Westward range extension of Burmese Python Python bivittatus in and around the Ganga Basin, India: a response to changing climatic factors

: The range extension of animals is influenced by various factors, particularly environmental variables and ecological requirements. In this study, we have attempted to quantify the potential current distribution range of the Burmese Python Python bivittatus in and around the Ganga Basin. We collected the Burmese Python sightings between 2007 and 2022 from various direct and indirect sources and recorded 38 individuals, including eight females and five males; the rest were not examined for their sex. Out of these, 12 individuals were rescued from human habitations. Most python sightings were observed in Uttarakhand and Uttar Pradesh (n = 12 each), followed by Bihar (n = 6). The expanded minimum convex polygon (MCP) range was calculated as 60,534.2 km 2 . In addition, we quantified the potential current distribution status of this species using 19 bioclimatic variables with the help of MaxEnt software and the SDM toolbox in Arc GIS. The suitable area for the python distribution was calculated as 1,03,547 km 2 . We found that the following variables influenced the python distribution in the range extended landscape: Annual Mean Temperature (20.9 %), Precipitation of Wettest Quarter (6.4 %), Precipitation of Driest Quarter (30.1 %), Precipitation of Warmest Quarter (0.3%), Isothermality (0.1%), Temperature Annual Range (18.7 %), Mean Temperature of Wettest Quarter (11.4 %), Mean Temperature of Driest Quarter (2.2 %), Land use/land cover (3.3 %), and Elevation (6.6 %). These results will support the field managers in rescuing individuals from conflict areas and rehabilitating them based on the appropriate geographical region.


INTRODUCTION
Reptiles are poikilothermic and are extremely sensitive to the thermal features of the environment (Carranza et al. 2018); hence highly vulnerable to climate change (Sinervo et al. 2016). Minute changes in the environmental temperatures also affect their daily activities, biology, and survival (Wilms et al. 2011;Ribeiro et al. 2012). Several studies have recorded the influence of climatic variables in the distribution of species, i.e., altitude (El-Gabbas et al. 2016), precipitation (Sanchooli 2017), temperature (Javed et al. 2017), and vegetation cover (Fattahi et al. 2014). Studies have concluded that reptiles are more influenced by climate-related variables rather than topographical variables (Guisan & Hofer 2003). Reptiles are being threatened for many reasons, including conversion and loss of habitat, invasive species, and the pet trade, apart from the changes in climate and topographical features, which adversely disturb their spatial distribution (Cox et al. 2012). Pythons, one of the largest reptile groups and apex predators, perform a significant role in the ecological system like other carnivores (Pearson et al. 2005), by controlling the population of ungulates, reptiles, birds, and other small mammals (Bhupathy et al. 2014). Identifying the potential distribution range of species and predicting future potential distribution based on changing environmental conditions have become necessary due to population declines and expansion (Todd et al. 2010;Urban 2015). Many species appear to adapt to rising temperatures associated with climate changes by shifting their ranges to higher latitudes or elevations (Chen et al. 2011;Jose & Nameer 2020) The Burmese Python Python bivittatus is considered one of the largest snake species in the world (Barker & Barker 2008), and it can grow up to a length of 6 m (20 ft) (Clark 2012). Kuhl (1820) has formally distinguished the Burmese Pythons from other python species. P. bivittatus is a squamate reptile of the Pythonidae family, the top of the body is dark brownish-or yellowish-grey, with a series of 30 to 40 large irregular squarish, black-edged, dark chocolate-grey blotches on the top and sides of the body; it has dark and dark grey dorsal and lateral spots; it has a sub-ocular stripe; and the belly is greyish with dark spots on the outer scale rows (Das 2012). The body is thick and cylindrical; the head is lance-shaped and distinct from the neck; sensory pits can be found in the rostrals as well as on some supralabials and infralabials (Das 2012). The spurs are small; the tail is short and prehensile; and there are cloacal spurs (Das 2012).
Python bivittatus is one of three native python  (Stuart et al. 2012). Also, they are included in Schedule-I (Part II) of the Indian Wild Life (Protection) Act, 1972 (IWPA) and Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Burmese Pythons occupy habitats ranging from hardwood forests to mangrove swamps in the introduced range in the USA (Walters et al. 2016), however in the native range, they dwell in the tropical lowlands, grassland forests and within areas modified for human use (Barker & Barker 2008;Cota 2010;Rahman et al. 2014).
In this study, we have attempted to quantify the potential current distribution range of Burmese Pythons in and around the Ganga Basin. Also, identified the bioclimatic variables that contributed to their range expansion.

