 16.1 % and 25.1 % for the R-factor after applying correction factors of 1.708 and 1.010, respectively, for the eastern and western part of China. The 1-in-10-year event erosivity had larger correction factors and MREs than did the R-factor, with the MREs being 22.1 % and 27.2 % after applying correction factors of 1.959 and 1.880, respectively, for the eastern and western part of China. This study pointed out that in the applications of gridded precipitation datasets, the empirical models established based on gauge precipitation data should not be used directly for the gridded data, or a bias correction process needed to be considered for the model outputs.

Maoqing Wang et al.

 
Status: closed (peer review stopped)
Status: closed (peer review stopped)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement
 
Status: closed (peer review stopped)
Status: closed (peer review stopped)
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Maoqing Wang et al.

Maoqing Wang et al.

Viewed

Total article views: 476 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
329 142 5 476 5 5
  • HTML: 329
  • PDF: 142
  • XML: 5
  • Total: 476
  • BibTeX: 5
  • EndNote: 5
Views and downloads (calculated since 10 Dec 2020)
Cumulative views and downloads (calculated since 10 Dec 2020)

Viewed (geographical distribution)

Total article views: 371 (including HTML, PDF, and XML) Thereof 370 with geography defined and 1 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 21 Oct 2021
Download
Short summary
This study quantified the bias of rainfall erosivity estimated from gridded precipitation data, and the results showed the grid-estimated mean annual rainfall erosivity were underestimated by 15–40 % in the eastern China. The scale difference between gridded data and gauge data was the main cause. In application, the empirical models established based on gauge data should not be used directly for gridded data, or a bias correction process needed to be considered for the model outputs.