[Dataset Title/Name]: Annual global land cover mapping at 30 m resolution from 1985 to 2022 with high temporal consistency

[Yelong-Z]

Contact Details

No response

Dataset description

This dataset is a global 30 meter resolution long-term time-series annual land cover dataset. This product is a long-term land cover mapping method based on the integration of multi temporal change detection algorithms and incremental dynamic updates. It combines multiple temporal change detection algorithm integration strategies, lightweight spatiotemporal fusion methods for key feature reconstruction, and a stable training sample transfer driven region adaptive random forest classification model to achieve high-precision annual land cover mapping. This dataset has the following advantages: (1) high classification accuracy, with an annual average overall accuracy of 78.63%; (2) The results of CCDC, BFAST Monitor, and LandTrendr mainstream change detection are integrated, and the Zou test is used to eliminate false change information. Therefore, the temporal consistency is good, which can support fine analysis of land cover change.
Can be downloaded at https://data.tpdc.ac.cn/en/disallow/5ece2a65-ce1f-4cb7-8a76-58a58eb4e33a
The data is in GeoTIFF raster format, with each file containing land cover classification results from 1985 to 2022 within a 1 °× 1 ° range. The file is named "AGLC_1985_2022_LonLat. tif", where "Lon" represents the longitude of the lower left corner of the image and "Lat" represents the latitude of the lower left corner of the image. At the same time, each file contains 38 bands named "land_comver_year", where year represents the year. The dataset contains 10 types of land cover categories, and the names and grayscale values of each category correspond to: cultivated land (10), forest land (20), grassland (30), shrub (40), wetland (50), water body (60), tundra (70), impermeable surface (80), bare land (90), and permanent ice and snow (100).

Earth Engine Snippet if dataset already in GEE

for example

var asset_name = ee.ImageCollection("path to your collection")

Sample Code: Add a sample code maybe just adding your datasets in the code editor

Enter license information

Enter License information ex: Attribution 4.0 International (CC BY 4.0)

Keywords

Land Use/Land Cover

Code of Conduct

• I agree to follow this project's Code of Conduct