Monitoring the dynamics of two Faidherbia albida parklands in North Cameroon by analysis of Google Earth images ; Seguimiento de la dinámica de dos parques de Faidherbia albida en el norte de Camerún, mediante el análisis de imágenes de Google Earth ; Suivi de la dynamique de deux parcs à Faidherbia albida du Nord-Cameroun, par analyse d’images Google Earth

In the Sudano-Sahelian zone of North Cameroon, the Développement Paysannal et Gestion de Terroir project has encouraged the restoration of Faidherbia albida agroforestry parks by subsidising their assisted natural regeneration (ANR), starting in 1994, through a levy on cotton revenues. In the following years, the Cameroonian (IRAD) and French (CIRAD) agricultural research institutes joined forces to study services and production in these parks, as well as their restoration dynamics, with the support of the Paris Institute of Technology for Life, Food and Environmental Sciences and the PRASAC project. Monitoring continued until 2012, through tree pruning trials, inventories and surveys. But since the Boko Haram security crisis (2015), access to the land has become difficult. This study aims to determine whether monitoring of tree cover change in two of these parklands would be possible by semi-automatic photo-interpretation of Google Earth images taken during the dry season, based on an object-oriented analysis, and how reliable this would be. Our comparison of data from an inventory carried out in these parks in 2012 with data from the interpretation of Google Earth images from 2013 shows that the latter are very inaccurate for estimations of the number of trees, particularly trees with a crown area of 2to100m2 and especially for F. albida, which lose some of their foliage very irregularly. In the Sirlawé and Gané areas, the average tree cover more than doubled between 2009 and 2018, from 246 to 587m2/ha (2.5% to 5.9%) and from 245 to 575m2/ha (2.5% to 5.8%), respectively. This increase in tree cover is mainly due to small trees. This study provides a basis for more accurate inventories with other types of satellite images, for comparing change in tree cover in villages benefiting or not benefiting from the subsidized ANR policy and for determining farmers' perceptions of the value of the ANR support policy and sustainable management of these agroforestry systems. ; En la zona sudanosaheliana del norte de ...