Assessing the Impact of Climate Change on the Water Resources of the Seyhan River Basin, Turkey
Total Page：16
File Type：pdf, Size：1020Kb
Assessing the Impact of Climate Change on the Water Resources of the Seyhan River Basin, Turkey Yoichi FUJIHARA1, Kenji TANAKA2, Tsugihiro WATANABE3, and Toshiharu KOJIRI4 1;3Research Institute for Humanity and Nature, 457-4 Kamigamo-motoyama, Kita-Ku, Kyoto 603-8047 2;4Disaster Prevention Research Institute, Kyoto University, Gokasho, Uji 611-0011 1. Introduction mm in the northern area. The annual inflow at the Seyhan Dam ranges between 3.7 and 7.3 Gm3 and The Intergovernmental Panel on Climate averages 5.5 Gm3. The Seyhan and Catalan Dams Change (IPCC) Third Assessment Report con- have storage capacities of 0.8 and 1.6 Gm3, re- cluded that there was evidence that most of the spectively. The stored water is used mainly for warming observed over the last 50 years is at- irrigation. According to the 1990 statistics, the tributable to human activities. With the expected amount of irrigation water used annually is about build-up of greenhouse gases in the atmosphere, 1.4 Gm3, and it is increasing annually (Figure 1). it is anticipated that the climate will continue to The amount of domestic water used annually is 0.1 change throughout the 21st century. Moreover, it Gm3 according to the 2003 statistics. is thought that global warming will have a signifi- ) 10.0 cant impact on the hydrology and water resources 3 Annual Inflow m 9 9.0 of river basins. Irrigation Water 8.0 Basins that have a large fraction of runoff driven Domestic Water 7.0 by snowmelt, such as the Seyhan River Basin 6.0 in Turkey, will be especially sensitive to global 5.0 warming, because the temperature determines the 4.0 fraction of precipitation that falls as snow and the 3.0 timing of snowmelt. In this paper, the climate 2.0 projected using two general circulation models 1.0 Annual Inflow at the Seyhan Dam (10 0.0 (GCMs) under the Special Report on Emissions 1990 1992 1994 1996 1998 2000 2002 2004 Scenarios (SRES) A2 emissions scenario was used Fig.1 Annual inflow, irrigation water, and to drive hydrologic models to assess the impact of domestic water use at the Seyhan Dam. climate change on the water resources of the Sey- han River Basin. 3. Approach 2. Study Basin 3.1 Downscaling Method The raw outputs of GCMs are inadequate for as- The Seyhan River Basin (21,700 km2) is lo- sessing the impact of climate change on the hy- cated in southern Turkey between 34.25-37.0±E drology and water resources of river basins, be- and 36.5-39.25±N. The lower basin is dominated cause the temporal and spatial resolution of GCMs by the Mediterranean climate, while the middle is too coarse compared to those of hydrologic and upper basins are influenced by the Continental models that are applied to river basins. This study climate. applied a dynamic downscaling method called The annual precipitation is about 700 mm in the pseudo warming (Sato et al., 2006) to connect the coastal area, increases to approximately 1,000 mm output of the raw GCMs and river basin hydro- at higher elevations, and decreases to about 400 logic models. 600 600 Observed Observed /s) Simulated (Present) Simulated (Present) /s) 3 3 400 400 200 200 Inflow at the Station 1818 (m Inflow at the Station 1818 (m 0 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 1 2 3 4 5 6 7 8 9 10 11 12 (a) Monthly inflow (b) 10-year average monthly inflow Fig.2 Simulated hydrograph at Station 1818. The pseudo warming downscaling method is as discharge. Nevertheless, since the input data were follows. For the current climate simulation, the downscaled data, the hydrologic models repro- pseudo-warming method uses reanalysis data as duce the river discharge at station 1818. The an- a boundary forcing of the regional climate model nual inflow at the Seyhan Dam is shown in Fig- (RCM). A specially created boundary condition, ure 3. This figure shows that the simulated results in which changes in meteorological variables pro- agree with the observed data. jected in a GCM simulation are added to reanaly- sis data, is used to simulate global warming. ) 10.0 3 Observed The GCMs used in this study were m 9 9.0 Simulated (Present) MRI-CGCM2 (Yukimoto et al., 2001) and 8.0 CCSR/NIES/FRCGC-MIROC (K-1 Model De- 7.0 velopers, 2004) under SRES A2. The downscaled 6.0 data covered two subset periods (the 10 years 5.0 present and the 10 years future; Kimura et al., 4.0 2007), and were used to drive hydrologic mod- 3.0 2.0 els to assess the impact of climate change on the 1.0 water resources of the Seyhan River