TY - JOUR
T1 - Assessing the impact of climate and land use changes on extreme floods in a large tropical catchment
AU - Jothityangkoon, Chatchai
AU - Hirunteeyakul, Chow
AU - Boonrawd, Kowit
AU - Sivapalan, Murugesu
N1 - Funding Information:
The authors are grateful to the Electricity Generating Authority of Thailand (EGAT) for financial and technical support for this project. Grateful acknowledgements are given to Thai Meteorological Department, Royal Irrigation Department, Department of Water Resources, Department of Groundwater Resources, and the Land Development Department for providing field data. Special thanks go to Assoc. Prof. Amnat Apichatvallop for useful comments.
PY - 2013/5/20
Y1 - 2013/5/20
N2 - In the wake of the recent catastrophic floods in Thailand, there is considerable concern about the safety of large dams designed and built some 50years ago. In this paper a distributed rainfall-runoff model appropriate for extreme flood conditions is used to generate revised estimates of the Probable Maximum Flood (PMF) for the Upper Ping River catchment (area 26,386km2) in northern Thailand, upstream of location of the large Bhumipol Dam. The model has two components: a continuous water balance model based on a configuration of parameters estimated from climate, soil and vegetation data and a distributed flood routing model based on non-linear storage-discharge relationships of the river network under extreme flood conditions. The model is implemented under several alternative scenarios regarding the Probable Maximum Precipitation (PMP) estimates and is also used to estimate the potential effects of both climate change and land use and land cover changes on the extreme floods. These new estimates are compared against estimates using other hydrological models, including the application of the original prediction methods under current conditions. Model simulations and sensitivity analyses indicate that a reasonable Probable Maximum Flood (PMF) at the dam site is 6311m3/s, which is only slightly higher than the original design flood of 6000m3/s. As part of an uncertainty assessment, the estimated PMF is sensitive to the design method, input PMP, land use changes and the floodplain inundation effect. The increase of PMP depth by 5% can cause a 7.5% increase in PMF. Deforestation by 10%, 20%, 30% can result in PMF increases of 3.1%, 6.2%, 9.2%, respectively. The modest increase of the estimated PMF (to just 6311m3/s) in spite of these changes is due to the factoring of the hydraulic effects of trees and buildings on the floodplain as the flood situation changes from normal floods to extreme floods, when over-bank flows may be the dominant flooding process, leading to a substantial reduction in the PMF estimates.
AB - In the wake of the recent catastrophic floods in Thailand, there is considerable concern about the safety of large dams designed and built some 50years ago. In this paper a distributed rainfall-runoff model appropriate for extreme flood conditions is used to generate revised estimates of the Probable Maximum Flood (PMF) for the Upper Ping River catchment (area 26,386km2) in northern Thailand, upstream of location of the large Bhumipol Dam. The model has two components: a continuous water balance model based on a configuration of parameters estimated from climate, soil and vegetation data and a distributed flood routing model based on non-linear storage-discharge relationships of the river network under extreme flood conditions. The model is implemented under several alternative scenarios regarding the Probable Maximum Precipitation (PMP) estimates and is also used to estimate the potential effects of both climate change and land use and land cover changes on the extreme floods. These new estimates are compared against estimates using other hydrological models, including the application of the original prediction methods under current conditions. Model simulations and sensitivity analyses indicate that a reasonable Probable Maximum Flood (PMF) at the dam site is 6311m3/s, which is only slightly higher than the original design flood of 6000m3/s. As part of an uncertainty assessment, the estimated PMF is sensitive to the design method, input PMP, land use changes and the floodplain inundation effect. The increase of PMP depth by 5% can cause a 7.5% increase in PMF. Deforestation by 10%, 20%, 30% can result in PMF increases of 3.1%, 6.2%, 9.2%, respectively. The modest increase of the estimated PMF (to just 6311m3/s) in spite of these changes is due to the factoring of the hydraulic effects of trees and buildings on the floodplain as the flood situation changes from normal floods to extreme floods, when over-bank flows may be the dominant flooding process, leading to a substantial reduction in the PMF estimates.
KW - Compound channel
KW - Dam safety
KW - Probable Maximum Flood (PMF)
KW - Probable Maximum Precipitation (PMP)
KW - Rainfall-runoff model
UR - http://www.scopus.com/inward/record.url?scp=84876823013&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84876823013&partnerID=8YFLogxK
U2 - 10.1016/j.jhydrol.2013.03.036
DO - 10.1016/j.jhydrol.2013.03.036
M3 - Article
AN - SCOPUS:84876823013
SN - 0022-1694
VL - 490
SP - 88
EP - 105
JO - Journal of Hydrology
JF - Journal of Hydrology
ER -