Preprints
https://doi.org/10.5194/hess-2018-36
https://doi.org/10.5194/hess-2018-36
15 Feb 2018
 | 15 Feb 2018
Status: this discussion paper is a preprint. It has been under review for the journal Hydrology and Earth System Sciences (HESS). The manuscript was not accepted for further review after discussion.

Exploring the Long-Term Reanalysis of Precipitation and the Contribution of Bias Correction to the Reduction of Uncertainty over South Korea: A Composite Gamma-Pareto Distribution Approach to the Bias Correction

Dong-Ik Kim, Hyun-Han Kwon, and Dawei Han

Abstract. The long-term record of precipitation data plays an important role in climate impact studies. The local observation is often considered to be the truth in regional-scale analyses, but the long-term meteorological record for a given catchment is very limited. Recently, ERA-20c, a century-long reanalysis of the data has been published by the European Centre for Medium-Range Weather Forecasts (ECMWF), which includes daily precipitation over the whole 20th century with high spatial resolution of 0.125° × 0.125°. Preliminary studies have already indicated that the ERA-20c can reproduce the mean reasonably well, but rainfall intensity was underestimated and wet-day frequency was overestimated. The primary focus of this study was to expand our sample size significantly for extreme rainfall analysis. Thus, we first adopted a relatively simple approach to adjust the frequency of wet-days by imposing an optimal lower threshold. We found that the systematic errors are fairly well captured by the conventional quantile mapping method with a gamma distribution, but the extremes in daily precipitation are still somewhat underestimated. In such a context, we introduced a quantile mapping approach based on a composite distribution of a generalized Pareto distribution for the upper tail (e.g. 95th and 99th percentile), and a gamma distribution for the interior part of the distribution. The proposed composite distributions provide a significant reduction of the biases compared with that of the conventional method for the extremes. We suggest a new interpolation method based on the parameter contour map for bias correction in ungauged catchments. The strength of this approach is that one can easily produce the bias-corrected daily precipitation in ungauged or poorly gauged catchments. A comparison of the corrected datasets using contour maps shows that the proposed modelling scheme can reliably reduce the systematic bias at a grid point that is not used in the process of parameter estimation. In particular, the contour map with the 99th percentile shows a more accurate representation of the observed daily rainfall than other combinations. The findings in this study suggest that the proposed approach can provide a useful alternative to readers who consider the bias correction of a regional-scale modelled data with a limited network of rain gauges. Although the study has been carried out in South Korea, the methodology has its potential to be applied in other parts of the world.

Dong-Ik Kim, Hyun-Han Kwon, and Dawei Han
Dong-Ik Kim, Hyun-Han Kwon, and Dawei Han
Dong-Ik Kim, Hyun-Han Kwon, and Dawei Han

Viewed

Total article views: 1,893 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
1,457 394 42 1,893 61 55
  • HTML: 1,457
  • PDF: 394
  • XML: 42
  • Total: 1,893
  • BibTeX: 61
  • EndNote: 55
Views and downloads (calculated since 15 Feb 2018)
Cumulative views and downloads (calculated since 15 Feb 2018)

Viewed (geographical distribution)

Total article views: 1,664 (including HTML, PDF, and XML) Thereof 1,636 with geography defined and 28 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 27 Feb 2024
Download
Short summary
This study introduces a new QM approach based on a composite distribution of a generalized Pareto distribution for the upper tail and a gamma distribution for the interior part of the distribution. The proposed composite distributions provide a significant reduction of the biases compared with that of the conventional method for the extremes. The proposed approach can provide a useful alternative for the bias correction of a regional-scale modeled data with a limited network of rain gauges.