Ge Nature YW) Conservation Nature Conservation 55: 67-82 (2024) DOI: 10.3897/natureconservation.55.114746 Research Article Climatic niche modelling and genetic analyses highlight conservation priorities for the Spotted Softshell Turtle (Pelodiscus variegatus) Minh Duc Le'??©, Dennis Rédder*®, Tao Thien Nguyen®®, Cuong The Pham®7®, Truong Quang Nguyen®”®, An Vinh Ong®®, Timothy E. M. McCormack”, Thang Tai Nguyen™®, Mai Huyen Le’®, Hanh Thi Ngo™2®, Thomas Ziegler’'2® 1 Department of Environmental Ecology, Faculty of Environmental Sciences, University of Science, Vietnam National University, 334 Nguyen Trai Road, Thanh Xuan District, Hanoi, Vietnam oe AN Do FP WwW DY Central Institute for Natural Resources and Environmental Studies, Vietnam National University, 19 Le Thanh Tong, Hoan Kiem District, Hanoi, Vietnam Department of Herpetology, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, USA LIB, Museum Koenig Bonn, Leibniz Institut zur Analyse des Biodiversitatswandels, Adenauerallee127, 53113 Bonn, Germany Institute of Genome Research, Vietnam Academy of Science and Technology, Hanoi 10072, Vietnam Institute of Ecology and Biological Resources, Vietnam Academy of Science and Technology, Hanoi 10072, Vietnam Graduate University of Science and Technology, Vietnam Academy of Science and Technology, Hanoi 10072, Vietnam Department of Zoology, Vinh University, 182 Le Duan St., Vinh City, Nghe An Province, Vietnam Asian Turtle Program - Indo-Myanmar Conservation, Room 1806, C14 Bac Ha Building, To Huu Road, Nam Tu Liem District, Hanoi, Vietnam 10 Department of Genetics, Faculty of Biology, University of Science, Vietnam National University, 334 Nguyen Trai Road, Thanh Xuan District, Hanoi, Vietnam 11 Institute of Zoology, University of Cologne, Ziilpicher Strasse 47B, 50674 Cologne, Germany 12 Cologne Zoo, Riehler StraBe 173, 50735 Cologne, Germany Corresponding authors: Minh Duc Le (le.duc.minh@hus.edu.vn); Thomas Ziegler (ziegler@koelnerzoo.de) OPEN ro. ACCESS Academic editor: Franco Andreone Received: 25 October 2023 Accepted: 30 January 2024 Published: 26 February 2024 ZooBank: https://zoobank. org/OA9E2BDD-283C-4C 56-BF2B- 9CF30658213A Citation: Le MD, Rodder D, Nguyen TT, The Pham C, Nguyen TQ, Ong AV, McCormack TEM, Nguyen TT, Le MH, Ngo HT, Ziegler T (2024) Climatic niche modelling and genetic analyses highlight conservation priorities for the Spotted Softshell Turtle (Pelodiscus variegatus). Nature Conservation 99: 67-82. https://doi.org/10.3897/ natureconservation.59.114746 Copyright: © Minh Duc Le et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International - CC BY 4.0). Abstract The Spotted Softshell Turtle (Pelodiscus variegatus) has been recognised since 2019 from Vietnam and Hainan Island, China, but little information about its population status and distribution range is currently available. The species has been provisionally listed as Crit- ically Endangered by the Turtle and Tortoise Working Group, although the status has not been officially accepted by the IUCN, due to the threats the species is facing, including habitat loss and degradation, overexploitation for food, competition with other non-na- tive softshell turtles and pollution. To identify conservation priority sites for P variegatus in mainland Indochina, this study combines molecular analyses and species distribution modelling. Our results show that, in Vietnam, Phong Nha-Ke Bang National Park has the largest suitable area and high probability of species occurrence, followed by Vu Quang National Park and Song Thanh and Ke Go Nature Reserves. In addition, the central prov- inces, from Thanh Hoa to Thua Thien Hue in Vietnam, constitute a key part of the species distribution and should be prioritised for conservation actions. According to the study's findings, although P variegatus is possibly found in Laos, the probability decreases sharply at the border between both countries and there is also a gap in the occurrence of wetlands, arguing for strong natural barriers. Unfortunately, to date, only part of the species potential distribution is protected, while no records are known from protected areas, highlighting the need for extended or even new reserves. To recover natural populations of the species and following the IUCN’s One Plan Approach to Conservation, breeding programmes have been established in Vietnam with a potential to expand to other facilities in the country and abroad. Once suitable sites are identified, offspring can be released into the protected ar- eas to improve the current conservation status of this highly-threatened softshell turtle. 67 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Key words: Cytochrome b, Laos, ND4, species distribution modelling, Vietnam Introduction The Chinese Softshell Turtle, Pelodiscus sinensis, was described nearly 200 years ago and was believed to represent a morphologically variable, geograph- ically widespread taxon (from the Russian Far East through the Korean Penin- sula, eastern and central China to Vietnam). The Northern Chinese Softshell Turtle, P maackii (Brandt, 1857), was described 23 years later, but was then thought to be a synonym of P sinensis. Only 35 years ago, the populations from the northernmost part of the P sinensis distribution range were shown to repre- sent a distinct species, based on osteological differences (Chkhikvadze 1987). Two additional species of this complex from central China were described in the 90s, based on morphological differences: the Hunan Softshell Turtle (P. axe- naria) and the Lesser Chinese Softshell Turtle (P. parviformis) (Zhou et al. 1991; Tang 1997). Recent molecular studies confirmed that the genus Pelodiscus constitutes a species complex (Fritz et al. 2010, see also Stuckas and Fritz (2011); Yang et al. (2011); Gong et al. (2018)). Based on genetic and morpho- logical analyses, three new taxa were described: the Spotted Softshell Turtle, P. variegatus from northern Vietnam and Hainan (China), the Horse-hoof Soft- shell Turtle, P huangshanensis from southern Anhui Province of China and the Chinese Stone Slap Softshell Turtle, P. shipian from Jiangxi and Hunan Provinc- es of China (Farkas et al. 2019; Gong et al. 2021; Gong et al. 2022). Thus, the Pelodiscus sinensis complex at the moment comprises seven species, with