Potential natural vegetation pattern based on major tree distribution modeling in the western Rif of Morocco

In a fragile ecological environment, understanding the potential spatial distribution of species and the underlying environmental conditions is crucial for resolving ecosystem management issues and predicting impacts of global change. This study aimed to determine the potential distribution of fourteen major Mediterranean taxa (seven conifers and seven broadleaves) native to northwestern Morocco using the Maximum Entropy Algorithm (MaxEnt). Subsequently, a Potential Natural Vegetation (PNV) map model was generated for the study area based on the species with the highest probability of occurrence. MaxEnt demonstrated high predictive power for all studied taxa, exhibiting high accuracy for conifers and deciduous trees, especially those with endemic characteristics and/or specialized habitats. Climatic variables (i.e., mean annual temperature, maximum temperature of the hottest month, mean annual precipitation, and minimum temperature of the coldest month) significantly contributed to the species distribution models. The potential distribution areas of all species were larger than their current areas, indicating the impact of deforestation, particularly in plains and mid-mountains. Relict species such as Abies marocana, Cedrus atlantica, and Pinus mauretanica, are confined to the coldest annual temperature range with very limited ranges (i.e., 1.7°C, 2.6°C, and 1.5°C, respectively), making them extremely vulnerable to global warming. The PNV model demonstrated good reliability (K = 0.67). This is the first application of this approach in the region, providing insights for reforestation and ecosystem rehabilitation efforts in Morocco, with potential applicability to other regions.