Amphibian fauna of Pakistan with notes on future prospects of research and conservation

Abstract Research on amphibians and their conservation have gained worldwide attention, as the group includes the highest number of threatened and Data Deficient species when compared to other vertebrates. However, amphibians have long been neglected in wildlife conservation, management decisions, policy making, and research agendas in Pakistan. In this paper, an annotated checklist of the 21 amphibian species of Pakistan, a key to their identification, and detailed discussions on variation in species, including the genera Minervarya and Sphaerotheca, are provided. We found a statistically significant difference in the morphometric measurements of males but non-significant difference in the females of the two forms (rusty dorsum and dotted dorsum) of S.maskeyi. Some genera, such as Microhyla, Uperodon, Minervarya, Allopaa, Chrysopaa, Euphlyctis, Nanorana, and Sphaerotheca, in Pakistan are in need of additional data for molecular and morphological comparisons with taxa in other South Asian countries. The predicaments of amphibian research in Pakistan are discussed, gaps identified, and suggestions are made. Although the occurrence of chytrid fungus in Pakistan is predicted of low likelihood, a lack of data merits studying the prevalence of the fungus, particularly in the northern regions of the country which exhibit complex and dynamic ecosystems. It is recommended that systematic and coordinated surveys are conducted throughout the country to build a database of species occurrences and distributions. Additionally, the monitoring of wild populations and threat mitigation, as well as appropriate legislation, are suggested as long-term measures. By adopting an inclusive wildlife conservation approach in Pakistan, amphibians could be integrated into wildlife conservation and management efforts.


Introduction
Amphibians are bioindicators of an ecosystem's health and may also serve as a biological control of crop and forest pests (Attademo et al. 2005;Kanagavel et al. 2017). Additionally, various important compounds have also been extracted from their skin and eggs for medical applications (Erspamer 1971;Clarke 1997). Amphibians are sometimes kept as pets (Gerson 2012) and are also a source of food (protein) for people in many regions of the world (Marineros 2007). The number of currently described amphibian species is 8378 (Frost 2021).
The First Herpetological Congress, organized in 1989, presented alarming findings about the decline in amphibian populations which was presumed to have started in the early 1970s in the United States, certain Central American countries, and in northeastern Australia (Czechura and Ingram 1990;Drost and Fellers 1996;Burrowes et al. 2004). Amphibians include the highest number of Data Deficient species (>1500 species) (Morais et al. 2013) and the highest percentage (>40 %) of threatened species among all vertebrate groups. Bishop et al. (2012) categorized threats to amphibians into two groups. The first group of threats included habitat destruction and fragmentation, exotic invasive species, and over-exploitation. The second group, which is more poorly understood, includes the threats of infectious diseases and global climate change. Approximately 700 amphibian species are known to have been affected globally by the chytrid fungus, Batrachochytrium dendrobatidis. This fungus has extirpated about 90 amphibian species and has caused population declines of over 500 species (Rosenblum et al. 2010;Lips 2016;Scheele et al. 2019).
This paper provides an annotated checklist of the 21 amphibian species of Pakistan and keys to their identification. The predicaments of amphibian research in Pakistan are discussed and knowledge gaps identified. Suggestions are made on how to proceed with research and conservation of amphibians in the country.

Materials and methods
The available historical as well as recent literature on the amphibians of Pakistan was critically reviewed. We collected data on the morphology of 10 amphibian species (N = 158) (Suppl. material 1, Table S1) beginning in 2015 from the areas of Rawalpindi, Islamabad, and Gilgit-Baltistan. We used published literature (Murray 1884;Khan and Tasnim 1989;Auffenberg and Rehman 1997;Dutta 1997;Stöck et al. 1999;Khan 2006;Dufresnes et al. 2019;Ali et al. 2020) on other species in the development of the identification keys.
We studied morphological differentiation of the two forms of Sphaerotheca maskeyi: uniform rusty-colored dorsum (n = 9, Fig. 3F) and dorsum olive with dotted pattern (n = 29, Fig. 3E). We performed a principal components analysis (PCA) on 23 morphometric measurements separately on males and females (Borzée et al. 2013) in XLSTAT to reduce the studied measurements into fewer significant variables (r > 0.90) (see variable 1-23 in Suppl. material 1, Table S2a, S2b). Principal components analysis (PCA) is a variable-reduction technique that shares many similarities to exploratory factor analysis. The aim is to reduce a larger set of variables into a smaller set of "artificial" variables, called "principal components", which account for most of the variance in the original variables. We then conducted a multivariate generalized linear model (one-way MANOVA) to examine if there were any differences (a = 0.05) between categorical predictor variables in the two forms (in males and females separately) on continuous response variables (obtained after PCA with r > 0.90) in SPSS 22.

