Incorporating Artificial Water Transfers from the Tigris to the Euphrates into the WaterGAP Global Hydrology Model

The Tigris-Euphrates River Basin (TERB) is a transboundary water system of critical importance in West Asia, spanning Turkey, Syria, Iraq, and Iran. Its water resources support agriculture, industries, and hydropower, and water flows and storages are significantly affected by human interventions, including dam construction, water abstractions, and artificial water transfers. While a few global hydrological models, such as the WaterGAP Global Hydrology Model (WGHM), include human interventions like reservoirs, as well as surface water and groundwater use, including water transfers between adjacent grid cells, long-distance artificial water transfers, are not simulated except by the global hydrological model H08 (Hanasaki et al. 2018). H08 includes the location of 55 aqueducts but all outside of the TERB. However, they assume, for lack of information on transferred water flows, that water flows correspond to the demand for surface water abstractions in grid cells connected to the aqueduct until the river flow at the origin of the aqueduct becomes zero. This assumption certainly does not represent most water transfers well. In this study, for the first time, we modeled long-distance artificial water transfer in WGHM as the Tigris-to-Euphrates water transfer via Lake Tharthar is crucial for the water flows and storages in the TERB. Based on an analysis of observed streamflow data for 2002-2021 at the Baghdad station (downstream of the lake on the Tigris) and the volume of water transferred to the lake from the Tigris at the Samarra site, we developed a diversion algorithm. The algorithm directs streamflow above a given threshold (534 m3/sec) from December to July to Lake Tharthar to maintain stable streamflow at the Baghdad station, consistent with observations. Lake Tharthar is treated as a regulated lake instead of an inland sink, with its outflow transferred to the Euphrates River. Results demonstrate that simulating water transfers between the Tigris and Euphrates Rivers improves the accuracy of streamflow simulations at the Baghdad station. The mean simulated streamflow at the Baghdad station for the standard WGHM simulation was 1052 m³/sec, which, after modification, was reduced to 561 m³/sec, bringing it much closer to the mean observed value of 525 m³/sec. Additionally, the variability of the simulated streamflow in relation to the observed values (the ratio of the standard deviation of the simulated streamflow to the standard deviation of the observed streamflow (117 m³/sec)) improved from 5.4 for the standard WGHM to 2.4 for the modified simulation. These findings highlight the necessity of integrating artificial water transfers into hydrological models to better capture the alteration of natural water flows and storages.References Hanasaki, N., Yoshikawa, S., Pokhrel, Y., & Kanae, S. (2018). A global hydrological simulation to specify the sources of water used by humans. Hydrology and Earth System Sciences, 22(2), 789–817. https://doi.org/10.5194/hess-22-789-2018