Flood modelling of the Phetchaburi river basin
The importance of spatially distributed precipitation for flood modelling
K. Chen (TU Delft - Civil Engineering & Geosciences)
T.A. Bogaard – Mentor (TU Delft - Surface and Groundwater Hydrology)
Olivier Hoes – Graduation committee member (TU Delft - Water Systems Monitoring & Modelling)
Punpim Puttaraksa Mapiam – Graduation committee member
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
As flood intensities are increasing, Flood Early Warning Systems (FEWS) are becoming more crucial so the right mitigation measures can be taken. This study focused on the Phetchaburi river basin in Thailand, which experiences floods yearly. Currently, precipitation data from rain gauges are used in the FEWS for this region. However, weather radar is also available in Thailand. Weather radar is usually able to capture the spatial variability of precipitation in more detail. Therefore, this study aimed to research the effect of spatially distributed precipitation on flood modelling in the Phetchaburi river basin.
Using HEC-RAS, a hydrodynamic rainfall-runoff model of the Phetchaburi river basin was made. The area of interest was the middle reach of the Phetchaburi river and its two tributaries, lying in-between three reservoir dams upstream and the Phet Diversion Dam downstream. The Phetchaburi river and its tributaries were calibrated using measured dam outflow and water level data. Three types of precipitation data were used as input, namely homogeneous precipitation data and spatially distributed precipitation data obtained from weather radar and from rain gauges.
High infiltration rates were found for the Phetchaburi river basin. When homogeneous precipitation data was used as input, precipitation intensity would be too low, allowing all precipitation to infiltrate into the subsurface. Using homogenous precipitation data as input results in a underestimation of floods. On the contrary, precipitation intensities in spatially distributed precipitation data were high enough to exceed the soil infiltration capacity, leading to surface runoff and floods. When solely looking at the water balance, using precipitation data from weather radar and rain gauges as input lead to similar results. However, when it came to the water levels at specific locations, precipitation data from weather radar performed better. The density of the rain gauge network in the Phetchaburi river basin is too low to capture the spatial variability of the precipitation events in detail. This resulted in floods or the
lack thereof at the wrong locations. Weather radar captures the spatial variability of precipitation in greater detail than rain gauges can. This study showed that using precipitation data from weather radar as input results in more accurate flood modelling in the Phetchaburi river basin.