Browsing by Subject "Flood forecast"
Now showing items 1-2 of 2
-
Gumbricht, T.; Wolski, P.; Frost, P.; McCarthy, T.S. (Elsevier http://www.doi:10.1016/j.jhydrol.2003.11.010, NaN, 2004)[more][less]
Abstract: The pristine Okavango Delta wetland of northern Botswana is potentially under threat due to water abstraction from its tributaries. We have developed a statistical model which makes it possible to predict the extent of wetland loss which will arise from water abstraction. The model also permits prediction of the maximum area of flooding, and its spatial distribution, three months in advance of the flood maximum. The model was calibrated using maximum areas of seasonal inundation extracted from satellite imagery covering the period 1985–2000, which were correlated with rainfall and total flood discharge. A technique was developed to translate the modelled flood area into a flood map. The methodology can predict maximum area of flooding and its distribution with better than 90% accuracy. An important, although relatively minor, source of error in the spatial distribution of the flood arises from a secular change in flood distribution in the distal Delta which has taken place over the last 15 years. Reconstruction of flooding history back to 1934 suggests that the Delta may be subject to a quasi 80 year climatic oscillation. If this oscillation continues, the extent of flooding will increase in the coming decades. URI: http://hdl.handle.net/10311/428 Files in this item: 1
Gumbricht_JH_2004.pdf (1.267Mb) -
Gumbricht, T.; Wolski, P.; Frost, P.; McCarthy, T.S. (Elservier, www.elsevier.com, NaN, 2004)[more][less]
Abstract: The pristine Okavango Delta wetland of northern Botswana is potentially under threat due to water abstraction from its tributaries. We have developed a statistical model which makes it possible to predict the extent of wetland loss which will arise from water abstraction. The model also permits prediction of the maximum area of flooding, and its spatial distribution, three months in advance of the flood maximum. The model was calibrated using maximum areas of seasonal inundation extracted from satellite imagery covering the period 1985-2000, which were correlated with rainfall and total flood discharge. A technique was developed to translate the modelled flood area into a flood map. The methodology can predict maximum area of flooding and its distribution with better than 90% accuracy. An important, although relatively minor, source of error in the spatial distribution of the flood arises from a secular change in flood distribution in the distal Delta which has taken place over the last 15 years. Reconstruction of flooding history back to 1934 suggests that the Delta may be subject to a quasi 80 year climatic oscillation. If this oscillation continues, the extent of flooding will increase in the coining decades. URI: http://hdl.handle.net/10311/948 Files in this item: 1
Gumbricht JOH 2004.pdf (2.449Mb)
Now showing items 1-2 of 2