The Use of Radar Data to Derive Areal Reduction Factors for South Africa

dc.contributor.advisorDu Plessis, J. A.en_ZA
dc.contributor.authorThomas, Jason Todden_ZA
dc.contributor.otherStellenbosch University. Faculty of Engineering. Dept. of Civil Engineering.en_ZA
dc.descriptionThesis (MEng)--Stellenbosch University, 2019.en_ZA
dc.description.abstractENGLISH ABSTRACT: An areal reduction factor (ARF) can be defined as a factor that is applied to point rainfall depths to convert these depths to an average rainfall over a specific catchment area. The concept of ARFs provide a powerful mechanism for the analyses of the spatial variability of various hydrological processes. However, a plethora of methods used to derive ARFs are dependent rainfall stations. With the decline in population of reliable rainfall stations, radar data has been reviewed as an alternative for applications in hydrology and consequently ARFs. Radar data appears to be more efficient than using rain gauge networks as radar data is able to capture the internal and spatial distribution of a rainfall event. This research made use of using the storm centred approach based on its incorporation of high spatial and temporal resolution, with radar imagery. Meteorological Data Volume (MDV) files were converted to network common data form (netcdf) files, for compatibility in ArcGIS. ArcGIS was used to delineate single cellular storms, find the maximum point rainfall and calculate the isohyetal rainfall, to produce ARFs. Precipitation Arrays were abstracted from the netcdf files, for the validation of ArcGIS storm outputs: storm area and isohyetal areal rainfall. The ArcGIS storm outputs, storm area and isohyetal areal rainfall, had significantly high correlation with Precipitation Array outputs, confirming the results thereof suitable for use in deriving ARFs. Furthermore, the ARFs that were obtained using Precipitation Arrays and ArcGIS had high correlation and significance. The influence of storm area on the radar derived ARFs was determined. It was found that storm area had a significant effect on the ARFs; stronger influence on storm durations of 3 hours, and weaker influence on storm durations of 24 hours. Satisfactory correlation was found between the maximum point rainfall and the resulting ARF. Moreover, a strong inverse proportional relationship was found to exist between the maximum point rainfall intensity and the resulting ARF, for storms of different durations. For various rainfall processes, convective rainfall regions produced higher ARFs, than the ARFs produced in frontal rainfall regions. A comparison between the radar derived ARFs and the ARFs currently implemented in South Africa was carried out. The radar derived ARFs were lower than ARFs in South Africa for storm areas less than 200 km2, for smaller catchments. For larger catchments, it was found that radar derived ARFs were lower than ARFs in South Africa for storm durations of more than 1 hour. The radar derived ARFs for the 1 hour storm events were found to be overestimated. In general, the radar derived ARFs were more conservative. Finally, using radar data for the derivation of ARF has exposed the high potential of its use in hydrology.en_ZA
dc.format.extent144 pages : illustrationsen_ZA
dc.publisherStellenbosch : Stellenbosch Universityen_ZA
dc.subjectAreal Reduction Factoren_ZA
dc.subjectRadar -- Dataen_ZA
dc.titleThe Use of Radar Data to Derive Areal Reduction Factors for South Africaen_ZA
dc.rights.holderStellenbosch Universityen_ZA

Files in this item


This item appears in the following Collection(s)