Presence probability of Hemiscorpius lepturus Peters, 1861 using maximum entropy approach in the western areas of Zagros Mountains, Iran

Research (Published online: 17-02-2020)

11. Presence probability of Hemiscorpius lepturus Peters, 1861 using maximum entropy approach in the western areas of Zagros Mountains, Iran

Ahmad Ali Hanafi-Bojd, Mona Sharififard, Elham Jahanifard, Shahrokh Navidpour and Babak Vazirianzadeh

Veterinary World, 13(2): 296-303

ABSTRACT

Aim: The purpose of this research was to use environmental variables for predicting the probability of Hemiscorpius lepturus existence in the provinces where situated in the west of the Zagros Mountains.

Materials and Methods: In this study, 64 occurrence records of the H. lepturus were extracted from the published documents available in electronic databases. MaxEnt model was used for predicting the ecological niches of this species. Normalized difference vegetation index (NDVI) and 19 climatic variables were used as the environmental variables affecting the distribution of this scorpion. The Jackknife test in the model was used to indicate the importance of variables to predict the probability of the presence of the studied species. The logistic threshold that was evaluated using a logistic regression algorithm showed the converting of the probability model into a binary model. The model was evaluated by area under the curve (AUC). The probability presence map of this scorpion was then prepared in ArcGIS 10.5 Software.

Results: The results of the analysis showed that the most important environmental factor on the distribution of H. lepturus was the maximum temperature of the warmest month (Bio5) with a contribution rate of 43% and permutation importance of 8%. The Jackknife test revealed that NDVI did not gain any value when it used independently in the model. The logistic threshold was reported 0.255 for the maximum test sensitivity plus specificity. The AUC of the model was 0.7698, shows an acceptable value for model validity. Overall the hot spots for this toxic scorpion seem to be in Khuzestan, Lorestan, and Ilam Provinces of the studied area.

Conclusion: Regarding our findings, MaxEnt algorithm, in combination with geographic information system contributed to revealing the effects of environmental variables on the probability of H. lepturus presence in the west of Zagros Mountains. These visualized maps as a warning alarm can be helpful to policymakers for managing, controlling, and monitoring the scorpionism in high-risk areas.

Keywords: ecological niches, Hemiscorpius lepturus, Iran, MaxEnt, Zagros mountains.