The global surface water range of lakes/reservoirs is a fundamental input data for many studies. Although there are currently some datasets, due to various reasons such as different data sources and dynamic changes in surface water, there are problems such as incompleteness or spatial inconsistency between datasets. Here, based on HydroLAKES, GRanD, and OpenStreetMap, a new global lake/reservoir water surface range dataset (GLRSED) has been created, which includes the spatial range and basic attributes (such as name, area, lake type, and source) of 2.17 million lakes/reservoirs. By spatially overlaying GLRSED with other auxiliary data, we identified mountainous lakes, inflow lakes, reservoirs, glacial lakes, and permafrost lakes. In addition, we also calculated the surface water and ocean topography (SWOT) trajectories passing through each lake. This dataset can provide basic data for global lake/reservoir monitoring and can also be used to study the impact of human activities and climate change on freshwater supply in lakes/reservoirs.
1. The HydroLAKES dataset is a combination of multiple data sources, including topographic maps and remote sensing datasets. It contains 1.43 million global lakes and reservoirs with an area greater than 0.1 square kilometers, and its attributes include surface area, perimeter, average depth, and volume. HydroLAKES is one of the most comprehensive and widely used datasets. The website is http://www.hydrosheds.org ；
2. OSM is based on geographic information collected and updated by volunteers. The data source comes from devices such as the Global Positioning System (GPS) and cadastral data, as well as manually digitizing medium to high-resolution satellite and aerial images, with global coverage and renewability. You can obtain it from the following website: https://www.openstreetmap.org/ ；
3. GRanD 1.3 version contains 7250 records of reservoirs and their related dams, with attributes including name, spatial coordinates, surface area, storage capacity, dam height, construction year, primary use, etc. The website is: http://sedac.ciesin.columbia.edu/pfs/grand.html .
Firstly, the data is preprocessed by extracting lakes and reservoirs from all types of water in the OSM for cleaning. The final processed OSM contains data from approximately 850000 lakes in total.
The preprocessing process for HydroLAKES and GRanD involves partitioning using HydroBASINS for subsequent processing. Then, HydroLAKES, OSM, and GRanD were integrated and processed in different watersheds, and spatially overlapped with other auxiliary data and GRanD. Use the "Endo" attribute in the 12th level dataset of HydroBASINS to analyze the internal flow of lakes. Endo "attribute variable, i.e.>0.
Use SWORD to identify lakes located on rivers. In addition, by aligning our dataset with the SWORD orbit, we calculated the number of orbits passing through each lake.
Mark the above lake types as attributes of the dataset and calculate the area and shoreline attributes of each object through geographic calculations. Meanwhile, the ID attributes of HydroLAKES and GRanD are retained. When merging two or more objects, only the first object is retained.
The above process is completed using ArcGIS.
During the process of developing the database, data discrepancies and record gaps were corrected. However, due to the lack of information on many lakes/reservoirs (such as small and medium-sized lakes/reservoirs), the database is still incomplete.
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:11738.11.NCDC.ZENODO.DB6645.2024
10.5281/zenodo.8121174