Basin. Annual Inflow at the Seyhan Dam (10 0.0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 3.2 Hydrologic Model Fig.3 Annual inflow at the Seyhan Dam. We used a land surface model (Simple Bio- sphere including Urban Canopy (SiBUC); Tanaka 3.3 Reservoir Models and Ikebuchi, 1994) to estimate the surface en- We developed reservoir models to simulate the ergy and water balance components. In addition, reservoir operations of the Seyhan and Catalan we used the stream flow rooting model of Hydro- Dams. We examined the historical record, includ- BEAM (Kojiri et al., 1998) to simulate river dis- ing the inflow, water level, and dam discharge, and charge and incorporated a reservoir model in this interviewed the dam operators about the actual op- flow rooting model. erations. From these analyses, we used the follow- The region simulated was a 2.75 × 2.75± ing operational rule as a basic rule: water is stored area (34.25-37.0±E and 36.5-39.25±N) with a 5- to maintain a target operational water level and the minute latitude-longitude spatial resolution (33 × demand water is released regardless of the level. 33 grids). The simulated hydrograph at station The simulated river discharge using the flow 1818 is shown in Figure 2. There were some rooting model described in section 3.1 is input into discrepancies between the simulated and observed the reservoir models. The target operational wa- 1800 1000 ) Observed 3 Observed m Simulated 6 800 Simulated 1500 600 ) 3 m 6 1200 400 200 900 1000 1000 Reservoir Volume (10 0 Observed /s) Observed /s) Simulated 800 3 Simulated 800 3 600 600 600 Reservoir Volume (10 400 400 300 200 200 Dam Discharge (m Dam Discharge (m 0 0 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 (a) Catalan Dam (b) Seyhan Dam Fig.4 Simulated reservoir volume and discharge. ter level is the average of historical operational 3.4 Land and Water Use Scenarios records, and the demand water is the actual water The land and water use at the present period withdrawal for irrigation and domestic use. were the actual conditions in the Seyhan River The simulated reservoir volume and dam dis- Basin. For the future period, the following three charge at the Seyhan and Catalan Dams are shown scenarios were used: in Figure 4. The simulated volume and discharge (a)Future: The land and water use are the same agreed with the observed values. Although no re- as at present. sults are shown in here, we found that the estab- (b)Adaptation 1: The land and water use are un- lished reservoir models also reproduced the hydro- der low investment conditions. The cropping pat- electric generation quite well. Figure 5 shows the tern in the Lower Seyhan Irrigation Project (LSIP) simulated inflows with and without the reservoir simulated by Umetsu et al. (2007) is used to es- models. This figure clearly indicates that the reser- timate the water demand. In addition, the effects voir models can reproduce the actual reservoir op- of global warming on the irrigation water require- erations. ments are considered using the SiBUC simulation. (c)Adaptation 2: The land and water use are un- 600 Observed /s) 3 Simulated (with reservoir model) Simulated (without reservoir model) 400 200 Inflow at the Seyhan Dam (m 0 1 2 3 4 5 6 7 8 9 10 11 12 Fig.5 Simulated hydrograph at the Seyhan Dam. der high investment conditions, in which 25% of 120 the rain-fed winter wheat is converted to irrigated 100 Present crop, and citrus is cultivated in this area. The crop- Future (MRI) Future (CCSR) ping pattern in the LSIP simulated by Umetsu et 80 al. (2007) is used as a future scenario to calculate the water demand. The effects of global warming 60 on the irrigation water requirements are also con- 40 sidered using the SiBUC simulation. Monthly Precipitation (mm) 20 4. Results 0 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. 4.1 Temperature, Precipitation and Stream Fig.7 Precipitation changes predicted under different models. Flow Changes The monthly mean temperatures are compared 400 in Figure 6. The average annual temperature change for the Seyhan River Basin was +2.0±C Present 300 Future (MRI) /s) in the Meteorological Research Institute GCM 3 Future (CCSR) (MRI) and +2.7±C in the Center for Climate Sys- tem Research GCM (CCSR). The monthly pre- 200 cipitation is compared in Figure 7. The aver- age annual precipitation change for the Seyhan Monthly Inflow (m 100 River Basin was -159 mm in MRI and -161 mm in CCSR. The decreases in precipitation in Jan- 0 uary, April, October, November, and December 1 2 3 4 5 6 7 8 9 10 11 12 were greater than in the other months.