The Climate Change Impact on Water Resources of Upper Indus Basin-Pakistan Institute of Geology University of the Punjab, Lahore
Total Page:16
File Type:pdf, Size:1020Kb
THE CLIMATE CHANGE IMPACT ON WATER RESOURCES OF UPPER INDUS BASIN-PAKISTAN By Muhammad Akhtar M.Sc (Applied Environmental Science) Under the Supervision of Prof. Dr. Nasir Ahmad M.Sc. (Pb), Ph.D. (U.K) A thesis submitted in the fulfillment of requirements for the degree of Doctor of Philosophy INSTITUTE OF GEOLOGY UNIVERSITY OF THE PUNJAB, LAHORE-PAKISTAN 2008 Dedicated to my parents CERTIFICATE It is hereby certified that this thesis is based on the results of modelling work carried out by Muhammad Akhtar under my supervision. I have personally gone through all the data/results/materials reported in the manuscript and certify their correctness/ authenticity. I further certify that the materials included in this thesis have not been used in part or full in a manuscript already submitted or in the process of submission in partial/complete fulfillment for the award of any other degree from any other institution. Mr. Akhtar has fulfilled all conditions established by the University for the submission of this dissertation and I endorse its evaluation for the award of PhD degree through the official procedure of the University. SUPERVISOR 0. cL- , Nasir Ahmad, PhD Professor Institute of Geology University of the Punjab Lahore, Pakistan ABSTRACT PRECIS (Providing REgional Climate for Impact Studies) model developed by the Hadley Centre is applied to simulate high resolution climate change scenarios. For the present climate, PRECIS is driven by the outputs of reanalyses ERA-40 data and HadAM3P global climate model (GCM). For the simulation of future climate (SRES B2), the PRECIS is nested with HadAM3P-B2 global forcing. In the present day simulations, climatic means and interannual variability are examined and biases are identified focusing on the most important parameters (precipitation and temperature) for hydrological modelling. In this study, both the meteorological station observations and results of the PRECIS RCM are used as input in the HBV hydrological model in order to investigate the effect of PRECIS simulated precipitation and temperature on the HBV predicted discharge in three river basins of UIB region. For this, three HBV model experiments are designed: HBV-Met, HBV-ERA and HBV-PRECIS where HBV is driven by meteorological station data and by the outputs from PRECIS nested with ERA-40 and HadAM3P data respectively. The robustness and uncertainties ranges of these models are tested. The future water resources are quantified using the two approaches of transferring the climate change signals i.e. delta change approach and direct use of PRECIS data. The future discharge is simulated for three stages of glacier coverage: 100 % glaciers, 50 % glaciers and 0 % glaciers. The PRECIS is able to reproduce the spatial patterns of the observed CRU mean temperature and precipitation. However, there are notable quantitative biases over some regions especially over the Hindukush-Karakorum-Himalaya (HKH) region, mainly due to the similar biases in the driving forcing. PRECIS simulations under future SRES B2 scenario indicate an increase in precipitation and temperature towards the end of 21st century. The calibration and validation results of the HBV model experiments show that the performance of HBV-Met is better than the HBV-ERA and HBV-PRECIS. However, using input data series from sources different from the data used in the model calibration shows that HBV-ERA and HBV-PRECIS are more robust compared to HBV-Met. The Gilgit and Astore river basins, for which discharges are depending on the preceding winter precipitation, have higher uncertainties compared to the Hunza river basin for which the discharge is driven by the energy inputs. The smaller uncertainties in the Hunza river i basin as compared to Gilgit and Astore river basins may be because of the stable behavior of the input temperature series compared to the precipitation series. The robustness and uncertainty ranges of the HBV models suggest that regional climate models may be used as input in hydrological models for climate scenarios studies. In a changed climate, the discharge will generally increase in both HBV-PRECIS and HBV-Met in the 100 % glacier coverage stage up to 65% and 44%, respectively. At the 50 % glacier coverage stage, the discharge is expected to reduce up to 24% as predicted by HBV-PRECIS and up to 30% as predicted by HBV-Met model. For the 0 % glacier coverage under climate change, a drastic decrease in water resources is forecasted by HBV-Met is up to 96 % and by HBV-PRECIS is up to 93%. At 100 % glacier coverage, the magnitude of flood peaks is likely to increase in the future which is an indication of higher risk of flood problems under climate change. There are huge outliers in annual maximum discharge simulated with HBV-Met. This shows that the prediction of hydrological conditions through the delta change approach is not ideal in the UIB region. HBV-PRECIS provides results on hydrological changes