Diagnostic assessment of climatic and anthropogenic drivers of drying of Hamun lake

Abstract:

Hamun lake, situated on border of Iran and Afghanistan, completely dried up in 2003 for the first time in four decades and has transformed to an intermittent lake ever since. The lake is fed by the Helmand river sourced from snowmelt‐driven streams of Hindu Kush mountains, as well as monsoon-driven seasonal floods. The Helmand river (1,150 km) is regulated by two storage dams and is diverted along the way to support irrigated agriculture in its valley. In this study, we investigate anthropogenic and climate drivers that led to drying of Hamun lake. To analyze climatic drivers, we use monthly precipitation accumulation and monthly temperature data; and to account for anthropogenic impacts, we use monthly surface area of irrigated cropland. We use several precipitation datasets including the Climate Prediction Center (CPC), Global Precipitation Climatology Center (GPCC), and Climatic Research Unit (CRU) and a number of temperature daily/monthly reanalysis datasets such as ERA5, TerraClimate, and NEX-GDDP. For both precipitation and temperature datasets intercomparison of error for different products is done using the total error variance obtained from Triple Collocation Analysis (TCA). To create a monthly time series of surface area of irrigated croplands, we develop a Google Earth Engine App using temporal distribution of spectral properties of irrigated cops. This App uses both Landsat and MODIS imagery to delineate irrigated croplands, and a Markov Chain Monte Carlo (MCMC) within a Bayesian framework is used for filling gaps in time series of Landsat moderate-resolution imagery with MODIS coarse-resolution imagery. These drivers are then fed to a generalized additive model to capture relationships between the drivers and the surface area of the Hamun lake. We identify the causal relationships among the drivers and the lake, and quantify the role of anthropogenic forcing in drying of the Hamun lake. We also look into GRACE total water storage anomalies and deficits for possible shifts in trend of total water storage in this region. Further, we will use nighttime light imagery to see how population might be relocated in response to drought and drying of the lake.