Caracal Caracal caracal (Schreber, 1776) (Mammalia: Carnivora: Felidae) in India

This article focuses on the historical and current extent of occurrence of the Caracal Caracal Caracal in India between 1616 and April 2020. We collated 134 reports during this period. Historically, the Caracal was reported in 13 Indian states in nine out of 26 biotic provinces. Since 2001, the Caracal’s presence has been reported in only three states and four biotic provinces, with only two possible viable populations. Before 1947, the Caracal was reported from an area of 793,927km2. Between 1948 and 2000, the Caracal’s reported extent of occurrence in India decreased by 47.99%. From 2001 to 2020, the reported extent of occurrence further decreased by 95.95%, with current presence restricted to 16,709km2, less than 5% of the Caracal’s reported extent of occurrence in the 1948–2000 period.


INTRODUCTION
The Caracal Caracal caracal is among the most widespread of small wild cats, distributed across at least 20 million km 2 including 42 African and 18 Asian countries (Avgan et al. 2016). The oldest report of the Caracal in the Indian subcontinent is a fossil from the Indus Valley Civilization dating to 3000-2000 BCE (Ghosh 1982). In Asia, the Caracal's historical range overlaps with small ungulate species such as Blackbuck Antilope cervicapra in India (Sunquist & Sunquist 2002) and Chinkara Gazella bennettii in Iran and India (Sunquist & Sunquist 2002;Farhadinia et al. 2007;Ghoddousi et al. 2009;Moqanaki et al. 2016). The Caracal is known to inhabit the drier parts of India (Kitchener 1991;Corbett & Hill 1992;Nowell & Jackson 1996). Knowledge of its conservation status, however, is largely outdated, especially for the Asian populations (Moqanaki et al. 2016).
The Caracal has a long and unique history with humans in Asia where it was valued for its litheness and ability to catch birds in flight (Vigne 1842;Lydekker 1907;Sunquist & Sunquist 2002). The Caracal's iconic large black ears with long tufts of hair at the tips are emphasized in its name, which originates in the Turkish word 'Karakulak', meaning 'Black Ear' (Buffon 1761). In India, it is vernacularly known as 'Siya Gosh', a Persian name meaning 'Black Ear' (Harting 1883). A Sanskrit fable accounts of a small wild cat named Dirgha-karan or 'long-eared' preying on a bird's chicks (Capeller 1891;Arnold 1893). This cat might be a Caracal. The Sanskrit name 'sas-karan' meaning 'rabbit-like ears' was proposed by Vira et al. (1953) in an attempt to establish a Sanskrit nomenclature for the fauna of India, Myanmar and Sri Lanka following the Linnaean system of classification.
In India, the Caracal was used as a coursing animal during the period of the Delhi Sultanate in the late medieval period (Divyabhanusinh 1993;Verma 1994;Thapar et al. 2013). In the 14 th Century, Firuz Shah Tughlaq was thought to have established a provision called 'Siyah-Goshdar Khana' solely dedicated to the maintenance of large numbers of coursing Caracal (Verma 1994). The Third Mughal Emperor Akbar furthered the reputation of the Caracal as a coursing animal and used it extensively for coursing (Blochmann 1873). It was also represented in illustrated simplified Persian adaptations of texts sourced from Sanskrit, Turkic and Arabic literature such as Anvar-i-Suhayli, Tutinama, as well as Persian poetry and epics such as Khamsa-e-Nizami and Shahnameh (Maurice 1953), which are full of wildlife fables. The Caracal's historical use as a coursing animal might have taken it far beyond its natural range in places like the Ladakh region in the Himalaya (Pocock 1939) along with Calcutta in West Bengal (Brandon-Jones 1997).
The Caracal in India has been considered rare in the wild since at least 1671 (Foster 1924(Foster , 1926, and several naturalists commented on its rarity (Hamilton 1727;Blyth 1842;Stoliczka 1872;Jerdon 1874;O'Malley 1909;Allen 1919;Sharma & Sankhala 1984a). We think that its rarity may be explained by the economic development of India. The country is primarily an agrarian economy, with 70% of the human population living in rural areas (Chand et al. 2017). In the 20 th Century alone, India's human population grew six-fold, which along with economic growth resulted in the total forest area decreasing from 1,000,000-810,000 km 2 (Tian et al. 2014). Agricultural land in India increased from 1,000,000-1,200,000 km 2 between 1880 and 1950 (Tian et al. 2014). Approximately, 200,000km 2 of grassland and shrub land, along with 260,000km 2 of forests are estimated to have been converted for agricultural use from 1880 to 2010 (Vanak et al. 2017).
Landscapes in India have significantly been transformed by such anthropogenic factors. Against the backdrop of these large-scale changes, we consider it important to examine the change in the extent of occurrence of the Caracal in India. The prevalence of coursing Caracal historically along with the seemingly elusive behaviour of wild Caracal makes this a challenging endeavour.
We collated all credible reports of the Caracal in India from the beginning of recorded history until April 2020, mapped its historical range and assessed changes in its present extent of occurrence.

