Vegetation cover changes analysis at the Cameroonian shores of Lake Chad and its hinterland (2000-2018) - Archive ouverte HAL Accéder directement au contenu
Poster De Conférence Année : 2019

Vegetation cover changes analysis at the Cameroonian shores of Lake Chad and its hinterland (2000-2018)

Résumé

Since the great droughts of the 1970s, the monitoring of vegetation cover in arid regions in general, and Sahelian in particular is a major environmental issue. Indeed, the 1970s and 1980s were marked in the Sahel region by severe droughts accompanied by significant environmental and social consequences that have accelerated the process of the Lake Chad drying up. The improvement of climatic conditions observed since the 1990s started the period of Sahel “regreening” observed by coupling weather records with satellite images NOAA (1981) and MODIS (2000). However, observations carried out at local scales show trends of evolution sometimes different from the results obtained in the Sahel. This study explores the quantification of changes in vegetation cover at the Cameroonian shores of Lake Chad and its hinterland. Its two complementary goals are the mapping of vegetation cover; and the statistical changes analysis to assess the degree of correlation between visual, quantified trends, and environmental parameters. The overall methodology combines remote sensing and statistical analyses in three steps with three complementary results. From the outset, vegetation cover is mapped using modis images (mod13q1) from 2000 to 2018. Then the variation of the said vegetation cover is assessed via the tome series method. Precisely, the test associates mann-Kendall's correlation with the significance test measured for Pvalue <0.001, and determines vegetation cover changes according to three alternative hypotheses: negative trend, positive trend and non-zero. Across the study area, stable trends occupy 65.08% negative trends 21.61% and positive trends 13.29%. In addition, a principal component analysis (PCA) is applied to the MODIS images to study seasonal and annual variations in chlorophyll activity. The first component describes seasonal contrasts between areas of low NDVI value (bare soils and permanent open water), and areas with high NDVI value (permanent wetlands). The second component contrasts the areas of low seasonal variation of NDVI located in the lake area and the areas with significant variations, particularly in the hinterland. Distinction between the vegetative variations related to the rains and the vegetation activity related to the variations of the lake is contained in the component three, which also differentiates areas with progressive evolution from those regressive evolution. Finally, to identify the environmental factors related to the different trends, the mann-Kendall map is reclassified into 5 classes, and crossed with the map of the soils potentiality which divides the area into landscape units, and describes them using the quality of the soil drainage. Among the landscape units covering the study area, only that of Makari well-drained, is dominated by stable and progressive trends (83.1% of its surface). The two other landscape units important in terms of area namely Bodo Kouda and Lake Chad, are poorly drained and dominated by stable and regressive trends corresponding to 87.47% and 93.3% of their respectively surface. Furthermore, The Mann-Kendall’s map is crossed with the phytogeography map. It appears that, among the 18 vegetation groups, the "vegetation of Lake Chad" presents the most important trends in terms of area. The regressive classes occupy 40%, the stable class corresponds to 36.2% and the progressive classes accumulate 23.88%. The processing performed and the results obtained have shown the advantages of well correlating visual maps with statistics trends/graphs and environmental parameters. Indeed, Lake Chad and its immediate shores were subject to positive and negative trends, and the hinterland around Cameroon were subject to stable and negative trends particularly in poorly drained area. Then, the vegetation cover variation and its spatial evolution are better interpretable when combining time series and remote sensing separately as tested in the period 2000-2018.
Fichier principal
Vignette du fichier
Poster_ILUS2019_Gbetkom_Gadal.pdf (1.06 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02396450 , version 1 (06-12-2019)

Identifiants

  • HAL Id : hal-02396450 , version 1

Citer

Paul Gérard Gbetkom, Sébastien Gadal, Ahmed El Aboudi, Julien Andrieu, Alfred Homère Ngandam Mfondoum, et al.. Vegetation cover changes analysis at the Cameroonian shores of Lake Chad and its hinterland (2000-2018). 3rd International Land Use Symposium 2019: Trends and projections, Dec 2019, Paris, France. , 2019. ⟨hal-02396450⟩
92 Consultations
43 Téléchargements

Partager

Gmail Facebook X LinkedIn More