Modelling The Impacts Of Land Use/Cover Changes On Flood Risk:Case Studies In China And Sub-saharan Africa

Flood disasters have been causing loss of human lives and affecting the economic development in many regions of the world in the last three decades.Flood events have been particularly more dangerous in urban areas affecting millions of people.So far,most of the flood risk management studies have focused on assessing the impact of land use/cover changes on flooding at the catchment scale.However,this approach of studying floods analyzes only one side of the problem,without considering the spatial heterogeneity of land cover within the small-scale level of the catchment(e.g.,sub-basin).As a result,flood mitigation strategies focus on developing early warning systems and other structural measures to reduce flood risk.Moreover,only few previous studies have tried to examine the effectiveness of various land cover management measures to reduce flood disasters,especially in sub-Saharan Africa countries.To address these gaps,the main objective of this dissertation was to investigate the characteristics of historical land use/cover dynamics,hydrological responses and their interactions at the basin and sub-basin scales.In addition,this dissertation investigated the effectiveness of various land cover management scenarios in reducing flooding.The assessment of the relationship between land use/cover(LULC)changes and flooding is important to better understand the impact of anthropogenic activities on flood occurrence.In the present dissertation,we answered our research questions using four case studies in both China and Sub-Saharan Africa.The four case studies have undergone various LULC changes and have been affected by recurrent flood events in the last three decades.The first two cases are in mountainous regions in western China,which were selected to assess the effect of reforestation driven land use/cover changes on flood peak discharge.The third one is the case of fast-urbanized catchment in eastern China,which was used to investigate the impact of urbanization driven LULC changes on streamflow.The fourth one is the case of Sub-Saharan Africa in Rwanda(characterized by unplanned urban development),which was used to explore a set of land use/cover management strategies with various spatial arrangements of land use/cover on flood reduction.The four cases have experienced LULC changes and are located in different geographical and climatic conditions,hence serve as the ideal areas to respond to our research questions.Remote sensing data and techniques are excellent ways of assessing changes in LULC.We used the maximum likelihood classifier algorithm to classify a series of Landsat satellite images for land use/cover map reconstruction in the selected case studies.This was useful to monitor and quantify changes in land use/cover.Moreover,the Hydrologic Models(Hydrologic Modeling System(HEC-HMS)and OpenLISEM)coupled with meteorological data were utilized to examine the impact of changing LULC on hydrological responses at both the basin and sub-basin scales.The models were calibrated and validated using ground data.Key findings were as follows:(1)major changes observed in land use/cover were conversion from agriculture land into forest land cover.The forest land cover type increased up to 18%from 1990 to 2017 in the two mountainous study catchment.Changes in forest land resulted in a decrease in flood peak discharge up to 14%at the basin scale and up to 19%at the sub-basin scale.This demonstrated the effectiveness of Grain to Green Program(GTGP)and Natural Forest Conservation Program(NFCP)programs introduced in the last three decades.(2)The anthropogenic effects(built-up,roads,etc.)were the main cause of the expansion in urban land cover in the Liangshuihe catchment in Eastern China.The areas covered by urban land increased up to 40.26%during 1985-2017.The increase in urban land cover resulted in various degrees of increase in flood peak discharge and volume.At the basin scale,we observed an increasing trend in flood peak discharge and flood volume up to 7.02%and 11.93%,respectively,during the study period.Furthermore,the impact of urban expansion on flood peak discharge and flood volume was more obvious at sub-basin level.The highly urbanized sub-basins demonstrated an increase in flood peak discharge and flood volume up to 24.44%and 25.88%,respectively.The most staggering findings from this analysis were that the changes in flood peak discharge and flood volume at the sub-basin scale were closely related to the variation in land cover types.The higher the increasing rate of forest and vegetation cover in the sub-basin,the greater the decreasing rate of flood volume and flood peak discharge.(3)The findings indicated that configuration of various land use/cover management measures affected surface runoff,flood volume and flooded area at the catchment outlet differently.We observed that land cover management measure(characterized by the combination of replacement of unplanned settlement located on steep slope by forest and vegetation cover and structural measures)would reduce surface runoff,flood volume and flooded areas up to 63%,47%and 36%,respectively.These findings suggested that the spatial distribution of land use/cover management measures in the catchment had a significant impact on the reduction of flooding at the outlet.To summarize,this dissertation focused on extracting land use/cover information from satellite images and investigated the impact of changing land cover on hydrological responses at small-scale level.In this sense,the most noticeable consideration of investigating the impact subbasin land use/cover changes on floods contributes to the growth of scientific field of floods related studies and can be applied in other parts of the world.Moreover,this dissertation provided a framework for better understanding the impact of LULC changes on spatial patterns of hydrological responses and could be useful for land use planners and policy makers for adopting suitable strategies to control flood related disaster in many regions of the World.