Linked to the research conducted under the Methods and Tools Component of the 50x2030 Initiative (https://www.50x2030.org/), this data deposit includes 10-meter spatial resolution maps for (i) areas cultivated with any crops, and (ii) areas cultivated with maize across Malawi for each rainy season during the period of 2016-2019. The maps are a product of the analyses conducted by Azzari et al. (2021), as part of the collaboration between the World Bank and Atlas AI, in support of one of the objectives of the 50x2030 Initiative to create guidelines for the collection of minimum-required survey data for training and validating remote sensing models for high-resolution crop type mapping and crop yield estimation. Azzari et al. (2021) integrate Sentinel-2 satellite imagery and complementary geospatial data with georeferenced plot-level data from national household surveys that were conducted by the Malawi National Statistical Office and the Central Statistical Agency of Ethiopia during the period of 2018-2020 in order to identify the optimal approach to collecting survey data for training a machine learning model to identify areas cultivated with maize. The best performing model estimated by Azzari et al. (2021) has been used to generate the 10-meter spatial resolution maps that are being made available here. For more information, please see the accompanying Basic Information Document and Azzari et al. (2021).
Reference: Azzari, G., Jain, S., Jeffries, G., Kilic, T., and Murray, S. (2021). "Understanding the Requirements for Surveys to Support Satellite-Based Crop Type Mapping: Evidence from Sub-Saharan Africa." World Bank Policy Research Working Paper No. 9609, LSMS Washington, DC: World Bank.