six species being distributed in China and four endemic to the country (Gong et al. 2021). The revisions have consequences on the species conservation, as taxonomic splitting implies that the range size and number of individuals decrease for each species. However, these new research results have yet to be reflected in the conservation assessment of the genus. Accord- ing to the IUCN (2023), only P. sinensis is listed as Vulnerable, but the data are greatly outdated with the last update in 2000 (Asian Turtle Trade Working Group 2000). The Turtle Taxonomy Working Group (TTWG) recently evaluated conservation status of six species in the genus and provisionally categorised three species as Data Deficient (P. axenaria, P. huangshanensis and P. maackii), one as Vulnerable (P. sinensis) and two as Critically Endangered (P. parviformis and P. variegatus; TTWG 2021). The last species, P. shipian, has not been as- sessed by any previous study. The exact range of the recently-described P. variegatus is still largely un- known, but historical records suggest that the species occupies lowland ar- eas of central and northern Vietnam and parts of southern China, viz. Hainan Province (Farkas et al. 2019; TTWG 2021). Recently, Ziegler et al. (2020) in- vestigated whether natural populations of P. variegatus still exist, since the de- scription of this species was mostly based on historical museum specimens. To find potential members of P. variegatus, this study focused on surveys of central lowland freshwater habitats, as well as local markets, restaurants and farms in Vietnam. Individuals with the species-specific dark blotched plas- tron pattern were identified and subsequently genetically screened to confirm their identity. By using this approach, Ziegler et al. (2020) demonstrated that Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 68 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle P. variegatus is still extant in the wild in Vietnam, both based on evidence from the trade and on surveys in the natural habitat. The study also showed that P variegatus occurs in the central provinces of Thanh Hoa, Nghe An, Ha Tinh and Quang Binh, primarily in lakes with flat shores consisting of mud and soft soil, rivers in agricultural landscape and medium-sized streams in secondary forests. As softshell turtles are common and prized as food where they occur, natu- ral populations are threatened by local hunting, with further threats of habitat loss and competition with introduced softshell turtles (Shi et al. 2008; Le Duc et al. 2020). At the moment, there is little evidence of any viable population of the P. variegatus existing in the wild. It is, therefore, imperative to implement conservation measures in priority areas, where natural populations still like- ly survive. To establish conservation priorities for the species in Vietnam, the present study performed climatic niche modelling, based on comprehensive distribution data derived from detailed molecular analyses of existing and new samples collected from the country, the Emys system (emys.geo.orst.edu), the Turtle Taxonomy Working Group (TTWG 2021) and the Global Biodiversity In- formation Facility (GBIF). Material and methods Molecular analyses To identify new samples collected in the field, we used a molecular approach. In total, 61 newly-collected samples were included in the analyses (Suppl. ma- terial 1). Sequences of other Pelodiscus species were downloaded from Gen- Bank. Two species, Dogania subplana and Palea steindachneri, were employed as outgroups (Le et al. 2014). We used the protocols of Le et al. (2006) for DNA extraction and amplification. Two mitochondrial genes, the nearly complete cy- tochrome b (1110 bp) and partial ND4 (673 bp), were amplified using primers listed in Table 1. Successful amplifications were purified to eliminate PCR com- ponents using GeneJET™ PCR Purification Kit, Thermo Fisher Scientific (Vilni- us, Lithuania). Purified PCR products were sent to 1*t BASE (Selangor, Malaysia) for sequencing. Sequences generated in this study were aligned using De Novo Assemble function in the program Geneious v.7.1.8 (Kearse et al. 2012). Data were then analysed using Maximum Likelihood (ML) as implemented in 1Q-TREE v.1.6.7.1 (Nguyen et al. 2015) and Bayesian Inference analysis (B)), as implemented in MrBayes v.3.2.7 (Ronquist et al. 2012). For ML analysis, we used a single model and 10,000 ultrafast bootstrap replications. The optimal Table 1. Primers used in this study. Primer Sequence Reference Gludg (f) 5'- TEACTTGAARAACCAYCGTTG - 3' Palumbi (1996) CB3 (r) 5'- GGCAAATAGGAAATATCATTC - 3' Palumbi (1996) CB534 (f) 5'- GACAATGCAACCCTAACACGE- 3' Engstrom et al. (2004) Tcytbthr (r) 5'- TTCTTTGGTTTACAAGACC - 3' Engstrom et al. (2004) ND4 672 (f) 5'- TEACTACCAAAAGCTCATGTAGAAGC - 3' Engstrom et al. (2002) Hist (r) 5'- CCTATTTTTAGAGCCACAGTCTAATG - 3' Arévalo et al. (1994) Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 69 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle model for nucleotide evolution employed in both methods was determined us- ing jModelTest v.2.1.4 (Darriba et al. 2012). The optimal model for nucleotide evolution was set to TIM2+G for ML and single-model Bayesian analyses. For the Bayesian Inference, two independent analyses with four Markov chains (one cold and three heated) were run simultaneously for 10 million generations with a random starting tree and sampled every 1000 generations. Log-likelihood scores of sample points were plotted against generation time to determine stationarity of Markov chains. Trees generated before log-likelihood scores reached stationarity were discarded from the final analyses using the burn-in function. The posterior probability values for all clades in the final majority rule consensus tree were provided. The cut-off point for the burn-in function was set to 71 in the Bayesian analysis, as —InL scores reached stationarity after 71,000 generations in both runs. Nodal support was also evaluated using ul- trafast bootstrap (UFB) in 1Q-TREE and posterior probabilities (PP) in MrBayes. PP and UBP = 95% were regarded as strong support for a clade (Ronquist et al. 2012; Nguyen et al. 2015). This study only employed mitochondrial genes to provide taxonomic identi- fication of samples