Study On Climate Change Impact On Flood Control Safety In High Risk Flood Areas In The Beijjiang River Basin, South China

Because of human activity, atmospheric concentrations of some greenhouse gases are increasing, and most climatologists believe that this is causing significant climate change. Projected increases in global temperatures are expected to affect the hydrological cycle—make the spatio-temporal distribution of rainfall more uneven and increase the frequency and magnitude of flood and drought events. Investigation of the changes and trends expected for future extreme precipitation and flooding is of great significance for flood control measures. In this study, we present an analysis of the implications of climate change on flood control in two typical high-risk flooding areas(Feilaixia reservoir and Guangzhou City) of the Beijiang River basin, South China. The main research work in this thesis is as follows:(1) We investigated the spatial and temporal distributions of trends in climate extremes in the upstream area of the Feilaixia reservoir(called the Feilaixia catchment) over the period 1969–2011, utilizing extreme climate indices developed by the joint CCl/CLIVAR/JCOMM Expert Team(ET) on Climate Change Detection and Indices(ETCCDI). Trends in extreme climate indices were calculated using Sen’s slope estimator. Trend stability was determined using the statistical significance of the Mann–Kendall(M-K) trend test. Analyses of extreme temperature indices detected significant and stable trends for most stations. In contrast, significant and stable positive trends in precipitation extremes were sporadically recorded in the study area. Furthermore, increasing trends were much more frequent than decreasing trends for most extreme precipitation indices. Overall, the climate of the Feilaixia catchment has tended to become warmer over the past several decades, while extreme precipitation events show a weak increasing trend.(2) Based on the observed hydrological data from 1969–2011, changes in heavy precipitation and flooding and the causes of flood risk changes in the Feilaixia catchment were investigated using Sen’s slope estimator, the M-K trend test, the Pettitt test, the moving t-test, and wavelet transformation methods. Results indicated weak increasing trends in annual extreme precipitation and flood events. Monthly extreme precipitation and flood events experienced positive(negative) trends mainly in June–July(April, May, and August), and mostly did not experience significant change points. Overall, changes in flood risk were mainly influenced by precipitation variability. However, human activities(including urbanisation and the construction of reservoirs, soil and water conservation measures, and land use change, among others) have probably affected the flooding process in the study area.(3) The implications of climate change on future flood hazard in the Feilaixia catchment were assessed using a Variable Infiltration Capacity(VIC) model, which demonstrates good performance in simulating extreme floods. Uncertainty was evaluated by employing eight climate models and four emission scenarios(A1B and representative concentration pathway(RCP) scenarios RCP2.6, RCP4.5, and RCP8.5). Credibility of the projected trends in floods was described using an uncertainty expression approach, as recommended by the Fifth Assessment Report(AR5) of the Intergovernmental Panel on Climate Change(IPCC). Global Cimate Models(GCMs) and emission scenarios are a large source of uncertainty in predictions of future floods over the study region. Temperature and extreme precipitation are expected to show increasing trends over the period 2020–2050(relative to the historical period 1970–2000). Projected ranges of annual maximum 1-day discharges and annual maximum 3-, 7-, and 15-day flood volumes show relatively large variability under different future scenarios in the eight climate models, but most models project an increase in discharge and flood volumes over the period 2020–2050(relative to the historical period 1970–2000).(4) Based on observed meteorological data and climate model data(emission scenarios A1 B, RCP2.6, RCP4.5, and RCP8.5), observed changes in and future projections of extreme temperature and precipitation events and relative sea levels in Guangzhou city were investigated. Future changes in relative sea levels were projected using a semi-empirical approach. In the past several decades, a significant warming trend was found for Guangzhou city, more intense and more frequent short duration rainstorms were observed, and relative sea levels showed significant increasing trends. For the period 2020–2050, temperature, extreme precipitation, and relative sea levels are expected to continue to show an increasing trend. Furthermore, greater increases are projected for the higher emission scenarios.(5) Using the eight climate models(emission scenarios A1 B, RCP2.6, RCP4.5, and RCP8.5), we applied the Archimedean copula functions to calculate joint distributions of extreme rainfall and tidal level, and then developed a combined risk probability model for Guangzhou city under the different emission scenarios. Compared to the historical period 1970–2000, combined risk probabilities of extreme rainfall and high tidal level are expected to increase under all the emission scenarios over the period 2020–2050. Overall, there is a very high likelihood that Guangzhou will suffer more flooding in the future.