A research team, including a scholar from Near East University, has developed a method for drought monitoring in Iran by downscaling Terrestrial Water Storage Anomaly (TWSA) data from the GRACE satellite using deep learning techniques. This study is pivotal in addressing the challenges of sustainable water management, sustainable cities, and life on land, resonating with global sustainable development goals.

The study utilized the growing neural gas (GNG) method for clustering TWSA data and applied deep learning models, including Convolutional Long Short-Term Memory (ConvLSTM), to downscale this data to a 10 km spatial resolution. This approach is significant for countries like Iran, which face acute water-related issues and lack comprehensive ground observation data. The deep learning method notably outperformed traditional models in downscaling GRACE data and generating accurate groundwater storage maps.

This research is crucial for understanding and managing Iran’s water resources more effectively. The results indicate that Iran has been experiencing severe drought conditions since 2010, exacerbated by agricultural expansion and excessive groundwater withdrawal. The study highlights that meteorological drought, while not the sole cause, is a significant trigger for the country’s water crisis.

The findings emphasize the need for integrated sustainable water resources management in Iran. This involves updating water supply networks, improving irrigation efficiency, and implementing appropriate strategies in the agriculture sector. Such measures are essential for ensuring the availability and sustainable management of water and sanitation for all, making cities and human settlements inclusive, safe, resilient, and sustainable, and protecting, restoring, and promoting sustainable use of terrestrial ecosystems.

In conclusion, the study co-authored by a researcher from Near East University provides crucial insights into Iran’s long-term drought and water resource challenges. It underscores the importance of leveraging advanced technologies like deep learning for environmental monitoring and management, contributing significantly to global efforts in sustainable development and ecological conservation.

For further details, access the original paper from the publisher’s link:

https://www.sciencedirect.com/science/article/pii/S0022169422014081