Guo et al. (2025) A three-decade lake dataset on the Mongolian plateau tracking water area and quality dynamics (1990–2020)

Identification

• Journal: Scientific Data
• Year: 2025
• Date: 2025-11-14
• Authors: Junjiang Guo, Kai Liu, Jiaming Na, Ge Liu, Zhigang Cao, Chenyu Fan, Bin Xue, Junchuan Huang, Chunqiao Song
• DOI: 10.1038/s41597-025-06059-5

Research Groups

• College of Civil Engineering, Nanjing Forestry University, China
• State Key Laboratory of Lake and Watershed Science for Water Security, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, China
• University of Chinese Academy of Sciences, Nanjing (UCASNJ), China
• Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, China
• University of Chinese Academy of Sciences, Beijing, China
• Research Center of Applied Geology of China Geological Survey, China

Short Summary

This study developed the Mongolian Plateau Lake Dataset (MPLD), the first open-access, long-term (1990-2020) dataset of lake area and water quality (Secchi Disk Depth, Total Suspended Matter, and Forel–Ule Index) for 1,161 lakes (> 1 km²) on the Mongolian Plateau, revealing initial shrinkage followed by partial recovery and widespread eutrophication.

Objective

• To provide a comprehensive understanding of the hydrological and water quality dynamics of lakes (including artificial reservoirs) on the Mongolian Plateau through long-term remote sensing satellite observations.
• To develop a robust monitoring framework for identifying spatial distribution and maximum extents of 1,161 lakes, and generating annual time series data for Secchi Disk Depth (SDD), Total Suspended Matter (TSM), and Forel–Ule Index (FUI).

Study Configuration

• Spatial Scale: Mongolian Plateau (37.25°–53.10°N, 87.50°–126.07°E), approximately 2.75 million km², covering 1,161 lakes (including artificial reservoirs) with a surface area greater than 1 km². Data provided at 30-meter spatial resolution.
• Temporal Scale: Annual records from 1990 to 2020 (31 years).

Methodology and Data

• Models used:
  • Lake Area: JRC Global Surface Water (GSW) dataset, refined with GLAD dataset and manual inspection.
  • Water Masking: Modified Normalized Difference Water Index (MNDWI) with a threshold of 0.
  • Secchi Disk Depth (SDD): Empirical model based on Landsat red band reflectance (Rs(R)): SDD = 0.46 * (Rs(R))^-1.26.
  • Total Suspended Matter (TSM): Exponential model based on the ratio of Landsat red to blue band reflectance (Rs(R)/Rs(B)): TSM = 8.0047 * e^(0.0933 * (Rs(R)/Rs(B))).
  • Forel–Ule Index (FUI): Calculated using the CIE XYZ color system, converting visible light RGB bands to CIE-XYZ tristimulus values, then to chromaticity coordinates (x, y), chromaticity angle (α), and finally mapped to FUI levels via a polynomial function and look-up table.
• Data sources:
  • Satellite Imagery: Landsat Collection 1 Tier 1 surface reflectance (SR) and top-of-atmosphere (TOA) datasets from Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI). Processed via Google Earth Engine (GEE).
  • Surface Water Datasets: Global Surface Water (GSW) dataset and Global Land Analysis and Discovery (GLAD) dataset.
  • Validation Data: High-resolution PlanetScope CubeSat satellite imagery (approximately 3 meters) for lake area validation. In-situ measurements of TSM and SDD from 19 lake systems in Inner Mongolia (2013–2020) provided by Nanjing Institute of Geography and Limnology and Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences. Benchmarked against NOAA CoastWatch’s Kd(490) product and Jiang et al. (2021) global TSS remote-sensing dataset.

Main Results

• The Mongolian Plateau Lake Dataset (MPLD) was developed, providing annual records (1990–2020) of lake extent and three water quality indicators (SDD, TSM, FUI) for 1,161 lakes (> 1 km²).
• Total lake area on the Mongolian Plateau exhibited an initial shrinkage followed by partial recovery, reaching 19,592.78 km² by 2020.
• By 2020, 71.83% of the monitored lakes were classified as eutrophic (FUI ≥ 10).
• Significant spatiotemporal heterogeneity was observed: lakes in western and northern regions expanded and showed improved water clarity, while those in eastern and central Inner Mongolia experienced contraction and water quality decline.
• Lake area validation against manually digitized data showed high accuracy with a Mean Absolute Percentage Error (MAPE) of 5.1% and a Root Mean Squared Error (RMSE) of 7.25 km².
• Water quality model validation demonstrated robust performance:
  • SDD model: R² = 0.83, RMSE = 7.22 cm, MAPE = 5.05%.
  • TSM model: R² = 0.74, RMSE = 18.04 mg/L, MAPE = 36.82%.

Contributions

• Provides the first open-access, high-precision, and spatially explicit integrated dataset of lake area and water quality dynamics for the entire Mongolian Plateau, filling critical spatiotemporal data gaps.
• Establishes a solid data foundation for comprehensive evaluation of lake water resources, supporting regional water resource management, climate change impact studies, and dryland ecosystem dynamics research.
• Enables integrated, long-term monitoring of hydrological and water quality changes, facilitating the precise identification of ecologically vulnerable lakes and informing conservation and sustainable management policies.
• Bridges the monitoring gap for small lakes (< 10 km²) and offers continuous temporal data, overcoming limitations of previous interval-based studies.

Funding

• National Key Research and Development Program of China (Grant No. 2022YFF0711603)
• National Key Research and Development Program of China (Grant No. 2019YFA0607101)
• Science and Technology Planning Project of NIGLAS (Grant No. 2022N-IIGLAS-CJH04)
• National Natural Science Foundation of China (Grant No. 42271450)
• Excellent Scientific Research and Innovation Team of Universities in Anhui Province (Grant No. 2023AH010071)

Citation

@article{Guo2025threedecade,
  author = {Guo, Junjiang and Liu, Kai and Na, Jiaming and Liu, Ge and Cao, Zhigang and Fan, Chenyu and Xue, Bin and Huang, Junchuan and Song, Chunqiao},
  title = {A three-decade lake dataset on the Mongolian plateau tracking water area and quality dynamics (1990–2020)},
  journal = {Scientific Data},
  year = {2025},
  doi = {10.1038/s41597-025-06059-5},
  url = {https://doi.org/10.1038/s41597-025-06059-5}
}