that are more consistent with climate change. This shows that the climate change signals in HBV-PRECIS are transmitted more realistically than in HBV-Met. Therefore, the direct use of RCM outputs in a hydrological model may be an alternative in areas where the quality of observed data is poor. The modeled changes in future discharge and changes in peak flows under climate change are not conclusive because more research is needed to evaluate the uncertainties in this approach. Moreover, this technique needs to be tested with other RCMs and hydrological models preferably to river basins in other parts of the world as well. ii ACKNOWLEDGEMENTS I would like to extend my sincere thanks to my research supervisor Prof. Dr. Nasir Ahmad (Director Institute of Geology, University of the Punjab) for his keen interest, proficient guidance, valuable suggestions, and encouraging attitude during the course of this research work. I would like to thank the PRECIS team at Hadley Centre, Meteorological Office, U.K, on providing training in PRECIS regional climate modelling system and extending continuous help in solving day to day operational simulation problems. Special thanks are due to David Hein who provided boundary data of different GCMs on behalf of Hadley Centre. The river discharge data and meteorological data have been taken from Water and Power Development Authority (WAPDA) and Pakistan Meteorological Department (PMD), respectively. I am grateful to the scientists at Swedish Meteorological and Hydrological Institute (SMHI) for their useful comments and valuable suggestions during the study. I am indebted to the PMD on granting study leave for doctoral study at the University. Financial support extended by the Higher Education Commission under the indigenous PhD Scholarship Scheme is most gratefully acknowledged. I am thankful to ICTP, Trieste, Italy, for providing two months fellowship, which enhanced my modelling capabilities. The valuable suggestions and technical skill provided by Dr. Jermy Paul during my stay at ICTP helped improve my understanding towards modelling technique. Suggestions and critical comments by Dr. Martijn Booij of the Twente University, Netherlands and Dr. David Hein and Dr. Wilfran-Moufouma of Hadley Centre, U.K. greatly improved quality of my research work. Finally, I would like to express my heartiest gratitude to my parents, wife, sisters, brothers and friends whose cooperation, prayers and well wishes strengthened my confidence to endure the hardships faced during this study. iii LIST OF TABLES Table 1.1 Dimensions of some large glaciers in the UIB region 6 Table 3.1 Description of PRECIS RCM experiment 28 Table 3.2 Biases in mean temperature (˚C) as simulated with the PRECIS- 39 Had, PRECIS-ERA and HadAM3P relative to CRU reference data for different seasons and seven sub regions of figure 3.3 (Summer= April-September, Winter = October-March) Table 3.3 Biases in mean precipitation (%) as simulated with the PRECIS- 47 Had, PRECIS-ERA and HadAM3P relative to CRU reference data for different seasons and seven sub regions of figure 3.3 (Summer= April-September, Winter = October-March) Table 3.4 Seasonal changes of mean temperature and precipitation under 54 SRES B2 scenario from PRECIS in 2071-2100 over the seven sub regions relative to 1961-1990 (Summer = April-September; Winter=October-March) Table 4.1 Characteristics of study area 58 Table 4.2 Temperature and precipitation during two monsoon events at 60 selected stations Table 4.3 Biases in mean temperature (˚C) as simulated with PRECIS RCMs 63 relative to CRU reference data for different seasons and river basins (Winter =October-March; Summer =April-September) Table 4.4 Biases in precipitation (%) as simulated with PRECIS RCMs 63 relative to CRU reference data for different seasons and river basins (Winter = October- March; Summer = April-September) Table 4.5 Values and range of important parameters found in different 75 studies using HBV model Table 4.6 Parameter values for HBV for three river basins with three 75 different input data sets Table 4.7 Performance of three HBV models during calibration and 76 validation periods in different river basins Table 4.8 Efficiency Y of three HBV models using data sources different 86 from the calibration sources during the hydrological years 1985 and 1986 in different river basins. The values of absolute relative deviations (ARD) are given in parentheses. The italic values indicate efficiency Y during calibration Table 5.1 Seasonal changes of mean temperature and precipitation under 94 PRECIS simulated SRES B2 scenario for the period 2071-2100 over three river basins relative to the period 1961-1990 (Summer = April-September; Winter=October-March) iv Table 5.2 Mean relative change in future discharge (2071-2100) in a 102 changed SRES B2 climate relative to the present discharge (1961- 1990) for three glaciations stages and for three river basins Table 5.3 Characteristics of future annual maximum discharge simulated by 106 two HBV models in a changed SRES B2 climate for the three glaciations stages and for three river basins.