Congress Resources: Papers, posters and presentations

< Return to abstract list

Investigating Climate Change Effects On Water Resources Of Metropolitan City Of Istanbul In Turkey

Congress: 2015
Author(s): Ismail Yucel (Ankara, Turkey), Fatih Kara
Fatih University1

Keyword(s): Sub-theme 17: Climate change, impacts and adaptation,
Abstract 1. Introduction Water is one of the main requirements for human life and the distribution of fresh water resources is being affected by the climatic conditions of the earth. Scientific studies have demonstrated that the climate of the earth has changed since the beginning of 20th century owing to anthropogenic factors. Consequently, the frequency and magnitude of extreme weather events such as floods and droughts, have increased across almost the whole world. As a result, it is important to determine possible changes in climates and their effects on the environment in order to take necessary precautions. It is expected that climate change will affect water resources in a negative way in Turkey. Particularly, in Turkey, Istanbul shows a sharp increase in population and therefore, together with consequences from climate change the water scarcity is already becoming a serious problem for residents. The Omerli Basin has been selected as the study area for this work because of its importance to the water resources of Istanbul. The main goal of this study is to determine the impact of climate change on the water resources of the Omerli Basin using a number of GCM/RCM combinations together with a downscaling method based on spatially weighted regression and a hydrological model. The research objectives in reaching this main goal are presented as follows: * Determine changes in precipitation patterns including extreme events in the current and future climate conditions. * Determine changes in runoff patterns including extreme events in the present and future climate conditions. * Assess the impacts of downscaling RCM results on average and extreme precipitation and runoff for current and future climate conditions. * Determine frequency behavior of extreme precipitation and runoff for current and future climate. 2. Methods/Materials Climate change is expected to increase the frequency and intensity of extreme precipitation events all around the world (IPCC, 2014). Several climate change studies have been conducted to assess the impact of these changing conditions (Fowler and Ekström, 2009). These studies are usually conducted using GCM (Global Circulation Model)/RCM (Regional Climate Model) combinations and downscaling methods. High-resolution data are used in hydrological models to estimate changes in hydrological variables. GCM data are widely used in climate change studies but the spatial resolution of these data is approximately 2°. It is not possible to use these coarse resolution data in hydrologic models. RCMs provide data in 25-km spatial resolution but these data may still be too coarse for watershed scales. Consequently, further downscaling operations might be needed for local-scale studies (Fowler et al. 2007). Hydrologic response of the Omerli Basin using the RCM precipitation input with and without downscaling method was determined by the application of HBV hydrologic model in daily time step. In this study, precipitation and discharge properties of the biggest fresh water resource of Istanbul, Omerli Basin, are investigated by GCM/RCM combinations for the past (1961--1990) and the future (2071--2100). Geographically weighted regression (GWR) is selected as the downscaling method as it considers local geo-physical paramaters that influence precipitation distribution. 3. Results and Discussion Comparisons were made between modeled variables (precipitation and streamflow) and observed data for model simulations (both RCM and HBV) with and without downscaling method to investigate the improvement (or otherwise) in RCMs' and HBV's abilities to describe the mean, extreme, and frequency values of surface precipitation and streamflow fields. The extreme value series were extracted using peak over threshold (POT) method for 1-yr and 5-yr threshold return periods. As expected, precipitation downscaling was found to greatly improve the simulated precipitation and streamflow values versus no downscaling method during winter, spring, summer, and fall seasons, and annually. The most striking improvement was in the accuracy with which models (RCMs and HBV) calculates the magnitude and frequency of extreme precipitation and streamflow events. However, there was a strong underestimation tendency from all RCMs in estimating daily precipitation for all seasons and this also appeared with simulated discharges. Extreme index values described as ratio between mean extreme series for reference (1960-1990) and mean extreme series for scenario (2070-2100) periods were calculated for each season to determine the changes from past to future for the magnitudes of precipitation and streamflow under the influence of downscaling technique. Frequency behaviors of simulated extreme daily precipitation and streamflow were shown to assess whether the extreme magnitude increases or decreases for a given return period for future period and how that influenced with downscaling for different RCMs. Generally, seasonal and annual changes in extreme variables were more enhanced with downscaling method for both threshold return periods. Also, with downscaling the variability in producing precipitation amount from RCMs was reduced especially for spring and fall seasons. Majority of RCMs regardless of downscaling method showed that magnitudes of extremes will increase for a given return period and the frequency will decrease for a magnitude in future. 4. Conclusions The primary conclusions that can be drawn on the basis of the research described in this thesis are as follows: * It was shown that downscaling had a strong influence on precipitation extremes. * All RCMs with varying magnitude showed underestimation tendency in mean and extreme precipitation. * Errors in simulating extremes were found to vary with the seasons. The lowest errors in predicting extreme precipitations were obtained in the winter season and the highest errors were obtained for the spring season. * Changes in extreme precipitation from the reference to the future increased for winter, spring, and summer seasons, and decreased for fall season and annually. These changes became more significant with downscaling. The fall season has a strong effect on annual extreme precipitation. * The HBV significantly underestimated daily mean runoff and extreme runoff events. Downscaled precipitation inputs provided improvements in runoff simulations particularly with extreme values. * Spring extreme runoffs produced the smallest errors in comparison with the other seasons. * Projections showed that the magnitudes of extreme discharges increased in the winter and spring while they decreased in the fall and summer from reference to future periods. * Downscaled precipitation input provided much better frequency distribution for extreme discharges than RCMs with no downscaling input. For a given probability of occurrence the underestimation problem was significantly improved. 1. Fowler, H. J. and Ekström, M. (2009). Multi-model ensemble estimates of climate change impacts on UK seasonal precipitation extremes. Int. J. Climatol., 29: 385–416. 2. Fowler, H. J., Blenkinsop, S., and Tebaldi, C. (2007). Linking climate change modelling to impacts studies: recent advances in downscaling techniques for hydrological modelling. Int. J. Climatol., 27: 1547–1578. 3. IPCC (2014). Climate Change 2014: Impact, Adaptation, and Vulnerability. Contri-bution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.
2011 IWRA - International Water Resources Association office@iwra.org - http://www.iwra.org - Admin