This month, Philippe Rufin, Adia Bey, Michelle Picoli and Patrick Meyfroidt published the article "Large-area mapping of active cropland and short-term fallows in smallholder landscapes using PlanetScope data" in the International Journal of Applied Earth Observation and Geoinformation. The research focuses on a study area in Northern Mozambique, where they presented a framework for mapping active cropland and short-term fallows in highly fragmented smallholder landscapes. The results of this study aim to minimize financial constraints and maximize the replicability of the methods. All code and maps are made available for further use.

To conduct the study, they used Planet Scope mosaics available through the NICFI data and integrated them into Google Earth Engine. Their methodology consisted of derived seasonal analysis-ready datasets from coregistered time series of PlanetScope mosaics and used iterative learning to map active and fallow cropland. The authors mentioned that because of the Planet Scope images' high granularity and temporal precision, the resulting cropland map enables more precise estimates of actively used cropland. 

Additionally, the authors highlighted the importance of NICFI program. "The continuation of the NICFI data program will allow for the enhancement of the methods presented here in terms of disentangling short-term and long-term fallows and providing detailed insights into year-to-year change processes." 

Click here to read the full article. 

The data and code is made available at https://doi.org/10.5281/zenodo.6907606 and https://github.com/philipperufin/eepypr/