Distribution and human‐caused mortality of Persian leopards Panthera pardus saxicolor in Iran, based on unpublished data and Farsi gray literature

Abstract Gray literature and data from unpublished sources can provide important scientific information that has not been published scientifically. The Persian leopard (hereafter leopard) Panthera pardus saxicolor is classed as endangered on the Red List of the International Union for Conservation of Nature and also is one of the least‐studied subspecies of leopard. It occurs in the Caucasus and Central and Southwest Asia. Iran contains more than 75% of the leopard's extant range, and the leopard population in this country serves as a source for neighboring countries. In this study, we determined the distribution and human‐caused mortality of leopards in Iran, by reviewing unpublished data and Farsi gray literature (which includes government reports) between 1 January 2010 and 30 December 2018. We created the most recent distribution map of the leopard in Iran. Our data display that human‐caused mortality of leopard in Iran mostly includes poaching and intentional poisoning, and roadkill.


| INTRODUC TI ON
The leopard Panthera pardus has the widest distribution of any large felid species and occurs across much of Africa (except the Sahara Desert) and Asia from the Middle East to the Pacific Ocean (Jacobson et al., 2016). The Persian leopard (hereafter leopard) P. p. tulliana (Valenciennes, 1856;1,039), includes ciscaucasica and saxicolor (Kitchener et al., 2017), one of nine subspecies, is classed as endangered on the Red List of the International Union for Conservation of Nature with an estimated population size of 871 to 1,290 individuals (Stein et al., 2016). It occurs in some central and southwest Asian and Caucasian countries (Stein et al., 2016).
Iran contains more than 75% of leopard's extant range (Jacobson et al., 2016) and serves as a source for neighboring countries (Sanei et al., 2016). Iranian leopards are comprised of a monophyletic clade and are of the same maternal origin (Farhadinia et al., 2015). Their taxonomy as a distinct subspecies still remains obscure until further evaluations, but for the time being this subspecies is traditionally assigned to P. p. saxicolor (Yusefi, Faizolahi, Darvish, Safi, & Brito, 2019), as some studies investigating craniological patterns and molecular genetics showed that Iran is inhabited only by P. p. saxicolor (Farhadinia et al., 2015). According to unsubstantiated estimates, between 550 and 850 leopards are distributed within about 6,050 km 2 in Iran, specifically in two great mountain ranges, the Zagros Mountains in the west and the Alborz Mountains in the north, which makes Iran the main stronghold of the leopard (Erfanian, Mirkarimi, Mahini, & Rezaei, 2013;Sharbafi, Farhadinia, Rezaie, & Braczkowski, 2016;Yusefi et al., 2019).
Gray literature is defined as material published and disseminated by governments, intergovernmental organizations, nongovernmental | 11973 PARCHIZADEH AnD ADIBI organizations, environmental consultancies, private companies, and freelance individuals that (a) is mostly made available on the Internet, (b) contains some useful data that is of lasting scientific value, and (c) is in addition to the scientific literature and not instead of it (Corlett, 2011). Similarly, unpublished data are original data from sources that have never been published. Here, we report leopard presence records across Iran obtained by reviewing unpublished data to create the most recent distribution map of the subspecies. Also, by obtaining mortality data from Farsi gray literature, we document humancaused threats to the survival of leopards in Iran and propose some approaches to their conservation.

| MATERIAL S AND ME THODS
We contacted 31 Provincial Offices of Iran's Department of the Environment (DoE; the governmental organization responsible for conserving Iran's fauna and flora) to obtain unpublished data on records of leopard presence between 1 January 2010 and 30 December 2018. According to the data sources, the presence points were recorded in the form of coordinates (i.e., longitudes and latitudes).
We used 1 January 2010 as our cutoff point, because we were searching for newer data compared to previous studies (Jacobson et al., 2016;Sanei et al., 2016).
We then used the Environmental Systems Research Institute Inc. We conducted a gray literature search on available Farsi Web sites and daily online newspapers using the equivalent search terms "leopard; equivalent Farsi word is ‫ﭘﻠﻨﮓ‬ and "Persian leopard; equivalent Farsi word is ‫اﯾﺮاﻧﯽ‬ ‫ﭘﻠﻨﮓ‬ to access all relevant information about leopard mortality causes in Iran between 1 January 2010 and 30 December 2018. Then, we reviewed every source to avoid any data redundancy. We categorized mortality causes to unnatural, unknown, and natural, and then used unnatural causes for further analyses in our paper.

