A groundbreaking study by Professor Di Long and his team at Tsinghua University’s Department of Hydraulic Engineering has unveiled that seasonality is the dominant factor shaping global lake-surface-extent dynamics. Published in Nature on May 28, 2025, the research marks a major advancement in the integration of artificial intelligence (AI) with remote sensing technologies for environmental monitoring.
Using an innovative deep-learning-based spatiotemporal fusion model, the research team successfully monitored nearly 1.4 million lakes worldwide on a monthly basis at 30-meter spatial resolution—a level of detail never before achieved on a global scale.
Why Seasonality Matters in Lake Dynamics
While previous studies have focused primarily on long-term and interannual lake changes, this research reveals that seasonal variations are the primary drivers for 59% of global lakes, accounting for approximately 66% of total global lake area. Even more striking, over 90% of the global population lives within drainage basins where lake dynamics are seasonally dominated. This correlation underlines the strong connection between human activity and seasonal hydrological cycles.
Seasonality-related extremes are not only intensifying but also increasingly capable of overturning long-term hydrological trends, posing risks to ecosystem adaptability, habitat stability, and water security. The findings offer a robust foundation for more accurate estimation of lake-related greenhouse gas emissions and provide scientific insights essential for managing extreme hydrological events.
How the Research Was Done: AI Meets Remote Sensing
Professor Long’s team addressed the long-standing technical gap in balancing high spatial resolution with frequent temporal updates in global lake monitoring. Their deep-learning model integrates the frequent temporal resolution of MODIS with the high spatial precision of the Global Surface Water (GSW) dataset, originally published in Nature in 2016.
This fusion led to the most comprehensive and accurate global lake dataset to date:
- 93% user accuracy, 96% producer accuracy
- Missing data reduced from 34% (GSW) to just 1.2%
- Continuous monthly observations for over 1.4 million lakes
Scientific & Global Impact
This pioneering dataset transforms global lake research from static yearly snapshots to dynamic, high-frequency environmental monitoring, advancing our understanding of how climate change and human activity shape freshwater resources.
It also provides actionable data for global stakeholders involved in climate adaptation, ecosystem conservation, and water management, while pushing the boundaries of AI-driven environmental science.
About the Authors & Publication
📄 Paper title: “Global dominance of seasonality in shaping lake-surface-extent dynamics”
📅 Published in: Nature, May 28, 2025
🔬 Lead author: Luoqi Li (Ph.D. candidate)
👨🏫 Corresponding author: Professor Di Long (Tsinghua University)
🤝 Collaborators: Yiming Wang (Tsinghua), Prof. R. Iestyn Woolway (Bangor University, UK)
🎓 Supported by: National Natural Science Foundation of China, Second Tibetan Plateau Scientific Expedition and Research (STEP) program
📘 Read the full paper here:
👉 Nature Article
🔗 Tsinghua Source
Keywords:
- global lake surface dynamics
- seasonality in hydrology
- Tsinghua University lake research
- deep learning satellite data fusion
- AI in environmental monitoring
- lake surface extent changes
- Nature lake surface study
- seasonal hydrological cycles
- lake monitoring with remote sensing
- global water resources AI
Tsinghua University reveals seasonality’s global dominance in lake surface dynamics with AI-based satellite data fusion. Published in Nature.
Abroad Africa AI Beijing Belt & Road BLCU BRICS China chinese CPC CSC Culture Economy education Iraq Korea Kültür Langauge movie Russia scholarship science Shanghai Sino Sino Turkish Sino Turkish Sino Turkish Sino Turkish Sino Turkish Studies Sino Turkish Studies Sino Turkish Studies Sino Turkish Studies space Syria Taiwan Tariff trump Turkiye Türkiye University USA Xi Jinping Xinjiang ZJUT Çin
