Estimation of the Carbon Sequestration Dynamics of Senegal’s Great Green Wall Based on Land Cover over the Past Three Decades

The severe drought observed in the Sahel during 1970s, 1980s and 1990s has deeply affected the population as well as the economies and the eco-systems of this climatic area. The GGW Initiative spearheaded by Africa Union in 2007 proposed to combat the land degradation and desertification by planting a wall of trees stretching from Dakar to Djibouti. A reforestation was then conducted in the Senegal’s GGW since 2006 as part as other areas in the Sahel. This paper aims to evaluate the carbon sequestration dynamics in the sites of the Senegal’s GGW over the last three decades. The method consists firstly of analyzing the evolution of land cover and land use dynamics based on ESA-CCI LC satellite data. There is an improvement of the surface areas of tree and shrub savanna of 11.40% (Tessekere), 8.25% (Syer) and 2.70% (Loughere-Thioly). The regreening of the different localities and a positive dynamic observed is explained by the return to normal rainfall and to reforestation actions, agroforestry practices, better management of natural resources undertaken. However, some non-reforested sites showed an opposite trend despite of the normal rainfall. Secondly, the results on land mapping are used as a proxy for the assessment of carbon stocks. The dynamic observed in vegetation cover since the beginning of the reforestation made it possible to sequester 5.8 million tons of carbon representing respectively 2.31% of African GGW. This gain in stored carbon is equivalent to 21.2 million tons of CO2 captured in the atmosphere. Through this study, it appears that carbon storage becomes significant 8 to 10 years after the start of reforestation. An urbanization without respect for the environmental factors could be a danger for the climate (case of Ballou).

Deep neural network ensembles for remote sensing land cover and land use classification

With the advancement of satellite technology,a considerable amount of very high-resolution imagery has become available to be used for the Land Cover and Land Use(LCLU)classification task aiming to categorize remotely sensed images based on their semantic content.Recently,Deep Neural Networks(DNNs)have been widely used for different applications in the field of remote sensing and they have profound impacts;however,improvement of the generalizability and robustness of the DNNs needs to be progressed further to achieve higher accuracy for a variety of sensing geometries and categories.We address this problem by deploying three different Deep Neural Network Ensemble(DNNE)methods and creating a comparative analysis for the LCLU classification task.DNNE enables improvement of the performance of DNNs by ensuring the diversity of the models that are combined.Thus,enhances the generalizability of the models and produces more robust and generalizable outcomes for LCLU classification tasks.The experimental results on NWPU-RESISC45 and AID datasets demonstrate that utilizing the aggregated information from multiple DNNs leads to an increase in classification performance,achieves state-of-the-art,and promotes researchers to make use of DNNE.