Study area
Historical sites with Caracal reports were spread across northwestern and central India to the states of Jharkand and Odisha in the east. This region contains four biogeographic zones with 10 biotic provinces (Table  1; Rodgers et al. 2002). The climate in this region is dominated by the south-west Asian monsoon with rain falling in the months of June to September (Prakash et al. 2015). During this season, the mean annual rainfall varies from 100-500 mm in the Thar Desert (Roy & Singhvi 2016) and increases eastwards to over 1,300mm (Prakash et al. 2015). Mean annual temperatures range from 3-10°C in the cold season to 45-50°C in the hot season (Roy & Singhvi 2016). It also must be stated that the international border between India and Pakistan passes through the states of Rajasthan, Gujarat, and Punjab. Permanent fencing began in the 1980s and has now almost been completed (Gupta 2018). Thus, while

Survey on literature, specimens in collections and interviews
We searched for literature about wild Caracal in India from the start of recorded history to 2020 including the writings of credible authors such as naturalists, zoologists, natural historians, historians, forest officers, gazetteers, chroniclers, erstwhile royalty, and army officers. Literature was sourced online and in the libraries of the India International Centre, Maharaja Fatehsinghrao Gaekwad Library at the WWF-India secretariat and at the Indira Gandhi National Centre for the Arts, all in New Delhi; and in the library of University of Rajasthan in Jaipur. Literature was also provided by Valmik Thapar, Divyabhanusinh Chavda, and Satish Sharma. We examined Caracal specimens deposited at the Bombay Natural History Society (BNHS), Zoological Survey of India (ZSI), the Natural History Museum in London, private trophy collections in India, and other museums.
We also conducted open-ended interviews with forest officers and biologists who observed the Caracal in the field and people who provided photographs.
We assessed the reliability of the information obtained and categorized reports into: confirmed reports based on tangible evidence like photographs, specimens including animal carcasses or body parts that can be accessed currently; confirmed reports based on direct sightings of live or dead individuals, specimens submitted to museums that are no longer accessible or missing, photographic reports that are no longer accessible, destroyed or missing; confirmed reports that indicate Caracal occurrence through species specific information which includes species description and the provision of distinct vernacular names; unconfirmed or questionable reports without any accompanying description, photos or erroneous description.
Reports of captive or coursing Caracal are strictly not included as their wild origins are unknown unless explicitly stated.

Camera trapping
Regular camera trapping using Cuddeback X-ChangeTM 1279 models was carried out in selected sites on the peripheries of Ranthambhore

Extent of occurrence maps
To account for international boundary changes that have occurred in the region, we sorted the collated information into three categories. The first category entails all reports from undivided India until 1947. The second category excludes Pakistan and entails reports from 1948 to 2000. Both categories are considered historical reports.
The third category comprises contemporary reports from 2001 to April 2020, a time period when camera trapping and photography of wildlife became more common and resulted in the availability of authentic information.
Coordinates of each report were plotted using ARC GIS 10.3 where possible. If it was not possible to determine coordinates, then the centre of the province, principality or state was plotted. We visited all locations in the third category to gather data on habitat types and water sources. The geotagged locations were used to build extent of occurrence maps in QGIS 3.12 Bucuresti version and are also shown on a map of the Biogeographic Classification of India by Rodgers et al. (2002).
The outermost geotagged locations on the map were connected to plot a minimum convex polygon. More than 50 locations in a protected area (PA) are comprised in a single polygon, so that the entire PA formed one geotagged polygon on the map and is represented by digits on both the table and the map. Multiple locations within the polygon are represented by Roman numerals in tables. Geotagged locations outside PAs in the same district were marked separately on the maps.
Historical reports of the Caracal from 1616 to 1947 extend over an area of 839,398km 2 (Figure 1), including an area of 45,471km 2 in Pakistan. If we subtract the area in Pakistan, the area within India's current borders extends over 793,927km 2 . Reports from1948 to 2000 extend over an area of 412,877km 2 ( Figure 2).