| RE SULTS
By reviewing unpublished data between 1 January 2010 and 30 December 2018, we obtained 238 records of leopard presence (provided in Table S1). Leopards occurred in every province of Iran, except Kurdistan Province in western Iran (Figures 1 and 2). Mazandaran Province had the highest number of leopard presence records, followed by Semnan and Golestan Provinces ( Figure 2, Table S1).
We reviewed 92 sources from Farsi gray literature and obtained the following results regarding the causes of mortality of leopards between 1 January 2010 and 30 December 2018.
Of 90 leopard mortalities recorded in 22 provinces of Iran (provided in Table S2), unnatural mortality (i.e., shooting, stoning, trapping, intentional poisoning, hunting dogs, and roadkill) was 51%, unknown mortality was 33%, and natural mortality (i.e., old age, intraspecific or interspecific fights, drowning, snakebite, hunger, disease, and internal bleeding) was 16%. Our mortality data consisted of 40% unspecified gender, 31% females, and 29% males. In unnatural mortality category, 76% were poached and intentionally poisoned, and the remaining 24% were killed in car accidents ( Figure 5, Table S2). Mazandaran Province proved to have the highest rate of poaching with 26% (Table S2). Golestan Province showed the highest roadkill rate with 37% (Table S2). In addition, leopards have large home ranges with a mean home range of 103.4 ± SE 51.8 km 2 , and they occur in the mountainous areas of northeastern Iraq's Kurdistan region along the border with Iran and Turkey (Jacobson et al., 2016;Stein et al., 2016). Since the Shahu-Kusalan Protected Area is located adjacent to Iraq's Kurdistan region, it is possible that the leopards have come from this area. Therefore, we recommend camera trap studies to determine whether this is the case. Sanei and Zakaria (2011)  We consider poaching and intentional poisoning, and roadkill, to be the main human-caused threats to the leopard in Iran. Naderi, Farashi, and Erdi (2018) concluded that poaching and intentional poisoning of food eaten by leopards were the main causes of leopard fatalities in Iran. Sanei et al. (2016) concluded that almost 70% (n = 50) of the leopard mortalities during their study period resulted from poaching and poisoning. We found that 76% of the reported human-caused leopard mortalities were due to poaching Livestock grazing is the main threat affecting large mammal distribution (Soofi et al., 2018), and diseases are major threats to livestock ) that make them vulnerable to leopard attacks, resulting in more human-leopard conflicts (Khorozyan, Soofi, Khaleghi Hamidi, Ghoddousi, & Waltert, 2015). The other reason for human-leopard conflicts, which has been overlooked in Iran for many years, is that leopards are killed for their skins. Huge profits can be made from trafficking luxury wildlife goods (Brashares et al., 2014),  Reyahi Khoram, Rizvandy, & Reyahi Khoram, 2014). Leopards who suffer from a lack of sufficient wild prey, in addition to old and young dispersing leopards attack domestic dogs and livestock (Babrgir, Farhadinia, & Moqanaki, 2017;. In response, rural residents and shepherds use both poisonous lures and guns to kill leopards to defend their livestock (Jacobson et al., 2016). The longterm survival of the leopard is worrisome, given the ongoing escalated rate of human-leopard conflicts as an important threat (Iranian Cheetah Society, 2014), as the loss of every leopard is a major setback in efforts for safeguarding this subspecies (Maharramova et al., 2018). Therefore, the Ma Insurance Company (https ://www. bimeh ma.com/), which is a for-profit insurer, started contributing to leopard conservation free of charges by establishing a program to compensate for any losses of livestock and herder dogs inflicted by this predator in Iran (Sanei et al., 2016;The Tehran Times, 2017).

| Poaching and intentional poisoning
In

| Roadkill
Roadkill has been reported to be the second most important humancaused threat to the leopard in Iran (Naderi et al., 2018;Sanei et al., 2016  The Persian leopard is one of the least-studied subspecies of leopard (Babrgir et al., 2017), and without taking active, targeted, and large-scale conservation measures, it is in imminent danger of extinction (Khalaf-von Jaffa, 2008;Khorozyan, Baryshnikov, & Abramov, 2006;Khorozyan, Malkhasyan, & Asmaryan, 2005). The

| CON CLUS IONS
Caucasian Leopard Working Group concluded that, without recovery of the Iranian populations (i.e., leopard and its prey), little hope is left for natural recolonization of the leopard across the Caucasus (Moqanaki, Breitenmoser, Kiabi, Masoud, & Bensch, 2013). In addition, Breitenmoser et al. (2014) concluded that not only does the high profile of leopard mean that it is a flagship for conservation in the region, but its ecological requirements, large areas of habitat and adequate prey, mean that efforts to restore the leopard will help conserve other threatened fauna and flora. Therefore, we suggest the DoE allocate necessary resources for conducting future surveys on the leopard populations in Iran and increase effective conservation measures, before it is too late.

ACK N OWLED G M ENTS
We would like to thank Dr. Jon E. Swenson for his full support and assistance in reviewing the manuscript. Our special thanks go to Dr.
Mohammad S. Farhadinia for his support during the review process.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N S
Jamshid Parchizadeh designed the work, obtained the Persian leopard mortality data, analyzed and interpreted the data, prepared the figures, and wrote the draft and final versions of the manuscript.
Mohammad Ali Adibi obtained the Persian leopard presence records and reviewed the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors agree to deposit their data in Dryad is https ://doi. org/10.5061/dryad.pp281n6.