Rain gauge network optimization in a tropical area towards efficient hydrological data acquisition / Mohd Zaharifudin Muhamad Ali

Mohd Zaharifudin , Muhamad Ali (2019) Rain gauge network optimization in a tropical area towards efficient hydrological data acquisition / Mohd Zaharifudin Muhamad Ali. PhD thesis, Universiti Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (199Kb)
    [img] PDF (Thesis PhD)
    Download (2242Kb)


      An adequate and reliable rain gauge network is essential for observing rainfall data in hydrology and water resource applications. For this purpose, rain gauge stations are installed in the catchment area of the river that forms a rain gauge network. Normally, a rain gauge network will be developed in accordance with the hydrological purpose and evaluated to extend to the network size in order to increase data accuracy. The increasing number of rain gauge network and augmentation of the existing rain gauge networks without proper planning and design has resulted in the high density of rain gauge stations compared to the recommendation made by the World Meteorological Organization (WMO). This resulted in increasing maintenance costs, but at the same time, creates the possibility of redundancy of stations within the catchment area. Due to this factor, it is the aim of this study to review the rain gauge network in a specific catchment to establish the optimal number of stations so that efficient rainfall data acquisition can be obtained. Two new optimization approaches have been developed in this study for the rain gauge network optimization and to prioritize the rain gauge stations, first by coupling the cross-validation technique with the geostatistical method (CV-Geo), and second, using the modified Particle Swarm Optimization (MPSO) technique. The spatial interpolation error of the spatial rainfall distribution, measured as the Root Mean Square Error (Erms) optimization criterion, is applied to a rain gauge network in a tropical urban area. The total daily rainfall data from the 55 rain gauge stations were used to perform the optimization process for seven flood events. The optimization aimed to reduce the number of rain gauge stations in the existing network that could be hypothetically redundant. By using the two new methods, CV-Geo and MPSO, the number of stations in the existing rain gauge network could be optimized based on the lowest Erms value of spatial interpolation error. The optimized rain gauge network exhibited a better semivariogram structure, especially in terms of nugget value that has been drastically improved. However, MPSO had shown a slightly better nugget value since it has recorded the lowest value of nugget. The rain gauge stations were prioritized based on their importance in the network. Four stations, namely T02, N03, N06, and N21 were considered ineffective and could, therefore, be relocated within the study area or eliminated from the existing network. A preliminary evaluation of the optimized network without the four stations showed satisfactory results in flood hydrograph simulation using a lump hydrologic model. Three out of four flood hydrograph simulations have yielded the NSE, r, and R2 values more than 0.75, which have indicated that the optimized network is efficient enough to produce rainfall data to simulate a flood hydrograph. The optimized rain gauge network exhibited a better semi variogram structure and lowered spatial interpolation error.

      Item Type: Thesis (PhD)
      Additional Information: Thesis (PhD) - Faculty of Engineering, Universiti Malaya, 2019.
      Uncontrolled Keywords: Rain gauge network; Cross-validation; Geostatistical analysis; Optimization; Particle swarm optimization
      Subjects: T Technology > TA Engineering (General). Civil engineering (General)
      Divisions: Faculty of Engineering
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 11 Mar 2022 09:41
      Last Modified: 11 Mar 2022 09:41
      URI: http://studentsrepo.um.edu.my/id/eprint/12966

      Actions (For repository staff only : Login required)

      View Item