Study Area
The P. bivittatus live in subtropical or tropical forests, which include dry forests, mangrove vegetation, swamps, moist montane grasslands, wetlands, and permanent freshwater marshes/pools (Stuart et al. 2012). According to the IUCN, the Burmese Python's distribution range as being in northeastern states of India, including West Bengal. The current study focuses on six major Indian states: Uttarakhand, Uttar Pradesh, Bihar, Jharkhand, West Bengal, and Odisha; all apart from Odisha are J TT situated in the Gangetic Basin, however, some Burmese Python sighting records have gathered from the Odisha as well, since it is a neighbouring state of West Bengal.
Ganga is the national river of India which passes through three separate biogeographic zones, the Himalaya, the Gangetic Plain, and the eastern coast, which has a unique biodiversity assemblage (NMCG-WII GBCI 2019). The Ganga River Basin occupies nearly one-third of the geographical area of India (Jain et al. 2007). Presently this region is experiencing a high urbanisation rate and almost 45% of India's population lives in the Ganga basin (Quadir 2022). The temperature of the Gangetic plain doesn't fall under an average of 21⁰C, the daily maximum temperature in the warmest month rises to 40⁰C (EMSF 2019); thus, the atmospheric temperature is very suitable for P. bivittatus. Here, we report the extended native range of P. bivittatus in and around the Gangetic Basin.

Methods and Analysis
The direct sightings of Burmese Pythons have been obtained with photographic evidence from various parts of the study area, with the help of forest staff, researchers, and local people (Image 1). Also, we collected secondary pieces of information from the published works (Table  2). With the available coordinates, a range extension map has been made and the expansion area was estimated by the minimum convex polygon (MCP) in Arc GIS (Supplementary Figure 4). Additionally, the current potential distribution status of this species has been identified with 19 bioclimatic layers, which were obtained from Worldclim dataset. Further, the layers were prepared with the SDM toolbox in Arc GIS and run the model with the help of MaxEnt (Figure 1).
Species distribution for the Burmese Python was modelled using MaxEnt (version 3.4.1.; Phillips et al. 2004Phillips et al. , 2006 because it is the most widely used and popular choice for species distribution modelling, providing high extrapolative accuracies even with low presence-only data (Bosso et al. 2018;Soucy et al. 2018;Zhang et al. 2018). This study has only used presence data and to generate pseudo-absences, 10,022 background points were randomly selected by the MaxEnt model.
The presence data was split into 75% random samples for calibrating the model and 25% for evaluating model performance. We used a subsampling technique to generate a stable model because of its advantages over cross-validation (Anderson & Raza 2010), and bootstrap (Rospleszcz et al. 2014), and three replications were chosen to run the model. Regularization multipliers are used to prevent overfitting of predicted values and to balance the model fit (Phillips & Dudík 2008). The model provides settings for assessing model complexity by varying feature classes and regularisation multipliers. Threshold selection was done, the logistic output format ranging between 0 (unsuitable) and 1 (maximum suitability), was used for the model results, which shows habitat suitability (presence probability) of

