GLCD250-MOD: Global Lakes Chlorophyll-a Daily 250 m based on MODIS — Part 3 of 3 (Lake IDs 2,001–1,427,688)

GLCD250-MOD is a global dataset of chlorophyll-a (Chl-a) concentrations for inland lakes.

Source
Zenodo
Published
2026
Region
Global
Format
multiple formats
Access
Open access
License
cc-by-4.0
Themes
Remote Sensing & Satellites, Fresh Water, Geospatial & GIS Data, Machine Learning & AI

Get the data from Zenodo

About this dataset

GLCD250-MOD is a global dataset of chlorophyll-a (Chl-a) concentrations for inland lakes. It provides daily Chl-a estimates at a nominal spatial resolution of 250 m over the period 2000–2024. Chl-a concentration is estimated by a trained transformer-based hierarchical deep learning model based on MODIS satellite observations. With its high-resolution Chl-a estimates, GLCD250-MOD improves data availability for small lakes compared to the MODIS Terra Level-3 product (4.6 km spatial resolution, daily). GLCD250-MOD covers 465,966 lakes selected from the HydroLAKES database. The dataset is expected to support analysis of algal bloom dynamics, enabling monitoring of spatiotemporal bloom patterns and detection of long-term changes in inland lake ecosystems.

GLCD250-MOD provides the variable ' chlor_a ', representing chlorophyll-a concentration in mg m⁻³. Each NetCDF file includes daily time coordinates, MODIS sinusoidal X and Y coordinates in metres, and metadata describing the lake identifier, temporal coverage, and model provenance. Missing predictions are represented by the fill value −999.

The complete dataset is distributed across three Zenodo records using independently extractable Zstandard-compressed TAR archives (.tar.zst). This record contains Part 3, covering Lake IDs 2,001–1,427,688, with 463,968 NetCDF files for 463,968 lakes. The total compressed size is approximately 26.937 GB .

Lakes with Hylak_id: 2,001–10,000. These lakes are provided in a single archive, 0002001-0010000.tar.zst , with one NetCDF file per lake.

Lakes with Hylak_id: 10,001–200,000. These lakes are organized into 19 archives, each covering a range of 10,000 IDs, from 0010001-0020000.tar.zst to 0190001-0200000.tar.zst . Each lake is represented by one NetCDF file.

Lakes with Hylak_id: 200,001–1,427,688. The remaining lakes are provided in a single archive, 0200001-1427688.tar.zst , with one NetCDF file per lake.

Each file covers the period from 24 February 2000 to 31 December 2024, with predictions available where valid satellite inputs exist. Each archive contains a top-level directory matching its filename without the .tar.zst extension. Lake-ID ranges in archive filenames include both endpoints. These ranges specify file allocation; they do not imply that every integer ID has an associated file throughout the complete dataset.

The remaining data are available in Part 1 (Lake IDs 1–50) and Part 2 (Lake IDs 51–2,000) .

This dataset is developed and maintained by the Water Environmental Management Lab at the University of Seoul. For inquiries or technical support, please contact YoonKyung Cha ( ). The dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

For further methodological details and background, users are encouraged to consult the following reference:

Yang S., Lee H., Lee G., Gu T., Park T., Park J., Shin J., Kim T., Cha Y. (2026).

How the data was collected

GLCD250-MOD is a global dataset of chlorophyll-a (Chl-a) concentrations for inland lakes. It provides daily Chl-a estimates at a nominal spatial resolution of 250 m over the period 2000–2024. Chl-a concentration is estimated by a trained transformer-based hierarchical deep learning model based on MODIS satellite observations. With its high-resolution Chl-a estimates, GLCD250-MOD improves data availability for small lakes compared to the MODIS Terra Level-3 product (4.6 km spatial resolution, daily). GLCD250-MOD covers 465,966 lakes selected from the HydroLAKES database. The dataset is expected to support analysis of algal bloom dynamics, enabling monitoring of spatiotemporal bloom patterns and detection of long-term changes in inland lake ecosystems.

GLCD250-MOD provides the variable ' chlor_a ', representing chlorophyll-a concentration in mg m⁻³. Each NetCDF file includes daily time coordinates, MODIS sinusoidal X and Y coordinates in metres, and metadata describing the lake identifier, temporal coverage, and model provenance. Missing predictions are represented by the fill value −999.

The complete dataset is distributed across three Zenodo records using independently extractable Zstandard-compressed TAR archives (.tar.zst). This record contains Part 3, covering Lake IDs 2,001–1,427,688, with 463,968 NetCDF files for 463,968 lakes. The total compressed size is approximately 26.937 GB .

Lakes with Hylak_id: 2,001–10,000. These lakes are provided in a single archive…

Lakes with…

La…

Browse sustainability datasets on SusanHub