Flood and drought assessment with dam infrastracture: A case study of the Ba River basin, Fiji
Year: 2015
Type: conference-paper
Abstract: In Fiji, frequent flood and drought events require implementation of preventive infrastructure and planning measures to reduce potential damages that may be caused by these events.The Ba River basin, which is located in the Western part of the main island, Viti Levu, has experienced severe floods and droughts.The flood record of the Ba River extends from 1871 to 2009 and reports flooding of about 127 times with the most disastrous floods in 1931in , 1956in , 1993in , 1999in and 2009in (McGree et al., 2010)).For example, the 1993 flood damages amounted to about 8 million USD.For the past droughts, the rain fed sugar cane agriculture is affected by shortages of precipitation with the most severe drought in 1998.In the Ba River basin, the sugar cane agriculture recorded a loss of 21.80 km 2 at the Rarawai Mill during the 1998 drought.For the flood hazard reduction, a dry dam infrastructure is proposed upstream of the Ba Town allowing for natural river flows while only reducing flood peaks.In addition, two multipurpose dams are proposed for the water supply during droughts as well as flood reduction.In our approach, we utilize numerical models to investigate the effectiveness of dam infrastructure for flood control and water supply, and standardized indices to characterize historical droughts.For the standardized indices, the precipitation record of the Ba River basin has been investigated with the use of Standardized Precipitation Index (SPI) from 1960 to 2009 and the Standardized Precipitation Evapotranspiration Index (SPEI) was obtained from global datasets.The 3-arcsecond (about 90-m) grid block-wise TOP (BTOP) and 15-arcsecond (about 450-m) grid rainfall-runoff-inundation (RRI) models were constructed for the Ba river basin of 957 km 2 to simulate flood river flows and inundation.For the BTOP model, monthly potential evapotranspiration and Normalized Difference Vegetation Index (NDVI) were obtained using climate forcing data CRU TS3.1.Both BTOP and RRI models with short-and long-term local precipitation data were calibrated and validated with the local available data at two river gauging stations.Using the simulated discharges of the calibrated BTOP model, we estimated flood discharges of the selected return periods with flood frequency analysis and conducted flood inundation simulations with Flood Inundation Model (FID) model.The estimated flood inundation area and flood exposure with and without infrastructure are utilized for the flood risk estimation.For flood risk, we estimated the number of flood affected people and agriculture and found a satisfactory comparison with the record of the 1993 flood.For the historical drought assessment, both SPI and SPEI demonstrated the most extreme drought conditions of the 1998 drought and are consistent with the affected agriculture record during the 1998 drought.These preliminary results indicate the usefulness of our approach and suggest it could be used for selecting effective counter-measures for flood and drought risk reduction in Ba River basin.
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Keywords: flood risk, drought, flood control, water supply, Standardized Precipitation Index, Standardized Precipitation Evapotranspiration Index, rainfall-runoff-inundation model, flood inundation modeling, flood frequency analysis, drought indices, sugarcane cultivation, flood damage, flood hazard reduction, Viti Levu, Fiji, return periods
Open Access status: bronze
OpenAlex ID: W4240503495
