Asghar Mobaraki; Malihe Erfani; Elham Abtin; Farhad Ataie
Volume 16, Issue 4 , January 2019, , Pages 47-62
Abstract
Introduction:
The mugger crocodile is one of the most vulnerable species in the recent IUCN Red list classification. Most of the populations of the species are in decline and extirpation due to the threats caused by human activities. Nowadays, species distribution modeling plays an essential role in ...
Read More
Introduction:
The mugger crocodile is one of the most vulnerable species in the recent IUCN Red list classification. Most of the populations of the species are in decline and extirpation due to the threats caused by human activities. Nowadays, species distribution modeling plays an essential role in their conservational biology and ecological studies. However, considering the lack of such data on the mugger crocodile in Iran, this study was conducted to evaluate the suitability of the crocodile habitats in the country.
Material and methods:
The watersheds of the Sarbaz and Kaju rivers in the Makran area are the westernmost part of the distribution range of the mugger crocodile, located in the southern part of Sistan and Baluchestan Province. Modeling the suitability of the crocodile habitats in this area was conducted using maximum entropy. The environmental variables of elevation, Normalized Difference Vegetation Index (NDVI), temperature, rainfall, and the distance from urban and rural settlements, roads, and rivers were included in the study. To eliminate high correlation variables (<0.70), Pearson correlation test was used. By using Moran’s I, the lack of spatial autocorrelation between the presence data of the species was assured. Sensitivity and evaluation of the relative importance of each variable were done by the percent contribution and permutation importance of each variable. Also, the response curves of the variables and the jackknife test were calculated. Maximum training sensitivity plus specificity (MTSS) and equal training sensitivity and specificity (ETSS) were used as thresholds to classify the suitable regions.
Results and discussion:
The amount of AUC was more than 0.8, indicating excellent performance of the model. Based on the percent of contribution and permutation importance of each variable and the results of the jackknife test, the distance from the rivers, temperature and elevation were the most important variables. The threshold values of ETSS and MTSS were 0.52 and 0.18, respectively. According to the thresholds, the suitability of the habitat was classified into two suitable and unsuitable classes with an area of 312 and 1629 km2, respectively. Comparison of the thresholds in classification showed that ETSS is more accurate. Considering that the major part of the suitable habitats is located outside of the Gandou Protected Area, revision of the borders of the protected area is proposed as a suggestion for management for the conservation of this species. The study of the effect of dams on the crocodile habitat suitability in the area showed that dams increased the habitat suitability, but their negative impacts, like the loss of nesting sites, should be noted. Also, the fragmentation of the suitable habitats was obvious in the study, and considering the migration behavior of the species, corridors should also be included in conservation plans.
Conclusion:
Distance from the river, as the main variable determining the suitability of the mugger crocodile habitat, is affected by climatic fluctuations and the water amount. Conflicts between the crocodiles and local people for resources (water and food) as well as the immigration of crocodiles between separate habitat patches would be more severe in the warm seasons and drought situations. Therefore, in addition to revising the area of the protected area and conservation of the habitats, increasing species monitoring, public education and participation of local communities in the conservation actions would be essential.
