Predicting ixodid tick distribution in Tamil Nadu domestic mammals using ensemble species distribution models

Background Tick borne diseases are re-emerging around the world, including India. Information about the occurrence of the tick vectors in different geographical locations is essential for controlling the diseases. Tick surveys have not been conducted in many parts of India and information on the current prevalence of tick vectors is not available in all states of India. Many studies have been carried out utilizing modelling methods to predict the distribution of tick species in other countries. The MaxEnt model is widely used for predicting tick species distribution using bioclimatic variables. Lyme disease vectors such as Ixodes sp ., Amblyomma sp., and Dermacentor sp. are the most commonly predicted tick species. However, very few studies have been carried out to predict the distribution of tick species in India. Haemaphysalis spinigera, the primary Kyasanur Forest Disease vector, was predicted along the Western Ghats using the MaxEnt model . Rhipicephalus ( Boophilus ) microplus was predicted across India using the generalized linear model (GLM). Identifying the tick vectors in transmitting the infection through conventional survey and identification methods is cumbersome due to the less number of experienced persons available. Prediction of tick vectors of public health concern, including other tick species in different geographical regions of Tamil Nadu, India, is essential for the prevention and control of tick-borne disease in humans and domestic animals. The present study adopts the package ‘SSDM’ (stacked species distribution models) with R software containing ensemble species distribution models to predict the distribution of tick species using different available environmental and climatic data. Results The categorical variables such as land use and land cover (LULC), soil type, elevation, Bio1, Bio10, Bio15, Bio19 and Bio8 contributed more to modelling the distribution of tick species. MaxEnt, GLM, GBM and GAM are suitable models for predicting the tick species distribution in the present study. Among these models, MaxEnt is the most suitable model for predicting tick species distribution in Tamil Nadu, India. Conclusions Our results suggest that MaxEnt is a suitable model for predicting the distribution of tick species. Both environmental factors such as LULC, elevation and soil type and bioclimatic factors such as temperature and precipitation contribute significantly to predicting tick species distribution in domestic animals in Tamil Nadu.