collected from the wild. Although maternally inherited mi- tochondrial loci cannot help detect hybridisation events, interbreeding between different species of softshell turtles has only been reported in turtle farms (Gong et al. 2018). In addition to genetic screening of captured animals, we morphologically identified the specimens using diagnostic characters. Species records and environmental variables For species records, we carefully checked potential records listed in recent stud- ies, including, Le Duc et al. (2020), Pham et al. (2020), Ducottend et al. (2023), Pham et al. (2023a) and Pham et al. (2023b). However, the papers followed the older version of the Turtle of the World Checklist (TTWG 2018), which omitted information about P. variegatus because the species was only described a year later in 2019. As a result, the records were all assigned to P. sinensis, although the species is now considered only occurring in China and Taiwan according to the new checklist (TTWG 2021). In addition, much of the information came from interviews with local people and is not taxonomically robust. In Pham et al. (2023b), some photos clearly belong to the P. sinensis form, while the others did not show diagnostic characters to allow identification of these individuals. We, therefore, did not add the data to our analyses. In the final dataset, we used 54 unique, georeferenced and genetically confirmed locations located in unique grid cells, as shown in Fig. 2 (see below). For environmental predictors, we used a combination of weather station-de- rived precipitation data and remote sensing data. The full list of variables in- cluding their interpretation is provided in Table 2. Average annual characteris- tics of precipitation regimes were obtained from the Worldclim database 2.1 as interpolated elements from different climate conditions collected over a period of 30 years (1970-2000) with a resolution of 30 arc seconds (Hijmans et al. 2005; Fick and Hijmans 2017). In order to characterise seasonal changes in land surface temperatures and in vegetation cover, we used 27 remote sens- ing predictors derived from the Moderate Resolution Imaging Spectroradiom- eter (MODIS) sensors of two NASA satellites available through the EDENext Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 70 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle project (temporal resolutions: 8-day averages (MOD11A2) and 16-day averages (MCD43B4), spatial resolution 30 arc seconds) (Mu et al. 2007). Availability of wetlands as surrogate of suitable microhabitats was assessed using a re- cent assessment of tropical wetlands provided by Gumbricht et al. (2017) as a categorical variable. As the wetlands dataset lacks river networks which may provide suitable microhabitats for the turtle, we added a high-resolution water layer as additional category (GRDC 2020). The wetlands dataset was resam- pled to the spatial resolution of the remote sensing predictors and only grid cells including water bodies were considered. Multi-collinearity amongst predictors may hamper successful model training and subsequent projection (Brun et al. 2020). Therefore, we computed pairwise Spearman rank correlations and selected each predictor pair with p? > 0.75 with only one predictor for final model training. The final set of variables comprised both climate and land-cover related variables, which are suitable to character- ise the species range limits and microhabitat preferences. As these operate on very different scales, we used a two-step approach for modelling: (1) determin- ing likely range limits with temperature and precipitation related variables and (2) using these range limits to train a second model using land-cover related variables to assess microhabitat preferences. For the first climate related mod- el, the variables comprised the annual mean surface temperature (ED15078_ bio1), maximum surface temperature of the warmest month (ED15078_bi06), surface temperature annual range (ED15078_bio7), annual mean precipitation (bio_12), precipitation of the driest month (bio_14) and precipitation of the warmest quarter (bio_18). Quantification of microhabitat preferences was de- rived from the annual mean Normalised Difference Vegetation Index (NDVI; ED1514_bio1), annual range of NDVI (ED1514_bio7), annual mean Enhanced Vegetation Index (EVI; ED1515_bio1), annual range of EVI (ED1515_bio7) and the categorical map of wetlands including the categories open water, man- groves, swamps, flood-out swamp, fens, riverine, floodplain, marshes, marsh- es—dryland/wetland and marshes—wet meadows. Species distribution modelling As the algorithm for climatic niche modelling development, we used Maxent v.3.4.0, which is specifically designed to derive potential distributions from presence-pseudoabsence data (Phillips et al. 2006; Phillips et al. 2017) and which can perform well even if the sample size is comparatively small (Hernan- dez et al. 2006). For the first model, we chose an area defined by a minimum convex polygon buffered with 5 km enclosing all species records as environ- mental background. Model selection followed the procedure described in Ginal et al. (2022). In Maxent, we allowed only linear, product and quadratic feature classes and used a regularisation multiplier of 0.8, as theses settings had, on average, the minimum delta AlCc (689.4 with 8 parameters) and revealed the most realistic response curves. Using these optimal settings for feature class- es and regularisation parameter and a bootstrap approach, we computed 100 models, each trained with 80% of the species records and using the remaining 20% for model evaluation via the area under the ROC (Receiver Operating Char- acteristic) curve [AUC] (Swets 1988). The average across all 100 replicates in cloglog format was used for further processing. Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 71 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle For the second land-cover niche modelling, we reclassified the potential distribution suggested by the first model applying the minimum training pres- ence threshold, which was used as environmental background. Model selec- tion followed again Ginal et al. (2022) and internal settings in Maxent were set to a regularisation parameter of 0.7 and linear, product and quadratic feature classes. The regularisation parameter of the categorical wetland predictor was set to 0.250. The average delta AlCc was 760.6 with 10 parameters. Again, the average across all 100 replicates in cloglog format was used for further processing. The joint effects of climatic suitability and microhabitat suitability were estimated by rescaling both average predictions to a scale of 0-1 after applying the minimum training presence threshold and multiplying both. The resulting map highlights areas where both climatic and microhabitat suitability are highest. Defining conservation priority sites We merged the occurrence data with existing protected areas (Reserves, Nation- al Parks etc.) in the country. Information of protected areas was obtained from the world dictionary of protected areas/protected planet (https://www.protect- edplanet.net). We selected the targeted area in north-central and central Viet- nam because most of distribution records were reported from the region. In total, there were 42 potential conservation units within the general area and, for each Reserve, we computed the number of suitable grid cells, the sum of probabilities and the mean probability in QGIS 3.14. Map resolution ca. 1 km (30 arc sec). Results The molecular matrix contained 1921 aligned characters. Both BI and ML anal- yses showed that new samples belong to Pelodiscus sinensis and P. variega- tus. The former species was only moderately supported (PP = 89%, UBF = 90%), while the latter received strong support from both BI and ML (PP = 100%, UBF = 99%). In total, 61 newly-collected and four GenBank samples were iden- tified as P. variegatus (see Suppl. material 1: fig. S1, for the full tree). The local- ities of the samples were used for the species distribution modelling. The niche modelling trained with only climatic variables had a good overall performance (AUC,,.. = 0.79; AUC, ning = 0-84, Suppl. material 1: fig. S2). The pre- dictor with the highest explanative power was precipitation of the driest month (bio_14: 77%), followed by the maximum surface temperature of the warmest month (ED15078_bio6: 11.8%), precipitation of the warmest quarter (bio_18: 5.3%), surface temperature annual range (ED15078_bio7: 3.9%), annual mean surface temperature (ED15078_bio1: 1.4%) and annual mean precipitation (bio_12: 0.7%). Based on climatic conditions, the potential distribution covers major parts of central Vietnam, wherein the probability of occurrence sharply decreases towards Laos (Fig. 2A). The niche modelling, trained with microhabitat variables, had a good overall performance (AUCtest = 0.73; AUCtraining = 0.80, Suppl. material 1: fig. S3). The predictor with the highest explanative power was the categorical map of wetlands (60.3%) (open water, mangroves, swamps, flood-out swamp, fens, riverine, floodplain, marshes, marshes—dryland/wetland and marshes—wet Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 72 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle meadows), followed by annual mean NDVI (ED1514_bio1: 6.6%), annual mean EVI (ED1515_bio1: 3.6%), annual range of EVI (ED1515_bio7: 1.9%) and annual range of NDVI (ED1514_bio7: 1.7%). Density of suitable microhabitats is high- est near the coast, with some suitable wetlands/water bodies in higher eleva- tions towards Laos (Fig. 2B). When integrating the probabilities of occurrence derived from climatic and land-cover variables, the most suitable habitats for P. variegatus are near the Vietnam coastline, where extensive freshwater wetlands exist (Fig. 2C). Rivers and other water bodies in mountainous areas are partly suitable, but they cover much less area. Only part of the potential distribution for P. variegatus in Vietnam is protect- ed and no known occurrence is directly located within reserves, although they are close, such as nearby Ke Go Nature Reserve and Vu Quang National Park (Fig. 2D). Table 3 has information on the relative ranking of protected areas, based on their IUCN status, the total area suitable for P variegatus, the sum of potentially suitable sites and the average potential for the species. The Table is sorted with descending total suitable areas per Reserve. Discussion Farkas et al. (2019) stated that, in Vietnam, most records of P. variegatus fall within the “Northeast Lowlands Subregion” of Bain and Hurley (2011). The zoogeographical affinities of Hainan are closely related to this area as well as mainland south-western China, while the southern portion of the purported range forms part of the “Central- South Vietnam Lowlands Subregion” as de- fined by Bain and Hurley (2011). However, our molecular data show that most distribution localities of the species occur in the north-central region of the country, except for records from Dong Mo Lake (Fig. 1). It is likely that a major- ity of extant populations of the species is restricted to this region in Vietnam, although it is possibly found in Laos, based on climatic niche modelling results. However, the probability based on climate decreases sharply at the border be- tween both countries and there is also a gap in the occurrence of wetlands. Thus, the border area may represent a strong natural barrier. Our habitat suitability analysis predicts that two most important protected ar- eas for P variegatus include Phong Nha-Ke Bang and Vu Quang National Parks in Vietnam. Other protected sites with the largest suitable sizes consist of Song Thanh and Ke Go Nature Reserves and Ben En National Park (Table 3, Fig. 2). In terms of Average Probability, Hue Saola Nature Reserve receives the highest value (0.422), followed by Phong Dien Nature Reserve, Bach Ma National Park and Ke Go Nature Reserve (Table 3). The findings show that the central provinc- es from Thanh Hoa to Thua Thien Hue form an important part of the species distribution. Although one record in our study suggests that the species occurs in Laos (Fig. 2), its approximate field coordinates could not be used to abso- lutely confirm the species’ presence in the country. It is, therefore, essential to conduct field surveys at suitable sites in Laos to verify its presence or absence. Prior to its discovery, PR variegatus was considered part of P. parviformis. The latter species was already assessed as threatened and included on appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). As its southern population has been split into a different species, Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 73 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle 86 P, maackii AY962573 Korea 67 100 P, maackii MKO73654 100 98 —-P. maackii MKO73655 98 FP. maackii MK073636 P. maackii MK0O73637 P.. sinensis MK073617 China: Dongting P. sinensis MTDT10674 Vietnam: Ma River P. sinensis MTDT11160 China: Hong Kong P. sinensis MTDT6031 China: Guangdong, Shaoguan, Ruyuan P. sinensis S2 Japan: Yoshisa Town, Shizouka Prefecture P. sinensis TNE4030 US: Oahu, Kailua, Maunawili Stream P. sinensis TNE4042 US: Oahu, Kailua, Maunawili Stream 96 P. sinensis PsT29 Laos 98 P. sinensis PsT30 Laos P. sinensis PsT31 Vietnam: Ke Go Lake P.. sinensis PsT32 Vietnam: Nghe An Prov., Que Phong P. sinensis PsT33 Vietnam: Ke Go Lake P.. sinensis PsT34 Vietnam: Ke Go Lake P.. sinensis PsT35 Vietnam: Ha Tinh Prov., Gam Xuyen, Rac River P. sinensis PsT36 Vietnam: Ke Go Lake P. sinensis PsT37 Vietnam: Ke Go Lake P.. sinensis PsT61 Vietnam: Krong Bong, Dak Lak P. sinensis MTDT16907 China: Guangdong, Wuhua, Meizhou 68 P. sinensis MTDT16908 China: Guangdong, Wuhua, Meizhou bq —?. sinensis MTDT6033 China: Guangdong, Heyuan, Zijin P. sinensis MTDT6097 China: Taiwan, Taiwan P. sinensis MTDT6113 Japan: Okinawa Prefecture, Okinawajima 89 P. sinensis S3 Japan: Kumamoto City, Kumamoto Prefecture 90] FP. sinensis_MTDT6110 Japan: Okinawa Prefecture, Okinawajima P. sinensis Ps11 Vietnam: Quang Nam Prov., Nong Son P. sinensis MTDT12675 Europe: Slovenia, Primorje, Fiesa iS P.. sinensis PsT38 Vietnam: Quang Tri Prov. 93Lp sinensis S1 Japan: Kyoto City, Kyoto Prefecture Pelodiscus sp.1 MTDT4197China: Anhui, Jiangxian 93F Pelodiscus sp.1 MTDT14912 China: Jiangxi, Gangkoucun 100P 7 Pelodiscus sp.1 MTDT14913 China: Jiangxi, Gangkoucun gg] 99] ~ Pelodiscus sp.1 MTDT16871 China: Gangkou,Fengxin 5 Pelodiscus sp.1 MTDT16879 China: Gangkou, Fengxin 100 Pelodiscus sp.1 MTDT16841 China: Zhejiang, Deging ran Pelodiscus sp.1 MTDT16860 China: Zhejiang, Kaihua 93 Pelodiscus sp.1 MTDT6991 China: Guangdong, Lianzhou P. variegatus MTDT15734 Vietnam: Da River P. variegatus MTDT15735 Vietnam: Da River 100 P. variegatus MTDD42534 Vietnam: Khanh Hoa 99 P. variegatus }MTDD44045 Vietnam: Quang Binh. P. variegatus Ps12 Vietnam: QN P, variegatus Ps13 Vietnam: Nghe An 50 P. variegatus Ps18 Vietnam:Nghe An P. variegatus Ps14 Vietnam: Nghe An P. variegatus PsT1 Vietnam: Nghe An Prov. P. variegatus PsT10 Vietnam: Ha Tinh Prov. P. variegatus PsT11 Vietnam: Ha Tinh Prov. P. variegatus PsT12 Vietnam: Nghe An Prov. P, variegatus PsT13 Vietnam: Ha Tinh Prov. P, variegatus PsT14 Vietnam: Ha Tinh Prov. P. variegatus PsT15 Vietnam: Ha Tinh Prov. P. variegatus PsT16 Vietnam: Nghe An Prov. P. variegatus PsT17 Vietnam: Ha Tinh Prov. P. variegatus PsT18 Vietnam: Thanh Hoa Prov. 63 P. variegatus PsT5 Vietnam: Ha Tinh Prov. 61 P. variegatus PsT19 Vietnam: Quang Binh Prov. P. variegatus PsT2 Vietnam: Ha Tinh Prov. P. variegatus PsT20 Vietnam: Ha Tinh Prov. . variegatus PsT21 Vietnam: Nghe An Prov. . variegatus PsT22 Vietnam: Nghe An Prov. . variegatus PsT23 Vietnam: Ha Tinh Prov. . varlegatus PsT24 Vietnam: Ha Tinh Prov. . varlegatus PsT25 Vietnam: Nghe An Prov. . variegatus PsT26 Vietnam: Ha Tinh Prov. . variegatus PsT27 Vietnam: Ha Tinh Prov. . variegatus PsT28 Vietnam: Nghe An Prov. P. variegatus PsT3 Vietnam: Nghe An Prov. oe 80 P. variegatus PsT39 Vietnam: Nghe An Prov. 76 P. variegatus PsT4 Vietnam: Nghe An Prov. P. variegatus PsT40 Vietnam: Ha Tinh Prov. P. variegatus PsT41 Vietnam: Ha Tinh Prov. P. variegatus PsT42 Vietnam: Nghe An Prov. P. variegatus PsT43 Vietnam: Ha Tinh Prov. P. variegatus PsT45 Vietnam: Ha Tinh Prov. P. variegatus PsT47 Vietnam: Ha Tinh Prov. P. variegatus PsT49 Vietnam: Ha Tinh Prov. . variegatus PsT6 Vietnam: Ha Tinh Prov. . variegatus PsT7 Vietnam: Ha Tinh Prov. . Variegatus PsT8 Vietnam: Ha Tinh Prov. . variegatus PsT9 Vietnam: Ha Tinh Prov. . variegatus Pv22 Vietnam: Hanoi, Dong Mo Lake . Variegatus Pv25 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv26 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv28 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv29 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv30 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv31 Vietnam: Hanoi, Dong Mo Lake . variegatus Pv32 Vietnam: Hanoi, Dong Mo Lake P. variegatus RSDM3 Vietnam: Hanoi, Dong Mo Lake 100; Pelodiscus sp.2 MTDT16919 China:Guangdong, Huihoa 99 Pelodiscus sp.2 ZMB38 China: Tiger River P. sinensis MKO73618 China: Dongting P. parviformis MTDT16884 China: Hunan, Yongzhou P. parviformis MTDT 16885 China: Hunan, Yongzhou 99 P. parviformis MTDT16897 China: Guangdong, Luka farm 93 P. parviformis MTDT16899 China: Guangdong, Luka farm 100 P. parviformis MTDT16904 China: Guangdong, Luka farm 700 P. parviformis MTDT16905 China: Guangdong, Luka farm P. parviformis MTDT16920 China: Guangdong, Huihoa 00 P. parviformis MTDT16858 China: Zhejiang, Kaihua 95"P. parviformis MTDT16891 China: Guangxi, Longsheng 100 P. axenaria 004 100 P. axenaria 007 100 98 100 100 DUD DD 100 100 VUVVVVVVVUVUVVN DV 0.01 Figure 1. Trimmed phylogram, based on the Bayesian analysis. Number above and below branches of major nodes are Bayesian posterior probabilities and ML ultrafast bootstrap values, respectively. Sample highlighted in bold and orange is the paratype of Pelodiscus variegatus. P variegatus, the overall population of each species becomes even smaller than previously thought. Rhodin et al. (2018) regarded P. parviformis as “Critically En- dangered (CR)”. Consequently, TTWG (2021) provisionally categorised P. variega- tus as CR, although the assessment has not been officially accepted by the IUCN Red List. The species was recently proposed to list as Vulnerable (VU) in the Viet- Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 74 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Climate suitability 1 0 Records @ Pelodiscus sinensis O Pelodiscus variegatus Climate & landcover suitability 1 0 Records @ Pelodiscus sinensis © Pelodiscus variegatus Landcover suitability 1 0 Records @ Pelodiscus sinensis © Pelodiscus vanegatus Suitable protected area max min M™) none / no priority Records ® Pelodiscus sinensis © Pelodiscus vanegatus Figure 2. Potential distribution of Pelodiscus variegatus in Vietnam based on A climate B land cover C climate and land cover and D coverage with protected areas as number of suitable grid cells. nam Red Data Book, based on estimation of population reduction approximately of over 30% in the past of 30 years (Nguyen, per. comm. 2023). Ziegler et al. (2020) found during their market and trade surveys in Viet- nam that human overexploitation of softshell turtles appears massive. In addition, interbreeding events were observed to occur amongst softshell tur- tle species in farms. Most of the inhabited freshwater bodies and their sur- roundings showed signs of human encroachment, such as fishing, vegetation transformation, conversion and pollution. Thus, both in situ and ex situ conser- vation measures seem essential for protecting P variegatus from extinction. Some of the investigated freshwater habitats are already located inside pro- tected areas, such as Ben En, Phong Nha-Ke Bang and Vu Quang National Parks and Ke Go Nature Reserve in central Vietnam. Some of the protected areas are well known internationally as either special bird areas (Ke Go Nature Reserve) or with spectacular mammals, including Saola (Pseudoryx nghetinhensis) and the Large-antlered Muntjac (Muntiacus vuquangensis) (Vu Quang National Park). Improving conservation in and around the protected areas will benefit a suite of critically-endangered and endemic species, including the P variegatus. Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 7 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Table 2. Variables used for climatic niche modeling computation. NDVI = Normalized Difference Vegetation Index, EVI = Enhanced Vegetation Index. Abbreviation Remote sensing variable Bioclimatic variable Derived variable Source bio_12 N/A Annual Precipitation N/A Worldclim 2.1 bio_13 N/A Precipitation of Wettest Month N/A Worldclim 2.1 bio_14 N/A Precipitation of Driest Month N/A Worldclim 2.1 bio_15 N/A Precipitation Seasonality (Coefficient of N/A Worldclim 2.1 Variation) bio_16 N/A Precipitation of Wettest Quarter N/A Worldclim 2.1 bio_17 N/A Precipitation of Driest Quarter N/A Worldclim 2.1 bio_18 N/A Precipitation of Warmest Quarter N/A Worldclim 2.1 bio_19 N/A Precipitation of Coldest Quarter N/A Worldclim 2.1 ED1514_bio1 MODIS V4 Band 14 Synoptic Months: NDVI BIO1 = Annual Mean Temperature Annual Mean of NDVI EDENext ED1514_bio2 MODIS V4 Band 14 Synoptic Months: NDVI BIO2 = Mean Diurnal Range (Mean of Mean Diurnal Range of EDENext monthly (max temp - min temp)) NDVI ED1514_bio3 | MODIS V4 Band 14 Synoptic Months: NDVI BIO3 = lsothermality (BIO2/BIO7) (x100) | Isothermaility (BIO2/BI07) EDENext (*100) of NDVI ED1514_bio4 MODIS V4 Band 14 Synoptic Months: NDVI | BIlO4 = Temperature Seasonality (standard Seasonality of NDVI EDENext deviation x100) ED1514_bio5 MODIS V4 Band 14 Synoptic Months: NDVI | BIO5 = Max Temperature of Warmest Month Max NDVI of Monthly EDENext Scores ED1514_bio6 MODIS V4 Band 14 Synoptic Months: NDVI | BIO6 = Min Temperature of Coldest Month Min NDVI of Monthly EDENext Scores ED1514_bio7 MODIS V4 Band 14 Synoptic Months: NDVI | BIO7 = Temperature Annual Range (BIO5- Annual Range of NDVI EDENext BIO6) ED1514_bio10 | MODIS V4 Band 14 Synoptic Months: NDVI BIO10 = Mean Temperature of Warmest Mean NDVI of Warmest EDENext Quarter Quarter ED1514_bio11 | MODIS V4 Band 14 Synoptic Months: NDVI BIO11 = Mean Temperature of Coldest Mean NDVI of Coldest EDENext Quarter Quarter ED1515_bio1 MODIS V4 Band 15 Synoptic Months: EVI BIO1 = Annual Mean Temperature Annual Mean of EVI EDENext ED1515_bio2 MODIS V4 Band 15 Synoptic Months: EVI BIO2 = Mean Diurnal Range (Mean of Mean Diurnal Range of EDENext monthly (max temp - min temp)) EVI ED1515_bio3 MODIS V4 Band 15 Synoptic Months: EVI BIO3 = Isothermality (BIO2/BI07) (x100) | lsothermaility (BIO2/BIO7) EDENext (*100) of EVI ED1515_bio4 MODIS V4 Band 15 Synoptic Months: EVI BIO4 = Temperature Seasonality (standard Seasonality of EVI EDENext deviation x100) ED1515_bio5 MODIS V4 Band 15 Synoptic Months: EVI | BIO5 = Max Temperature of Warmest Month Max EVI of Monthly EDENext Scores ED1515_bio6 MODIS V4 Band 15 Synoptic Months: EVI BIO6 = Min Temperature of Coldest Month | Min EVI of Monthly Scores EDENext ED1515_bio7 MODIS V4 Band 15 Synoptic Months: EVI BIO7 = Temperature Annual Range (BIO5- Annual Range of EVI EDENext BIO6) ED1515_bio10 | MODIS V4 Band 15 Synoptic Months: EVI BIO10 = Mean Temperature of Warmest Mean EVI of Warmest EDENext Quarter Quarter ED1515_bio11 | MODIS V4 Band 15 Synoptic Months: EVI BIO11 = Mean Temperature of Coldest Mean EVI of Coldest EDENext Quarter Quarter ED15078_bio1 MODIS V4 Band 07+08 Synoptic Months: BIO1 = Annual Mean Temperature Annual Mean Temperature EDENext day- & nighttime land surface temperature ED15078_bio2 | MODIS V4 Band 07+08 Synoptic Months: BIO2 = Mean Diurnal Range (Mean of Mean Diurnal Range of EDENext day- & nighttime land surface temperature monthly (max temp - min temp)) Temperature ED15078_bio3 | MODIS V4 Band 07+08 Synoptic Months: BIO3 = Isothermality (BIO2/BI07) (x100) lsothermality (Bio2/Bio7) EDENext day- & nighttime land surface temperature (*100) ED15078_bio4 | MODIS V4 Band 07+08 Synoptic Months: BIO4 = Temperature Seasonality (standard Seasonality EDENext day- & nighttime land surface temperature deviation x100) ED15078_bio5 | MODIS V4 Band 07+08 Synoptic Months: | BIOS = Max Temperature of Warmest Month Max Temperature of EDENext day- & nighttime land surface temperature Warmest Month Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 76 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Abbreviation ED15078_bio6 Remote sensing variable MODIS V4 Band 07+08 Synoptic Months: day- & nighttime land surface temperature ED15078_bio7 MODIS V4 Band 07+08 Synoptic Months: day- & nighttime land surface temperature ED15078_bio10 MODIS V4 Band 07+08 Synoptic Months: day- & nighttime land surface temperature ED15078_bio11 MODIS V4 Band 07+08 Synoptic Months: day- & nighttime land surface temperature Bioclimatic variable BlIO6 = Min Temperature of Coldest Month BIO7 = Temperature Annual Range (BIO5- BIO6) BIO10 = Mean Temperature of Warmest Quarter BlO11 = Mean Temperature of Coldest Quarter