Journal cover Journal topic
Natural Hazards and Earth System Sciences An interactive open-access journal of the European Geosciences Union
Journal topic

Journal metrics

IF value: 3.102
IF3.102
IF 5-year value: 3.284
IF 5-year
3.284
CiteScore value: 5.1
CiteScore
5.1
SNIP value: 1.37
SNIP1.37
IPP value: 3.21
IPP3.21
SJR value: 1.005
SJR1.005
Scimago H <br class='widget-line-break'>index value: 90
Scimago H
index
90
h5-index value: 42
h5-index42
Volume 6, issue 3
Nat. Hazards Earth Syst. Sci., 6, 323–342, 2006
https://doi.org/10.5194/nhess-6-323-2006
© Author(s) 2006. This work is licensed under
the Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License.

Special issue: Advances in radar, multi-sensor and hydrological modelling...

Nat. Hazards Earth Syst. Sci., 6, 323–342, 2006
https://doi.org/10.5194/nhess-6-323-2006
© Author(s) 2006. This work is licensed under
the Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License.

  10 May 2006

10 May 2006

Improved radar rainfall estimation at ground level

S. M. Wesson and G. G. S. Pegram S. M. Wesson and G. G. S. Pegram
  • Civil Engineering, University of KwaZulu-Natal, Howard College Campus, Durban, South Africa

Abstract. A technique has been developed to provide an estimate of the rainfall reaching the earth's surface by extrapolating radar data contained aloft to ground level, simultaneously estimating unknown data in the radar volume scan. The technique has been developed so as to be computationally fast, to work in real time and comprises the following steps. A rainfall classification algorithm is applied to separate the rainfall into two separate types: convective and stratiform rainfall. Climatological semivariograms based on the rainfall type are then defined and justified by testing, which result in a fast and effective means of determining the semivariogram parameters anywhere in the radar volume scan. Then, extrapolations to ground level are computed by utilising 3-D Universal and Ordinary Cascade Kriging; computational efficiency and stability in Kriging are ensured by using a nearest neighbours approach and a Singular Value Decomposition (SVD) matrix rank reduction technique. To validate the proposed technique, a statistical comparison between the temporally accumulated radar estimates and the Block Kriged raingauge estimates is carried out over matching areas, for selected rainfall events, to determine the quality of the rainfall estimates at ground level.

Publications Copernicus
Download
Citation
Altmetrics