Rapid urbanization increasingly influences ecosystem stability and regional urban sustainability. Pan-Third Pole cities located in high-altitude regions around the Tibetan Plateau, Pamir, and Indo Kush have experienced rapid urbanization during the last few decades; however, an understanding of the urbanization rate and its driving mechanism remains lacking. Here, we investigate the urban land transformation dynamics of 14 major Pan-Third Pole cities and analyse their driving factors over two periods (2000-2010 and 2010-2017) based on the interpretation of historical Google Earth imagery using deep learning techniques. Our results show that the urbanized area has increased from 2809 km2 to 4380 km2, with an annual growth rate of nearly 3% during 2000-2017, which is probably the largest increase in the world. The analysis of potential driving factors reveals that population growth dominates urban expansion, elevation is a constrained topographical factor, and neighbourhoods to roads facilitate urbanization. Our findings would benefit policy making for urban sustainability in the most fragile region on Earth.
Keywords: Deep learning; Driving factors; Environment impacts; Pan-Third Pole; Urban expansion.
Copyright © 2020 Elsevier B.V. All rights reserved.