Lophira lanceolata est une espèce ligneuse autochtone à usages multiples, utile pour la stabilité des écosystèmes et essentielle pour la subsistance des communautés locales. Face aux effets néfastes de la péjoration climatique sur les habitats naturels, cette étude vise principalement à analyser les divergences entre cinq algorithmes de modélisation de niche écologique et leurs implications pour l'évaluation des habitats favorables de Lophira lanceolata au Burkina Faso. Les projections de la distribution potentielle actuelle et future aux horizons 2050 et 2070 sont utilisées comme support pour comparer la sensibilité des modèles selon différents scénarios climatiques. Les cinq méthodes de modélisation adoptées sont : Generalized Linear Model, Generalized Additive Model, Random Forest, Maximum Entropy et Boosted Regression Trees. Les données d'occurrence de l'espèce, obtenues des bases de Global Biodiversity Information Facility et de Rainbio, ont été filtrées puis croisées avec des variables environnementales sélectionnées sur la base de leur faible colinéarité. Les résultats ont montré que les modèles évalués à l'aide des indices Area under the curve, True skill statistics et kappa sont satisfaisants avec une forte variabilité spatiale selon les algorithmes. Random Forest s'est montré très restrictif, tandis que Boosted Regression Trees a révélé les surfaces les plus favorables. Les précipitations annuelles ont été identifiées comme la variable la plus influente. Dans les aires protégées, Generalized Linear Model et Generalized Additive Model projettent des superficies favorables plus importantes que Maximum Entropy et Random Forest. Les projections révèlent d'importantes divergences dans la distribution future de l'espèce : Random Forest prédit un net déclin des habitats très favorables, Maxent révèle une sensibilité marquée aux scénarios climatiques, alors que Boosted Regression Trees indique une stabilité. Generalized Linear Model et Generalized Additive Model suggèrent une expansion des zones favorables, en particulier sous le scenario MIROC6. Ces divergences parfois radicales selon les algorithmes confrontés, montrent les limites d'un usage non critique des outils de modélisation de niche écologique et appellent à une lecture prudente de leurs sorties pour guider les priorités de conservation de Lophira lanceolata face au changement climatique.
Les travaux sur les brises de mer et de terre dans la région de la Corne de l'Afrique sont rares. À partir de données météorologiques horaires (2010-2021) et de l'imagerie satellitaire (NOAA-AVHRR) disponible pour la période 2014-2022, la présente étude apporte des informations originales sur les caractéristiques de ces brises dans l'agglomération de Djibouti. La vitesse moyenne des brises varie entre 2 et 8 m/s et leur direction dominante alterne de l'E (le jour) à l'O (la nuit). Leur fréquence oscille de 66 % en été à 93 % pendant la saison moins chaude. La baisse relative de la fréquence des brises en été s'explique par l'impact du vent synoptique, plus actif durant cette saison (jusqu'à 13 m/s en juin, puis 16 m/s en juillet et août). Le contraste thermique entre le continent et la mer étant plus fort pendant le jour, les brises diurnes (de mer) ont des vitesses plus élevées que celles nocturnes (de terre), qui ne dépassent pas 3 m/s.
Observations made in northern Gaspé since the early 1980s show that niveo-aeolian processes can play a very important role in the dynamic of scree slopes composed of small fragments. During blizzards, characterized by winds that can exceed 100 km/h, wind gusts can displace debris ranging in size from gravel to small pebbles. The sediments displaced by the wind originate from the summit rockwall and the scree slope itself when it is not entirely covered by snow. Depending on snow conditions, snow can play a role in two ways: 1/ when it is blown along with sediments, it produces mixed stratified deposits, whose mineral content consists mainly of silt and sand; 2/ when snow is covered by a layer of ice, it acts as a conveyor belt for particles the size of gravel and small pebbles that are blown onto the ice by wind gusts. Wind can thus contribute to the formation of long stratified basal concavities characterized by reverse grain sorting, with the concavity composed of finer debris than the scree slope. A significant portion of the sediments mobilized by the wind is blown into the adjacent forest. The intensity of niveo-aeolian processes varies considerably on a regional scale. They are highest near the coast, where annual sedimentation rates range from few millimeters to few centimeters, and decrease into the valleys. These dynamics are favored by windy climate, heavy snowfalls, frequent freeze-thaw cycles, and frequent rain-on-snow events. Given their rapidly changing weather conditions over relatively short periods of time, cold temperate and boreal maritime climates seem to be the most conducive to this types of dynamic.
This study analyses spatial changes and landscape fragmentation in western Burkina Faso between 1990 and 2022, using Landsat TM (1990 and 2007) and OLI (2022) satellite imagery. Supervised classifications, performed with the Random Forest algorithm, achieve high overall accuracies ranging from 90 % to 95 %, ensuring the robustness of the results. Land-use dynamics were assessed through annual and overall rates of change and transition matrices. Landscape structure was analysed using the total area, the number of patches and their mean size, complemented by the decision tree developed by J. BOGAERT et al. (2004). Local perceptions were collected on an indicative basis to support and better understand the observed land-use dynamics.The results reveal a continuous decline in dense vegetation formations, whose share of the landscape decreases from 37.7 % in 1990 to 11.7 % in 2022. Open vegetation formations also decline, from 51,0 % to 43.1 %. In contrast, agricultural areas experience a marked expansion, increasing from about 10 % to more than 43 % of the territory, with a clear acceleration after 2007.Structural analysis highlights a progressive dissection of open vegetation formations, characterised by an increase in the number of patches and a reduction in their mean size. Dense vegetation formations undergo severe fragmentation: the ratio of aeras 2022/1990 drops to 0.41, while the number of patches decreases from 1,379,763 to 595,433. Agricultural areas show simultaneous increases in total area and number of patches, reflecting their spatial diffusion and consolidation. These transformations, driven by agricultural expansion and increasing land pressure, indicate a rapid degradation of natural environments and underscore the urgency of sustainable management strategies aimed at preserving ecological continuity and limiting landscape fragmentation in western Burkina Faso.
The island of Santiago has many naturalized species, among which some of them are invasive. This is especially the case of Acacia holosericea, a shrub originated from Australia and introduced in the 1980s to stabilize soils and erosion. This shrub has now spread widely in the environment. It also colonizes agricultural lands, thus causing additional work for farmers who must remove it from crops. Using high-resolution multispectral drone imagery, we extracted the Normalized Difference Vegetation Index (NDVI) for Acacia holosericea (n = 421). For comparison, we carried out the same operation for Prosopis juliflora (n = 267), an exotic tree widely planted for reforestation on the island. NDVI values provide indirect information on photosynthesis activity and plant growth. In this paper, we tested whether some simple physical parameters (elevation, slope steepness, and exposure) could influence the NDVI values of Acacia holosericea. Our statistical results show that, for this species, 5 to 10 % of the NDVI variance are related to elevation and slope steepness, respectively. Acacia holosericea presents a better NDVI on sloping areas located around 495 m altitude, which correspond to the upper part of the mesophilic zone, also called subhumid zone. This zone can therefore be viewed as a preferential physical environment for this acacia. Thus, our results lead us to predict an important extension of Acacia holosericea on the mesophilic zone in the future.
