Tecomella undulata under threat: The impact of climate change on the distribution of a valuable tree species using a machine learning model

PLOS One | https://doi.org/10.1371/journal.pone.0326609 July 9, 2025 2 / 18 impacting not only the intricate web of life on Earth but also posing direct and indirect consequences for humanity [7]. Climate change, characterized by shifts in temperature and precipitation patterns, is already instigating range shifts in various species, a trend projected to intensify in the future [8]. These climatic alterations disrupt the delicate balance that shapes species distribution, affecting their reproductive success and overall viability. The fragmentation of habitats, driven by global warming, is pushing some species towards extinction while forcing others to migrate in search of more suitable climates [9,10]. The urgency of this situation is underscored by the Intergovernmental Panel on Climate Change’s (IPCC) sixth assessment report, which reveals alarming trends in global temperature rise and paints a concerning picture of future climatic conditions [11]. Understanding the intricate relationship between climate change and species dis- tribution is paramount. As species respond to climatic shifts through changes in their geographic range, phenology, and other ecological attributes, we witness fluctuations in their abundance and, in some cases, their complete disappearance [12]. There- fore, the ability to predict the impact of future climate change on species distribution is essential for the development of effective conservation strategies. Technological advancements in machine learning, ecological theory, and geo-

graphic information systems (GIS) are revolutionizing the fields of ecology and con- servation, offering novel tools to address these challenges [13,14]. Species distribution modeling (SDM) stands at the forefront of these innovations, enabling researchers to generate predictive maps of species distribution based on their known locations and environmental data. By unraveling the complex interplay between abiotic environmental factors and species occurrence, SDMs provide invalu- able insights into the ecological niches of species [14–16]. As modeling algorithms undergo continuous refinement, the use of habitat utility models is becoming increas- ingly prevalent in the fields of environmental science, biogeography, conservation biology, and species management [17]. Among the various SDM algorithms, MaxEnt has garnered widespread recogni- tion for its ability to produce accurate predictive models [14,17–19]. This modeling approach has proven effective for various species, but it holds particular value for endangered species. It consistently outperforms other presence-only models in generating accurate predictive models, especially when dealing with limited input data [19]. The urgency for safeguarding endangered species becomes particularly pro- nounced in harsh climates and areas characterized by low biodiversity. The Desert teak (Tecomella undulata), a deciduous shrub or small tree belonging to the Bigno- niaceae family, exemplifies such a species. This plant, also referred to as Marwar teak or Devil’s pomegranate (Anar Sheitan in Persian), holds significant economic

and medicinal value [20]. There are usually small, pure masses growing along rivers and in relatively flat areas, its global distribution spans South Asia, including India, Pakistan, southern Iran, and the Arab states of the Persian Gulf, extending into North Africa [21]. Within southern Iran, it finds its niche in the provinces of Fars, Kerman, Competing interests: The authors have declared that no competing interests exist.