Machine learning ensemble species distribution modeling of an endangered arid land tree Tecomella undulata : a global appraisal

Tecomella undulata is a valuable tree that is threatened owing to unlawful harvesting and habitat fragmentation. The current study looked at this species’ ecological niches in hot, arid locations around the world. Ensemble modeling was used in this study to assess the species’ global distribution based on current and future bio-climatic (2050 and 2070) and four green house (RCPs 2.6, 4.5, 6.0, and 8.5) scenarios, as well as soil attributes. Our findings suggest that bioclimatic factors, rather than soil, are the primary constraint on this species’ spread. Isothermality and precipitation seasonality influenced the spread of this species. In 2050 and 2070, the largest region covered by the optimal and moderate classes dropped from RCP 2.6 to RCP 8.5. When current climatic circumstances are taken into account, optimal habitat suitability falls from − 13.09% in 2050 RCP2.6 to − 50.1% in 2050 and 2070 RCP8.5. Habitat loss in 2050 was greater than in RCP4.5 and 6.0 for 2070. When analyzing RCP combinations for this species, we came upon an unusual circumstance. Combining RCP6.0 and RCP8.5 with 2050 yielded the best results, whereas combining RCP 4.5 and 6.0 produced the worst. The findings may be useful to government and non-profit forest management organizations.