Statistical comparisons of Sphaerotheca populations
Of the studied 23 morphometric measurements in S. maskeyi, we obtained from our PCA 10 and eight significant variables (r > 0.90) in males having uniform rusty-colored dorsum and having dotted pattern, respectively. We obtained 10 and one significant variables (r > 0.90) in females, respectively. Eigen value, variability (%), cumulative variability (%), and factor loadings of the 23 morphometric measurements are given Table 1.

Discussion
The true toads of Pakistan are represented by two genera: Duttaphrynus Frost et al., 2006 andBufotes Rafinesque, 1815. Duttaphrynus is characterized by prominent ridges on the head, while Bufotes lacks such ridges but bears conspicuous pattern of irregularly shaped, darker, green or greenish-olive spots. Considering other taxa, Faiz et al. (2018) reported three amphibians, including Allopaa barmoachensis from Toli Pir National Park, Pakistan. Dubois (1992) considered A. barmoachensis synonymous with Allopaa hazarensis, but Khan (2004) regarded the two as distinct. However, Ohler and Dubois (2006) reiterated that the species is conspecific with A. hazarensis. As no molecular data exist to separate the two species, there is no evidence for separation. The species complex of Euphlyctis also needs detailed study. Dutta (1997) provided a record of Euphlyctis hexadactylus from Pakistan which needs confirmation. Murray (1884) reported and described Tomopterna strachani from Sindh, Pakistan, and Khan (2006) reported the species as Sphaerotheca breviceps. Dubois (1999) suggested that S. breviceps is a small-sized species based on the study of the name-bearing specimens. For the large-sized species of South Asia, he suggested that the names Rana variegata (Gravenhorst, 1829); Pyxicepha- lus fodiens (Jerdon, 1853); Pyxicephalus pluvialis (Jerdon, 1853); Sphaerotheca strigata (Günther, 1859); Rana dobsonii (Boulenger, 1882); Tomopterna strachani (Murray, 1884); Rana leuchorhynchus (Rao, 1937); and Rana swani (Myers and Leviton, 1956) are available. Dubois (1999) also regarded Tomopterna maskeyi to be a provisional synonym of these large-size taxa. Recently, Deepak et al. (2020) provided distribution records for Sphaerotheca pashchima from India and morphological descriptions of their samples of S. pashchima match samples collected for the present study. Deepak et al. (2020) reported similar morphological variation in S. pashchima among the samples collected from India. Khatiwada et al. (2021) demonstrated high similarity between topotypical material of S. maskeyi with name-bearing types of S. pashchima and considered the later name a synonym of T. maskeyi, valid as S. maskeyi. Molecular identification of our samples also confirms their identity as S. maskeyi (see Akram et al. 2021). Therefore, we conclude that Sphaerotheca maskeyi occurs in Pakistan, and not Sphaerotheca breviceps as reported by Murray (1884) and Khan (2006). Borthakur et al. (2007) studied cricket frog species in Assam, northwest India (Fejervarya nepalensis, F. pierrei Dubois, 1975, F. syhadrensis Annandale, 1919, and F. teraiensis Dubois, 1984), which have been assigned to other genera in Nepal  Table 1. Eigen value, variability (%), cumulative variability (%) and factor loadings of the 23 morphometric measurements of the two forms (uniform rusty-colored dorsum and dorsum olive with dotted pattern) of Sphaerotheca maskeyi. The factor loadings with absolute correlation values greater than 0.90 were considered significant (in bold).  Ahmed et al. (2009) andTiwari (2004). Rawat et al. (2020) reported Minervarya species from extreme southwestern Nepal in the Shuklaphanta National Park. Molecular identification of our samples confirms their identity as Minervarya pierrei (see Akram et al. 2021). Two distinct forms of Minervarya are known, one with a mid-dorsal stripe and another without it. Dubois (1974) has reported such variation. Hence, we suggest conducting country-wide surveys and use a molecular approach to confirm presence of other species of Euphlyctis, Sphaerotheca, and Minervarya from Pakistan. The inclusion of Uperodon systoma in the list of amphibians of Pakistan is based on two reports. Baig and Gvozdik (1998) reported this species from a torrent stream in the Shakarparian Hills, Islamabad Capital Territory (ICT), and Masroor (2011) recorded this species from a subtropical, semi-evergreen forest in Margalla Hills National Park (ICT). We consider U. systoma to be very rare in Pakistan. No historical quantitative data has been found to date. Some species assessed as Least Concern by the IUCN, such as U. systoma, are considered rare in Pakistan, compared to elsewhere in their global range. Hence, we caution the use of global conservation status for the amphibian species that occur within Pakistan.

