Mozambique 10 m resolution cropland map for 2017 to 2019

Description

The classification was carried out on Google Earth Engine (GEE) platform using random forest classifier. Big data analysis and cloud computing techniques were employed for analysis of massive volume of multi-source remote sensing data and thousands of samples. All available Sentinel 2, and Chinese Satellite Gaofen-1 and 2 high resolution imageries covering Mozambique over the last 3 years were integrated for this research. Seasonal (dry season, & rainy season) composite, annual composite, composition based on percentiles, as well as greenest composite and other methods were applied to extract informative layers for classification. A variety of features collections such as spectrum, phenological, and texture information were jointly utilized for the 10m resolution national cropland mapping for Mozambique. The training of the classifiers was carried at both locally and nationally.

This work is jointly done by CropWatch team in Aerospace Information Research Institute, Chinese Academy of Sciences (AIRCAS) and Catholic University of Mozambique (UCM). This scientific output is jointly supported by the Big Earth Data Programme (CASEarth) and Dryland Ecosystem International Programme (Global-DEP) of Chinese Academy of Sciences, Key Research & Development Program for Global Change and Adaptation of Ministry of Science and Technology, and the international cooperation project under National Natural Science Foundation – United Nations Environmental Programme (NSFC - UNEP).

Technical info (English)

Dataset users shall clearly indicate the source of the "Mozambique 10-meter resolution cropland product" in the research output including published or unpublished papers, thesis, reports, dataset and other academic output. Data producers are not responsible for any losses caused by the use of the data. The boundaries and marks used in the maps do not represent any official endorsement or opinion by the data producer.

Users shall also add the following references when using the dataset:

Bofana, J., Zhang, M., Nabil, M., et al. (2020). Comparison of different cropland classification methods under diversified agroecological conditions in the Zambezi River Basin. Remote Sensing, 12(13), 2096.

Files

Additional details