In:
Hydrology and Earth System Sciences, Copernicus GmbH, Vol. 24, No. 4 ( 2020-04-17), p. 1939-1956
Abstract:
Abstract. During the last decades, the endorheic Lake Urmia basin in northwestern Iran has suffered from declining groundwater tables and a very strong recent reduction in the volume of Lake Urmia. For the case of Lake Urmia basin, this study explores the value of different locally and globally available observation data for adjusting a global hydrological model such that it can be used for distinguishing the impacts of human water use and climate variations. The WaterGAP Global
Hydrology Model (WGHM) was for the first time calibrated against multiple
in situ and spaceborne data to analyze the decreasing lake water volume,
lake river inflow, loss of groundwater, and total water storage in the
entire basin during 2003–2013. The calibration process was done using an
automated approach including a genetic algorithm (GA) and non-dominated
sorting genetic algorithm II (NSGA-II). Then the best-performing calibrated
models were run with and without considering water use to quantify the
impact of human water use. Observations encompass remote-sensing-based time
series of annual irrigated areas in the basin from MODIS, monthly total
water storage anomaly (TWSA) from GRACE satellites, and monthly lake volume
anomalies. In situ observations include time series of annual inflow into
the lake and basin averages of groundwater level variations based on 284
wells. In addition, local estimates of sectoral water withdrawals in 2009
and return flow fractions were utilized. Calibration against MODIS and GRACE data alone improved simulated inflow into Lake Urmia but inflow and lake volume loss were still overestimated, while groundwater loss was underestimated and seasonality of groundwater storage was shifted as
compared to observations. Lake and groundwater dynamics could only be
simulated well if calibration against groundwater levels led to an adjustment of the fractions of human water use from groundwater and surface
water. Thus, in some basins, globally available satellite-derived observations may
not suffice for improving the simulation of human water use. According to
WGHM simulations with 18 optimal parameter sets, human water use was the
reason for 52 %–57 % of the total basin water loss of about 10 km3 during 2003–2013, for 39 %–43 % of the Lake Urmia water loss of about 8 km3, and for up to 87 %–90 % of the groundwater loss. Lake inflow was 39 %–45 % less than it would have been without human water use. The study shows that even without human water use Lake Urmia would not have recovered from the significant loss of lake water volume caused by the drought year 2008. These findings can support water management in the basin and more specifically Lake Urmia restoration plans.
Type of Medium:
Online Resource
ISSN:
1607-7938
DOI:
10.5194/hess-24-1939-2020
DOI:
10.5194/hess-24-1939-2020-supplement
Language:
English
Publisher:
Copernicus GmbH
Publication Date:
2020
detail.hit.zdb_id:
2100610-6
Permalink