Assessing the impact of dams and unsustainable agricultural development on Lake Urmia drying up using multi satellite sensor and deep learning

Lake Urmia, the largest inland lake in Iran, has been drying up in recent years. This research aims to investigate the cause of Lake Urmia’s drying using multiple satellite sensors and deep learning, providing novel insights into both anthropogenic and natural factors driving its decline. Lake Urmia ecosystem and area changes was conducted using Landsat images and the long short-term memory (LSTM) method. In late 2015, the lake reached its lowest area of 1962 km² due to the reduction in the amount of water entering the lake caused by the operation of approximately 34 dams. Based on LSTM’s analysis, it has been identified that the agricultural area increased between 1985 and 2021. Groundwater storage changes for the Urmia catchment was calculated using the gravity recovery and climate experiment satellites as well as a hydrological model, which showed a decreasing trend of 0.43 ± 0.1 cm/year. The drying of Lake Urmia is attributed to three primary factors: the unsustainable development of the agricultural sector, the construction of several dams around the lake that exploit the basin’s water, and climate change intensification coupled with drought occurrence, which are among the other two factors.