Preprints
https://doi.org/10.5194/hess-2023-122
https://doi.org/10.5194/hess-2023-122
12 Jun 2023
 | 12 Jun 2023
Status: this preprint is currently under review for the journal HESS.

Merging with crowdsourced rain gauge data improves pan-European radar precipitation estimates

Aart Overeem, Hidde Leijnse, Gerard van der Schrier, Else van den Besselaar, Irene Garcia-Marti, and Lotte Wilhelmina de Vos

Abstract. Ground-based radar precipitation products typically need adjustment with rain gauge accumulations to achieve a reasonable accuracy. This is certainly the case for the pan-European radar precipitation products. The density of (near) real-time rain gauge accumulations from official networks is often relatively low. Crowdsourced rain gauge networks have a much higher density than conventional ones and are a potentially interesting (complementary) source to merge with radar precipitation accumulations. Here, a 1-year personal weather station (PWS) rain gauge dataset of ~5 min accumulations is obtained from the private company Netatmo over the period September 1, 2019–August 31, 2020, which is subjected to quality control using neighbouring PWSs and on 1-h accumulations using unadjusted radar data. The PWS 1-h gauge accumulations are employed to spatially adjust OPERA radar accumulations, covering 78 % of geographical Europe. The performance of the merged dataset is evaluated against daily and disaggregated 1-h gauge accumulations from weather stations in the European Climate Assessment & Dataset (ECA&D). Results are contrasted to those from an unadjusted OPERA-based radar dataset and from EURADCLIM. The severe average underestimation for daily precipitation of ~28 % from the unadjusted radar dataset diminishes to ~3 % for the merged radar-PWS dataset. A station-based spatial verification shows that the relative bias in 1-h precipitation is still quite variable and suggests stronger underestimations for colder climates. A dedicated evaluation with scatter density plots reveals that the performance is indeed less good for lower temperatures, which points to limitations in observing solid precipitation by PWS gauges. The outcome of this study confirms the potential of crowdsourcing to improve radar precipitation products in (near) real-time.

Aart Overeem et al.

Status: open (until 12 Oct 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Aart Overeem et al.

Data sets

EURADCLIM: The European climatological gauge-adjusted radar precipitation dataset (1-h accumulations) Aart Overeem, Else van den Besselaar, Gerard van der Schrier, Jan Fokke Meirink, Emiel van der Plas, Hidde Leijnse https://doi.org/10.21944/7ypj-wn68

EURADCLIM: The European climatological gauge-adjusted radar precipitation dataset (24-h accumulations) Aart Overeem, Else van den Besselaar, Gerard van der Schrier, Jan Fokke Meirink, Emiel van der Plas, Hidde Leijnse https://doi.org/10.21944/1a54-gg96

Model code and software

EURADCLIM-tools Aart Overeem https://doi.org/10.5281/zenodo.7473816

Aart Overeem et al.

Viewed

Total article views: 443 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
339 90 14 443 10 8
  • HTML: 339
  • PDF: 90
  • XML: 14
  • Total: 443
  • BibTeX: 10
  • EndNote: 8
Views and downloads (calculated since 12 Jun 2023)
Cumulative views and downloads (calculated since 12 Jun 2023)

Viewed (geographical distribution)

Total article views: 419 (including HTML, PDF, and XML) Thereof 419 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 26 Sep 2023
Download
Short summary
Ground-based radar precipitation products typically need adjustment with rain gauge accumulations to achieve a reasonable accuracy. Crowdsourced rain gauge networks have a much higher density than conventional ones. Here, a 1-year personal weather station (PWS) gauge dataset is obtained. After quality control, the 1-h PWS gauge accumulations are merged with pan-European radar accumulations. The potential of crowdsourcing to improve radar precipitation products in (near) real-time is confirmed.