Caracal specimens in collections
We found 13 Caracal specimens in collections ( Table  4). Six of these specimens are known to have originated in Rajasthan, Gujarat, Madhya Pradesh, and Uttar Pradesh. While the Bombay Natural History Society (BNHS) had eight Caracal specimens in its collection from 1888 to 1907, only two of them remain today (Curator, BNHS in. litt. 2019).

Records from 2001 onwards
We obtained authentic Caracal records between 2001 and 2020 in Rajasthan (Table 5), Gujarat (Table  6) and Madhya Pradesh (Table 7). These records are mapped in Figure 3.

17178
J TT 2020 at 23 locations, clubbed into six different areas in and around Ranthambhore Tiger Reserve (Table 5). b. The Caracal in Gujarat: We found 19 reports of the Caracal in the state of Gujarat, all in the Kutch District. Nine of these reports are authenticated by photographs (Table 6).
c. The Caracal in Madhya Pradesh: We traced three Caracal reports in Madhya Pradesh at two locations (Table 7). Since none of these reports are supported by photographic evidence, they are category accounts.

DISCUSSION
Our collation of literature revealed that knowledge of the Caracal's presence in India until the end of the 19 th century was based on just 17 locality reports. The locations of these reports are scattered over the states of Rajasthan, Delhi, Haryana, Punjab, Gujarat, Madhya Pradesh, Uttar Pradesh, Maharashtra, Andhra Pradesh, Telangana, Odisha, Jharkhand, and Chhattisgarh. It is possible that the rarity of reports was the reason for early naturalists assuming that the Caracal is rare in India. The Caracal's historic range in India is very putative, as many reports are not supported by photographic evidence and can, therefore, neither be corroborated nor used to draw inferences. Although Jerdon (1874) reported to have obtained specimens in Odisha and northern Andhra Pradesh, Blanford (1888-91) and Lydekker (1907) assumed that its presence is limited to northwestern and central India. Examination of literature on rock painting sites in Rajasthan, Gujarat, Madhya Pradesh and Uttar Pradesh revealed no clues on the Caracal, and hence there is no pre-historic report (Chakravarty & Bednarik 1997;Wakankar 2005Wakankar , 2008Saleem 2014;Sharma 2014;Gupta 2019).
Reports of the Caracal until the end of the 20 th Century increased four-fold, indicating that the extent of the Caracal's occurrence putatively stretched over an area of 1,040,598km 2 in 13 states and nine biotic provinces. After 2001, its presence has been documented in just three states until April 2020.
According to Mukherjee (1998), the Caracal occurred in five biotic provinces in India, viz, 3A: Thar Desert, 3B: Kutch Desert, 4B: Semi-arid Gujarat Rajputana, 6A: Central Highlands and 6D: Central Plateau of the Deccan Peninsula. From our collation of reports starting in 1616, the Caracal was reported from four more biotic provinces in addition to those listed by Mukherjee (1998). These are 4A: Semi-arid Punjab Plains, 7A: Upper Gangetic Plain, 6B: Chotta Nagpur and 6C: Eastern Highlands in the Deccan Peninsula.
Authenticated reports supported by specimens and photographs are from 4B: Semi-arid Gujarat Rajputana,

J TT
4A: Semi-arid Punjab Plains, 3B: Kutch Desert and 6A: Deccan Peninsula Central Highlands. As many of the historical records are without specimens or photographs, mistaken identification with the Jungle Cat is possible. Such misidentifications are common even today, with some faulty reports being perpetuated only because they are published, e.g., Gogate (1998).