RESULTS AND DISCUSSION
We collected the details of Burmese Pythons in the Ganga Basin and adjacent areas. The data has been collected from both direct and indirect sources (Table 2). A total of 38 sighting records were obtained, including eight females, five males, and the remaining unsexed. The pythons were identified using photographs and morphological features from the field guide by Whitaker & Captain (2004).
The Burmese Pythons are known as the sister species of Indian Rock Python P. molurus and the Burmese Python differs from the Rock Python in several ways. Supralabials touching the eye, the tongue, and some parts of the head are pale pinkish in Indian Rock Python. The supralabials are separated from the eye by subocular scales in the Burmese Python and the tongue is bluishblack with no pink colour on the head (Whitaker & Captain 2004). Also, the Indian Python being 'yellowish' while the Burmese Python is 'greyish' in colour (Whitaker & Captain 2004).
From these, 10 individuals were rescued from human habitations. Also, a mating event was observed in August by the NMCG Team of WII, and a brooding female was observed by Rashid & Khan (2018) in May. Das et al. (2012) reported earlier breeding records of Burmese Python such as egg shell remains and earlier nesting activities in the Gangetic Basin, at the Katerniaghat and Dudhwa regions. Most of the python sightings were recorded from the state of Uttarakhand and Uttar Pradesh (n = 12 each), followed by Bihar (n = 6), West Bengal (n = 4), Odisha (n = 3), and Jharkhand (n = 1) respectively. The expanded MCP range was calculated as 60,534.2 km 2 (Supplementary Figure 4). The most python sighting records were obtained in the year of 2017 (n = 8), followed by the year 2021 (n = 6) ( Figure 3).
In addition, we found that some environmental variables have a considerable role in the distribution of P. bivittatus (Table 1 and Supplementary Table 1). The suitable area for the potential distribution of P. bivittatus was predicted as 1,03,547 km 2 . The Gangetic Plain and its adjacent places have a favourable temperature for the species rapid expansion.
From the Jackknife evaluation, these results were consistent. The model output yielded satisfactory results with the training and test data; the final model had accuracy with an AUC value of 0.865.
The present model outputs show that 10 variables influence the python distribution. Some variables, however, have a high proportion. Temperature and precipitation both play a significant role in their distribution.
An optimum temperature is essential for their survival and dispersal. According to research, excessively cold temperatures make it difficult for pythons to survive (Mazzotti et al. 2011). The reported lethal temperature in the low land species is approximately 38-42˚C (Brattstrom 1968;Snyder & Weathers 1975), and the increased temperatures can affect the population sex ratios of reptiles (Bickford et al. 2010). A study conducted in China among 50 snake species found that the distribution of species was related to changes

J TT
in the thermal index and precipitation or potential evapotranspiration (Wu 2016). The Jackknife evaluation results revealed that the Wettest Quarter Mean Temperature, Annual Mean Temperature, and Driest Quarter Precipitation were the primary factors influencing the P. bivittatus distribution (Figure 1). The percent contribution values are given in Table 1. A proper field survey in the remaining area would yield more sightings across the basin.
According to the findings, the Driest Quarter Precipitation (30.1%) is a significant influencing factor for the range extension of the Burmese Python in the Gangetic Basin, however, Penman et al. (2010) discovered that the Driest Quarter Precipitation is a major bioclimatic variable that has a significant impact on the distribution of the most endangered Hoplocephalus bungaroides snake species in Australia.
Similarly, Annual Mean Temperature is a significant variable that influences species distribution. Annual Mean Temperature contributed 20.9% to the Burmese Python distribution in the study area. Annual Mean Temperature is a significant bioclimatic factor for the species ( influenced its distribution. The contribution of elevation was 6.6% and landcover was found to have 3.3%. The elevation also plays a role ecologically since it affects the temperature (Ananjeva et al. 2014;Hosseinzadeh et al. 2014). Studies have concluded that with a gain in elevation, species richness among reptiles would decline (Chettri et al. 2010).
Our findings show a trend in the westward range extension of the Burmese Python in the study area, which could be attributed to a response to changing climatic factors. In the United States, some studies have proven that less body temperature during the cold snap leads to physiological stress on this species and may lead to mortality (Mazzotti et al. 2011;Stahl et al. 2016). Jacobson et al. (2012) observed that the Burmese Pythons are projected to spread northward in response to warming winter temperature regimes. Nevertheless, Van Moorter et al. (2016) stated that animal movement is directly connected to resource use, such as habitat selection. However, recent records justify that this species having a good population along the Gangetic plain (Rashid & Khan 2018;Shafi et al. 2020).
Scarce SDM studies were conducted among reptile species in India; the primary reason is the only way to know about the occurrence localities of their collections is through publications of researchers. In many cases, a direct visit to the particular institutes is the only way to get the required data, which takes considerable time (Das & Pramanic 2018). In addition, finding them in the field is very difficult due to their highly camouflaged behaviour.  The Journal of Threatened Taxa (JoTT) is dedicated to building evidence for conservation globally by publishing peer-reviewed articles online every month at a reasonably rapid rate at www.threatenedtaxa.org. All articles published in JoTT are registered under Creative Commons Attribution 4.0 International License unless otherwise mentioned. JoTT allows allows unrestricted use, reproduction, and distribution of articles in any medium by providing adequate credit to the author(s) and the source of publication.