Derived variable Min Temperature of Coldest Month Temperature Annual Range Mean Temperature of Warmest Quarter Mean Temperature of Coldest Quarter Source EDENext EDENext EDENext EDENext Table 3. Protected areas in Vietnam with projected proper climatic conditions and land-cover (Average Probability > 0.1) for Pelodiscus variegatus, sorted by size of suitable area. A full list of protected areas in China, Laos and Vietnam is presented in the Suppl. material 1. Name Phong Nha-Ke Bang Vu Quang Song Thanh Ke Go Ben En Phong Dien Bach Ma Nui Coc Hoa Lu Bac Huong Hoa Deo Ca-Hon Nua Than Sa-Phuong Hoang Nui Thanh Dakrong Dao Ho Song Da Bac Me Hue Sao La Hon Ba Ngoc Linh (Quang Nam) Quy Hoa-Ghenh Rang Ba Na-Nui Chua Nui Chung Bai Tu Long Son Tra Type National Park National Park Nature Reserve Nature Reserve National Park Nature Reserve National Park Cultural and Historical Site Cultural and Historical Site Nature Reserve Cultural and Historical Site Nature Reserve Cultural and Historical Site Nature Reserve Cultural and Historical Site Nature Reserve Nature Reserve Nature Reserve Nature Reserve Cultural and Historical Site Nature Reserve Cultural and Historical Site National Park Nature Reserve IUCN category II II Not Reported Not Reported Not Reported V Not Reported V Not Reported V Not Reported Not Reported Not Reported Not Reported Not Reported IV Not Reported Not Reported Not Reported Area [km?] 1222.825 591.661 890.589 239.724 141.816 407.762 375.506 90.263 65.819 235.057 215.611 136.517 64.139 387.232 78.889 87.109 377.864 198.213 192.496 52.102 268.922 2.031 64.906 38.368 Average Probability Suitable Area [km?] 0.303 424.578 0.275 168.536 0.110 117.079 0.364 91.145 0.123 85.311 0.417 85.060 0.389 79.732 0.200 60.395 0.126 59.431 0.162 56.869 0.122 56.472 0.112 35.724 0.134 34.971 0.351 32.474 0.130 31.908 0.122 24.266 0.422 17.208 0.146 14.560 0.146 10.710 0.115 7.707 0.226 6.019 0.295 2.020 0.214 1.401 0.198 1.213 As a first measure, based on the genetically-identified individuals, a conser- vation breeding programme has been established. This is following IUCN’s One Plan Approach to Conservation, developed by the Conservation Planning Spe- cialist Group (CPSG), which combines in situ with ex situ conservation mea- sures for the optimum protection of a given species (Byers et al. 2013). For the build-up of the conservation breeding programme, the individuals identified as P. variegatus were transferred to the Melinh Station for Biodiversity of the Insti- tute of Ecology and Biological Resources (IEBR), Hanoi. In the Station, located in Vinh Phuc Province, northern Vietnam, besides existing outdoor tank facil- ities, an exclusive softshell turtle breeding facility was constructed recently. Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 7/ Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle To maximise positive outcomes and for security reasons, viz., to extend the conservation breeding network, another group of the genetically-identified P variegatus was provided to another softshell breeding facility in northern Vietnam. Successful breeding has already been observed in the colony and off- spring are ready for release to the original habitat sites. To extend the conser- vation breeding programme and, thus, contribute to the build-up of a stable as- surance colony and conservation breeding network, a plan has been developed to transfer a limited number of surplus offspring to other facilities in Vietnam and overseas. In addition to these ex situ conservation measures already being in place, focus should now be directed to improving in situ conservation of this beautiful, but threatened softshell turtle species. Note added in proof In the late 2023, 50 young and healthy spotted softshell turtles from the in-country breeding program initiated by the Institute of Ecology and Biological Resources (IEBR), Vietnam, together with the Cologne Zoo, Germany, were suc- cessfully released to a site in northern Vietnam. Acknowledgements We thank Jonathan Fong, Luca Luiselli and two anonymous reviewers for their insightful comments on the earlier version of the paper and John B. Iverson and Anders G. J. Rhodin for providing valuable distribution data. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statement No ethical statement was reported. Funding This study was partially supported by the Ministry of Education and Training of Viet- nam (Project Code B2020-TDV-07) and the US Fish and Wildlife Service. Previous ge- netic screening was funded by Cologne Zoo and the European Union of Aquarium Cu- rators (EUAC). Cologne Zoo is partner of the World Association of Zoos and Aquariums (WAZA): Conservation Projects 07011, 07012 (Herpetodiversity Research, Amphibian and Reptilian Breeding and Rescue Stations). Author contributions TZ, MDL, TTN, and TQN conceptualized the study; CTR AVO, TEMM, and TTN conducted the fieldwork; DR, HTN, MDL, MHL, and TTN led the data analysis, data curation; TZ, MDL, DR led the writing and all authors edited and approved the manuscript. Author ORCIDs Minh Duc Le © https://orcid.org/0000-0002-2953-2815 Dennis Rédder © https://orcid.org/0000-0002-6108-1639 Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 78 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Tao Thien Nguyen © https://orcid.org/0000-0002-5640-4536 Cuong The Pham @® hitps://orcid.org/0000-0001-5158-4526 Truong Quang Nguyen © https://orcid.org/0000-0002-6601-0880 An Vinh Ong ® https://orcid.org/0000-0003-3683-3832 Timothy E. M. McCormack ® hittps://orcid.org/0009-0003-7565-2345 Thang Tai Nguyen © https://orcid.org/0009-0002-9088-3515 Mai Huyen Le © https://orcid.org/0009-0002-1536-1274 Hanh Thi Ngo ® https://orcid.org/0000-0002-5283-6243 Thomas Ziegler © https://orcid.org/0000-0002-4797-609X Data availability All of the data that support the findings of this study are available in the main text or Supplementary Information. References Arévalo E, Davis SK, Sites JW (1994) Mitochondrial DNA sequence divergence and phylo- genetic relationships among eight chromosome races of the Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Systematic Biology 43(3): 387-418. https://doi.org/10.1093/sysbio0/43.3.387 Asian Turtle Trade Working Group (2000) Pelodiscus sinensis (errata version published in 2016). The IUCN Red List