J TT Unconfirmed records of the Caracal
We found 33 reports that we categorised as unconfirmed.
Two originate in the Ladakh region of Jammu & Kashmir. The first is based on a drawing of a captive Caracal in Baltistan (Blyth 1842), and the second on a skin seen in a Srinagar shop by Ward (1923). Stockley (1928) and Pocock (1939) held the view that neither one is evidence for the Caracal's occurrence in the erstwhile state. Ward (1923) also accounted of shooting a Caracal in western Dun, Uttarakhand. In a map showing the extent of occurrence of the Felidae in the western Himalaya, Sinha (1995) included the Caracal in Dehradun without providing any related information. This location matches with the account by Ward (1923). In view of Ward's earlier claim from Ladakh, we doubt the credibility of this account.
The British army officer Osborn claimed that a Caracal was sighted in the Kangra District of Himachal Pradesh in 1904 (Government of Punjab 1904;Dodsworth 1913). A purported specimen was submitted by Osborn to the museum of the BNHS in 1907 (Bell 1907). Considering that Osborn provided neither details of the sighting nor of the submitted specimen, it is possible that this could have occurred in an area either in or bordering modern day Punjab.
Two publications refer to the rarity of the Caracal in the erstwhile Madras Presidency.
McMaster (1871) accounted of a Caracal presented by the Rajah of Karvetinagar to the people's park of Madras in September 1868, but did not clarify from where the Caracal originated. Thurston (1913) wrote that the Caracal or Red Lynx had become rare, but without providing information whether and where it was ever sighted or hunted in the region.
Briggs (1861) wrote about the presence of the Caracal in Surroo Nagar, currently in Telangana, but merely mentioned Lynx along with a host of other wildlife. The word 'Lynx' was at times also used for the Jungle Cat. Briggs (1861) neither provided a description nor information about a sighting. Behura & Guru (1969) reported the occurrence of the Caracal in Mayurbhanj District on the basis of a newspaper article dating to 18 April 1962 (Acharjyo 1998). This report was further included in a ZSI publication by Das et al. (1993). Acharjyo (1998), however, conceded that no other reports of its occurrence in the state were known at the time, nor had the Nandankanan Zoo received a wild-caught individual from any part of Odisha.
Two reports of Caracals around the Ludhiana area of Punjab in 1977-79 are unconfirmed (Gurmit Singh pers. comm. 2019). Parihar (1989) reported seeing a Caracal on the night of 16 March 1987 aided by a searchlight in the forests of Panna District in Madhya Pradesh. He acknowledged that the Jungle Cat is common there, but was certain that he saw the front and rear of a Caracal, although he "could not see the ear tuft" and described the tip of the tail being about 10cm long and darker than the rest of the tail. This description raises doubts, as Caracal tail length in India has been reported ranging from 17.5 to 29.9 cm (Blyth 1842;Jerdon 1874;Sterndale 1884;Allen 1919;Ward 1923;Prater 1948;Dharmendra Khandal pers. obs. 2019;Sonia Mondal in. litt. 2019). The tip of the tail measures approximately 1-2 cm with hair that is darker than the rest of the tail, but such hair is not present on the tails of all Caracals (Dharmendra Khandal pers. obs. 2019).
Parihar (1989) Desai (1974) included the Caracal in the list of animals in Gir National Park, and Singh (1998) claimed a sighting of a Caracal in tall grass in Saurashtra, Gujarat. Evidence of occurrence in both areas does not exist, despite regular camera trapping surveys today, and so we consider both reports unconfirmed. Chakraborty & Agarwal (2000) referred to 10 individual Caracals in Narayan Sarovar Chinkara Sanctuary listed in a report by Forest Department of Gujarat. This report is based on the annual waterhole census method for wild animals, which has serious limitations (Karanth & Ramaswamy 2006).
Two separate Caracal sightings were reported in the Dhakana and Gatang ranges of Melghat Tiger Reserve in Maharashtra (Gogate 1998). These were later proven to be Jungle Cats upon examination of photographic evidence (Shomita Mukherjee pers. comm. 2020). Singh & Soni (1999) mentioned the presence of the Caracal in the salt pans of Wild Ass Sanctuary located in Little Rann of Kutch, Gujarat, based on local people's accounts. Singh & Soni (1999), however, stated to not have sighted a Caracal themselves. It must be added that the salt pans are a highly improbable habitat for the Caracal. Kolipaka (2011) listed 11 sightings of the Caracal across India along with their purported GPS coordinates. The sources of these reports are not provided. Nine of the reports are verifiably non-specific. The errors in the locations listed and associated coordinates provided are significant, such as two different locations being presented as one, e.g., "Ranthambhore & Sariska" and "Sasaipura, Bhind". These errors make it impossible to verify the alleged sightings.
In 2015, whilst conducting a field survey in the Bagpat Reserve in the taluka of Nakhatrana in Kutch to assess the "Status and Distribution of Caracal in Gujarat", a team from the Gujarat Institute of Desert Ecology (GUIDE) claimed to have sighted a Caracal hiding 70-75 m away in a bush. The animal escaped before it could be photographed. The team, however, claimed that the animal left 'clear pugmarks for our confirmation' and provided a photograph of the pugmark with a pen placed next to it for size reference. They also surveyed the area for 10 more days but could not find the animal (Joshi et al. 2015). The length of the pen appeared to be 12-13 cm long, the average length of a pen. By comparison, the pugmark appeared to be 8-9 cm long, which is far too large for a Caracal. In an earlier update on the same survey, GUIDE provided the lengths of the right front pugmark and the right hind pugmark of a Caracal, presumably obtained in their survey, and these measured 4.7cm and 5.5cm respectively (Asrari et al. 2013-14). Skinner & Chimimba (2005) provide 5cm as the upper limit for the lengths of the right fore footprint and right hind footprint for southern African Caracals. We are, thus, of the opinion that the pugmark report is erroneous, and that the pugmark photographed was that of a Leopard. Kazmi (2020) interpreted Sterndale (1884) to have reported "raising a young Caracal cub he had caught from the wild" in Seoni, Madhya Pradesh. In fact, Sterndale (1884) merely wrote that "They are easily tamed. I had a young one at Seonee and the natives of some parts are said to train them for sporting purposes in the manner in which the hunting leopard (read Cheetah) is trained". He did not explicitly state that this cub was 'caught from the wild' in Seoni itself. We are, therefore, of the opinion that the information by Sterndale (1884) is insufficient to draw a conclusion of the cub's origin.

