The limited anthropogenic activities on the Tibetan Plateau make this an ideal natural laboratory to elucidate how climate change impacts lake changes. Previous studies have mainly focused on decadal lake changes, yet their rapid evolutions at short temporal intervals and the associated atmospheric origins remain elusive. Here, we produce a new lake area change dataset at monthly sampling over 2015-2020 from 16,801 satellite images. Our estimates achieve an accuracy of <30 m, as evidenced by in-situ GPS field survey validations of representative lake shorelines. We found contrasting patterns in recent rapid area changes: deaccelerating in the north and accelerating in the south. Such contrasting pattern was unprecedented in the last two decades and is likely caused by recent precipitation anomalies, indicating that lakes in TP may experience a tipping point. Lakes are found to store only a small portion (<5 %) of net precipitation in summer, increased to ∼11 % for years with heavy precipitation, which helps understand the water mass budget for lakes over there. Our study highlights the importance of investigating short-term lake area changes as a climate proxy to study their rapid responses to intra- and inter-annual climate variability.
Keywords: Acceleration; Deep learning; Field campaign; Lake area; Summer expansion.
Copyright © 2023. Published by Elsevier B.V.