Modeling the distribution of i>Avicennia marina (Forssk.) Vierh. in the Khamir and Qeshm mangrove forests, Iran using the maximum entropy model (MaxEnt)

Background and objectives: In recent decades, the Khamir and Qeshm mangrove forests in southern Iran have suffered significant destruction due to various human activities, including the construction of docks, commercial and tourism ports, deforestation, and unplanned tourism development. These forests, which provide valuable ecosystem services, have faced indiscriminate and unplanned development, as well as insufficient attention from local communities. Therefore, understanding the spatial distribution of plant species in this area is crucial for effective planning and enhancing the protection of these valuable biological resources. This study aims to identify suitable areas and model the presence of the mangrove tree species Avicennia marina (Forssk.) Vierh. in the Khamir and Qeshm mangrove forests to facilitate the restoration and proper distribution of this species.
Methodology: The habitat suitability map for A. marina was prepared using MaxEnt modeling. A total of 234 points were randomly recorded using the Global Positioning System (GPS). To investigate the spatial distribution of A. marina, various environmental factors affecting its geographical distribution were analyzed using ArcGIS 10.8 software to generate maps and environmental variables. The environmental variables were selected based on theoretical foundations, previous studies, and expert opinions. The model included 18 climatic variables and 5 physical variables affecting the distribution of A. marina. The physical variables included maximum wave height, beach slope, tidal fluctuations, water salinity, and beach material (sandy-gravel, muddy, back beach (sand material)). All layers were prepared in ASCII format with a cell size of 1 km, and MaxEnt 3.4.4 software was used for modeling.
Results: The area under the curve (AUC) obtained from the MaxEnt algorithm indicated excellent predictive power for the presence of A. marina in the study area, demonstrating the model’s ability to distinguish between suitable and unsuitable areas. The overlap of training and test data also confirmed the model’s accuracy. The contribution analysis of each environmental variable in the model showed that maximum wave height, annual average temperature, tidal fluctuations, and water salinity were the most influential variables. Specifically, maximum wave height and annual average temperature had the largest influence on the distribution of A. marina, while the minimum temperature of the coldest month had the least effect. The response curves indicated that maximum wave height (1 to 2 m) and average annual temperature (26.8°C) were the most important independent variables, with an inverse relationship to the probability of A. marina presence. As wave height and average annual temperature increased, the probability of species presence decreased. The suitability map for A. marina in the Khamir and Qeshm mangrove forests showed that the highest suitability was in areas with minimal wave height and less exposure to tidal fluctuations. Favorable areas for the presence and development of A. marina included the northern areas of Khorkhoran Islands, Mardove Island, and the northeast of the Khamir-Lashtghan habitat. In general, the species was more dispersed in areas with low wave height and minimal water level fluctuations.
Conclusion: The study provides key information on the impact of environmental variables on the distribution of A. marina, offering an important strategy for protecting the biodiversity and valuable resources of the species in the Khamir and Qeshm mangrove forests. The suitability map serves as essential information for planting and reviving these natural habitats.