Contemporary reports
Since 2001, the Caracal was reported in only four biotic provinces, namely 3B: Kutch Desert, 4B: Semiarid Gujarat Rajputana, 4A: Semi-arid Punjab Plains, and 6A: Central Highlands in the Deccan Peninsula (Figure 3). Judging by historical and contemporary reports, 4B: Semi-arid Gujarat Rajputana seems to be the stronghold for the Caracal. The biotic province 3B: Kutch Desert has more numerous reports in the current period than in the historical period. This is likely due to greater accessibility of this biotic province today than in the past. Very few historical records are known in 3A: Thar Desert, and no contemporary reports are known despite far greater access to this region today. Prakash (1994b) considered the Caracal to have been 'very common' in the Thar Desert in the beginning of the 20 th Century. He, however, does not provide any evidence to validate this statement. The report from 4A: Semi-arid Punjab Plains is very close to the border with 4B: Semi-arid Gujarat Rajputana, and the report in 6A: Central Highlands is a stand-alone report with no photographic documentation. Therefore, it appears that Caracal populations existing in 4B: Semi-arid Gujarat Rajputana and 3B: Kutch Desert could be the only viable populations in India today. Nevertheless, more targeted surveys are needed in addition to the intensive camera trapping targeting the Tiger in 4B: Semi-arid Gujarat Rajputana. It must, however, be noted that the multiple J TT    The putative extent of occurrence of the Caracal decreased by 47.99% in the period before 1947 to the period between 1947 and 2000, and the putative extent of occurrence area of the latter period accounted for 52% of the period until 1947.
The locations with verifiable reports from 2001 onwards are within a total area of 16,709km 2 , which is J TT  There, however, have been practically no attempts to survey the Caracal in most regions of its past reported range and the much higher effort put into camera-trapping and photographing in and around Ranthambhore and Kutch could be a reason for the higher numbers reported. Physical connectivity between these four landscapes is highly fragmented with potential impact on gene flow and population connectivity for the Caracal. This range encompasses Sariska Tiger Reserve, Ranthambhore Tiger Reserve, the districts of Udaipur and Chittorgarh in Rajasthan, the Kutch region in Gujarat and the Chambal ravine area in Madhya Pradesh.
No focused surveys for the Caracal were carried out in Maharashtra, Madhya Pradesh, Uttar Pradesh and in eastern parts of India. It may be possible that it is present there but under-reported. In that case, the putative reduction of its range needs to be verified and adjusted based on robust data. The common feature of these areas is dry deciduous thorn forest with waterbodies. If a river is present, the ground cover is usually sparse along severely eroded riverbanks and adjacent ravine habitat, while evergreen riparian vegetation is completely absent. While both the Kutch Desert and Ranthambhore Tiger Reserve exhibit dry habitats, 97 small rivers originate in the former and the latter has many rivers and rivulets that criss-cross the landscape and meet the Chambal river (Khandal & Khandal 2017).
Our findings indicate that the Caracal is indeed present in dry open habitats with some vegetation but is absent in 'true' desert as described by Sunquist & Sunquist (2002). This habitat use is consistent with records obtained in arid and semi-arid mountains and hilly terrain in Iran (Farhadinia et al. 2007;Ghoddousi et al. 2009;Moqanaki et al. 2016) and in Uzbekistan (Gritsina 2019).
With the exception of field work carried out by