Future prospects in amphibian research and conservation in Pakistan
Pakistan represents the westernmost limit of the geographic range of Duttaphrynus melanostictus. This species has been introduced outside its natural range into many parts of the world, and in these places it is considered a nuisance predator, a potential disease vector, and the cause of many other ecological problems (Labisko et al. 2015;Piludu et al. 2015). Studying the ecology and biology of D. melanostictus in its native range could help manage this species in Pakistan as well as elsewhere.
The chytrid fungus Batrachochytrium dendrobatidis affects amphibians worldwide. The likelihood of this fungus occurring in Pakistan is predicted to be low (<30%) (Olson et al. 2013;Rodder et al. 2010) by models which did not include samples of anurans from Pakistan. This lack of data may produce inaccurate results in models, which use no direct observational data. Therefore, the study of the prevalence of chytrid fungus in countries such as Pakistan is important to fill in these data gaps. Furthermore, the northern regions of Pakistan have complex and dynamic ecosystems (Roberts 1997) and therefore more diverse amphibian assemblages. Diversity of amphibians in an ecosystem has been linked to increased probability of the introduction and spread of chytrid fungus (Olson et al. 2013). This correlation with amphibian diversity and the lack of data in the Middle East and South Asia creates an urgency to perform risk assessments on amphibian communities in these regions.
There is also a dire need to change social attitudes towards amphibians in our society. This could be achieved by initiating community awareness by outreach, school, and citizen-science programs. While designing research projects, special attention should be given to include components of outreach. For instance, people working in agroecosystems can organize field activities with farmers and local communities. Likewise, the ongoing 10 Billion Tree Tsunami project by the Ministry of Climate Change, Government of Pakistan, should integrate consideration for herpetofauna species, particularly anuran species such as Allopaa hazarensis and Allopaa barmoachensis, which are endemic to forested montane wetlands. The development of android applications and websites could help reach out to the public. This, however, would be limited to those people who have access to the internet, but their participation would inevitably enhance the documentation of species occurrence and distribution records in the country. Collection and archiving quantitative data on anuran abundance would also help determine the current conservation status of our anuran species.
We suggest setting research priorities and to devise strategy for the conservation of amphibians of Pakistan when manageable anthropogenic threats exist, such as habitat destruction, urbanization, pollution, and unsustainable utilization, so that amphibian populations can be better controlled by utilizing less financial, administrative, and human resources. This can be achieved through short-, medium-, and long-term actions. Short-term actions could include the establishment of a network or people currently engaged in amphibian related research. A conservation assessment and management plan workshop should be organized wherein experts and researchers could provide their opinions and draft recommendations for medium-and long-term actions.
A medium-term action plan may include carrying out systematic and coordinated surveys throughout the country to establish a database on occurrence and distribution of species and the identification of their threats. It is recommended to use modern taxonomic tools, such as DNA barcoding, to determine taxonomy and initiate research on phylogenetic affinities, biogeography systematics, especially on endemic species. This approach can expect to yield additional amphibian species as a result. Some genera, such as Microhyla Tschudi, 1838, Uperodon Duméril & Bibron, 1841, Minervarya Dubois et al., 2001, Allopaa, Chrysopaa Ohler & Dubois, 2006, Euphlyctis Fitzinger, 1843, Nanorana, and Sphaerotheca Günther, 1896, which occur in Pakistan need additonal data for molecular taxonomy and detailed comparisons with taxa in other South Asian countries.
Long-term actions would entail monitoring of amphibian populations, threat mitigation, and appropriate legislation. Amphibians have been excluded from all current legislative and policy decisions of the country. The National Climate Change Policy (GoP 2012), the Pakistan National Biodiversity Strategy and Action Plan (GoP 2015), the Biodiversity Action Plan of Pakistan (GoP 2000), and the Pakistan National Conservation Strategy (GoP 1992) do not currently support the need to carry out research and conserve amphibians. Likewise, amphibians are not protected under any law (Shafiq 2005). Hence, the legislation pertaining to threatened and endemic species needs to be updated, particularly in need of revision is Schedule III, which includes protected species, of provincial and federal wildlife laws as well as the CITES appendices.
Wildlife conservation projects in Pakistan mainly focus on carnivores, ungulates, and birds. Shehzad et al. (2012) reported the occurrence of Nanorana vicina in the diet of Prionailurus bengalensis in Ayobia National Park, Khyber Pakhtunkhwa, Pakistan. Such studies usually lack the mandate of investigating whether a particular food item was eaten directly or through an alternate dietary item. Whatever the case, this explicitly signifies the role of amphibians in the food chain and could be used as an impetus to incorporate amphibians in such research projects and conservation programs. Therefore, it should be proposed to adopt an inclusive wildlife conservation approach in Pakistan. The approach would advocate the integration of poorly documented taxa, such as amphibians, in wildlife conservation and management projects to enhance the significance of their existence and the intrinsic values of all wildlife species which would eventually ensure their continued survival.

Availability of data
The data underpinning the analysis reported in this paper are deposited in the Dryad Data Repository at Dryad (https://doi.org/10.5061/dryad.mkkwh7118).