Integrating remote sensing and species distribution modeling to assess the impacts of flood disturbance on the species habitat in the central Zagros

• Published in: Discover applied sciences Vol. 7; no. 9; pp. 935 - 21
• Main Authors: Moradi, Soheyl, Moradi, Hossein, Rezvani, Azita
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.09.2025; Springer Nature B.V; Springer
• Subjects: Applied and Technical Physics; Bears; Chemistry/Food Science; Climate change; Conservation planning; Degradation; Earth Sciences; Ecological effects; Engineering; Environment; Environmental degradation; Flood damage; Flood impacts; Flood management; Flooding; Floods; Geographical distribution; Goats; Habitat degradation; Habitats; Keystone species; Materials Science; Mountains; Natural resources; Normalized difference vegetative index; Ponds; Prediction models; Protected areas; Protected species; Real time; Remote sensing; Small mammals; Species distribution modeling; Squirrels; Sus scrofa; Threatened species; Tortoises; Turtles; Urban areas; Vegetation; Water bodies; Wildlife; Wildlife conservation; Wildlife habitats; Zagros mountains; Iran; Zagros Mountains; Zagros mountains; Species distribution modeling; Flood impacts; Conservation planning; Remote sensing; Habitat degradation
• ISSN: 3004-9261; 2523-3963; 3004-9261; 2523-3971
• DOI: 10.1007/s42452-025-07582-1

Floods cause severe damage to natural resources and urban areas. Many studies have focused on the impacts of floods in urban areas, but research assessing flooding and its impacts on wildlife habitats is scarce. In recent years, the scientific community has shown how remote sensing can play a vital role in flood model calibration, validation, and real-time monitoring. The Sefidkuh Protected Area in the Zagros Mountains of Iran is an important habitat for diverse wildlife species. The year 2019 had one of the worst floods in recent decades, destroying the Zagros region, which was flooded at about 210 km 2 (30% of the study area), as determined from Sentinel-1 pre- and post-flood images. The present study aims to evaluate the ecological consequences of the flood in assessing habitat degradation for some selected keystone species in the protected area using the Species Distribution Models that may include the brown bear ( Ursus arctos ), wild goat ( Capra aegagrus ), Persian squirrel ( Sciurus anomalus ), wild boar ( Sus scrofa ), Caspian pond turtle ( Mauremys caspica ), and Greek tortoise ( Testudo graeca ). Flood extent was extracted using the Sentinel-1 Flood Service method in the European Space Agency’s SNAP software with a threshold of 0.01 for water body separation. Four remotely sensed indices—Normalized Difference Vegetation Index, Soil-Adjusted Vegetation Index, Normalized Difference Water Index, and Normalized Difference Pond Index—along with human presence and topography, were used to evaluate pre- and post-flood habitat suitability for the species. While for both the 2018 and 2019 Species Distribution Models, the MaxEnt model demonstrated strong predictive performance with AUC values above 0.8. In contrast, wild boar had a lower AUC, which was limited by the number of presence points. The disturbances occurred in the most suitable habitats, with wild goat suffering the most degradation, 86.39% of the suitable habitat affected (0.95 km 2 ), followed by Persian squirrel (22.23%, 0.45 km 2 ). Other species, such as brown bear (10.08%, 6.52 km 2 ), wild boar (5.02%, 0.34 km 2 ), Caspian pond turtle (8.38%, 0.51 km 2 ), and Greek tortoise (21.21%, 1.41 km 2 ), were affected. Jackknife tests highlighted key environmental variables that influence distribution for many species, such as slope, roughness, and NDVI. As climate change causes more unexpected severe flooding, the habitats of key mountain species are threatened. By integrating Sentinel products and habitat suitability modeling, this study highlights the value of remote sensing and predictive models for conservation planning in protected areas amid extreme climate change events.