Our PhD student Dan Kanmegne Tamga has published his first paper on “Modelling the spatial distribution of the classifcation error of remote sensing data in cocoa agroforestry systems” in cooperation with his supervisory team and World Agroforestry (ICRAF). This work has been performed in the frame work of WASCAL-DE-Coop.
From the abstract:
Cocoa growing is one of the main activities in humid West Africa, which is mainly grown in pure stands. It is the main driver of deforestation and encroachment in protected areas. Cocoa agroforestry systems which have been promoted to mitigate deforestation, needs to be accurately delineated to support a valid monitoring system. Therefore, the aim of this research is to model the spatial distribution of uncertainties in the classifcation cocoa agroforestry. The study was carried out in Côte d’Ivoire, close to the Taï National Park. The analysis followed three steps (i) image classifcation based on texture parameters and vegetation indices from Sentinel-1 and -2 data respectively, to train a random forest algorithm. A classifed map with the associated probability maps was generated. (ii) Shannon entropy was calculated from the probability maps, to get the error maps at diferent thresholds (0.2, 0.3, 0.4 and 0.5). Then, (iii) the generated error maps were analysed using a Geographically Weighted Regression model to check for spatial autocorrelation. From the results, a producer accuracy (0.88) and a user’s accuracy (0.91) were obtained. A small threshold value overestimates the classifcation error, while a larger threshold will underestimate it. The optimal value was found to be between 0.3 and 0.4. There was no evidence of spatial autocorrelation except for a smaller threshold (0.2). The approach differentiated cocoa from other landcover and detected encroachment in forest. Even though some information was lost in the process, the method is efective for mapping cocoa plantations in Côte d’Ivoire.