of Threatened Species 2000: e.T39620A97401140. https://doi.org/10.2305/IUCN.UK.2000.RLTS.T39620A10251914.en Bain RH, Hurley MM (2011) A biogeographic synthesis of the amphibians and reptiles of Indochina. Bulletin of the American Museum of Natural History 360: 1-138. https:// doi.org/10.1206/360.1 Brun P, Thuiller W, Chauvier Y, Pellissier L, Wiest RO, Wang Z, Zimmermann NE (2020) Model complexity affects species distribution projections under climate change. Journal of Biogeography 47(1): 130-142. https://doi.org/10.1111/jbi.13734 Byers O, Lees C, Wilcken J, Schwitzer C (2013) The One Plan approach: The philosophy and implementation of CBSG’s approach to integrated species conservation plan- ning. WAZA Magazine 14: 2-5. Chkhikvadze VM (1987) On systematic position of the USSR Far East soft-shelled turtle. Bulletin of the Academy of Sciences of the Georgian Soviet Socialist Republic 128: 609-611. [In Russian with Georgian and English summary] Darriba D, Taboada GL, Doallo R, Posada D (2012) jmodelTest 2: More models, new heuristics and high-performance computing. Nature Methods 9(8): e772. https://doi. org/10.1038/nmeth.2109 Ducotterd C, Le Duc O, Pham TV, Leprince B, Bordes C, Ly TN, Ho PT, Le AT, Tran BQ, Luu VQ, Luiselli L (2023) Previously unrecorded invasive species and the unsatisfying knowledge of turtle communities in northern Vietnam. Conservation 3: 1-13. https:// doi.org/10.3390/conservation3010001 Engstrom TN, Shaffer HB, Mccord WP (2002) Phylogenetic diversity of endangered and critically endangered southeast Asian softshell turtles (Trionychidae: Chi- tra). Biological Conservation 104(2): 173-179. https://doi.org/10.1016/SO006- 3207(01)00161-6 Engstrom TN, Shaffer HB, Mccord WP (2004) Multiple data sets, high homoplasy, and the phylogeny of softshell turtles (Testudines: Trionychidae). Systematic Biology 53(5): 693-710. https://doi.org/10.1080/10635150490503053 Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 79 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Farkas B, Ziegler T, Pham CT, Ong AV, Fritz U (2019) A new species of Pelodiscus from northeastern Indochina (Testudines: Trionychidae). ZooKeys 824: 71-86. https://doi. org/10.3897/zookeys.824.31376 Fick SE, Hijmans RJ (2017) WorldClim 2: New 1-km spatial resolution climate surfac- es for global land areas. International Journal of Climatology 37(12): 4302-4315. https://doi.org/10.1002/joc.5086 Fritz U, Gong S, Auer M, Kuchling G, Schneeweiss N, Hundsdorfer AK (2010) The world’s economically most important chelonians represent a diverse species complex (Testudines: Trionychidae: Pelodiscus). Organisms, Diversity & Evolution 10(3): 227- 242. https://doi.org/10.1007/s13127-010-0007-1 Ginal P. Tan WC, Rédder D (2022) Invasive risk assessment and expansion of the re- alized niche of the Calotes versicolor species complex (Daudin, 1802). Frontiers of Biogeography 14(3): e54299. https://doi.org/10.21425/F5FBG54299 Gong S, Vamberger M, Auer M, Praschag P Fritz U (2018) Millennium-old farm breeding of Chinese softshell turtles (Pelodiscus spp.) results in massive erosion of biodiversi- ty. Naturwissenschaften 105(5-6): 34. https://doi.org/10.1007/s00114-018-1558-9 Gong Y-A, Peng L-F, Huang S, Lin Y-F, Huang R-Y, Xu Y-H, Yang D-C, Nie L-W (2021) A new species of the Genus Pelodiscus Fitzinger, 1835 (Testudines: Trionychidae) from Huangshan, Anhui, China. Zootaxa 5060(1): 137-145. https://doi.org/10.11646/zoo- taxa.5060.1.7 Gong §S, Fritz U, Vamberger M, Gao Y, Farkas B (2022) Disentangling the Pelodiscus ax- enaria complex, with the description of a new Chinese species and neotype designa- tion for P axenaria (Zhou, Zhang & Fang, 1991). Zootaxa 5125(2): 131-143. https:// doi.org/10.11646/zootaxa.5125.2.2 GRDC (2020) WMO Basins and Sub-Basins. from Global Runoff Data Centre. https://gdk. gdi-de.org/geonetwork/srv/eng/catalog.search#/metadata/fcd3c1 9f-3332-447e- 8e03-c428c0cd1693 Gumbricht T, Roman-Cuesta RM, Verchot L, Herold M, Wittmann F, Householder E, Herold N, Murdiyarso D (2017) An expert system model for mapping tropical wetlands and peatlands reveals South America as the largest contributor. Global Change Biology 23(9): 3581-3599. https://doi.org/10.1111/gcb.13689 Hernandez PA, Graham CH, Master LL, Albert DL (2006) The effect of sample size and spe- cies characteristics on performance of different species distribution modeling meth- ods. Ecography 29(5): 773-785. https://doi.org/10.1111/j.0906-7590.2006.04700.x Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution inter- polated climate surfaces for global land areas. International Journal of Climatology 25(15): 1965-1978. https://doi.org/10.1002/joc.1276 IUCN (2023) The IUCN Red List of Threatened Species. Version 2022-2. https://www. iucnredlist.org Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Coope A, Markowitz S, Duran C, Thierer T, Ashton B, Mentjies P Drummond A (2012) Geneious Basic: An integrated and extendable desktop software platform for the organiza- tion and analysis of sequence data. Bioinformatics 28(12): 1647-1649. https://doi. org/10.1093/bioinformatics/bts199 Le M, Raxworthy CJ, McCord WP, Mertz L (2006) A molecular phylogeny of tortois- es (Testudines: Testudinidae) based on mitochondrial and nuclear genes. Mo- lecular Phylogenetics and Evolution 40(2): 517-531. https://doi.org/10.1016/j. ympev.2006.03.003 Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 80 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Le M, Duong HT, Dinh LD, Nguyen TQ, Pritchard PCH, McCormack T (2014) A phylogeny of softshell turtles (Testudines: Trionychidae) with reference to the taxonomic status of the critically endangered, giant softshell turtle, Rafetus swinhoei. Organisms, Diver- sity & Evolution 14(3): 279-293. https://doi.org/10.1007/s13127-014-0169-3 Le Duc O, Pham TV, Leprince B, Bordes C, Luu VQ, Lo OV, Nguyen ATT, Luong LTK, Pham SN, Luiselli L (2020) Farming characteristics and the ecology of Palea steindachneri (Trionychidae) in Vietnam. Russian Journal of Herpetology 27(2): 97-108. https:// doi.org/10.30906/1026-2296-2020-27-2-97-108 Mu Q, Heinsch FA, Zhao M, Running SW (2007) Development of a global evapotranspi- ration algorithm based on MODIS and global meteorology data. Remote Sensing of Environment 28(4): 519-536. https://doi.org/10.1016/j.rse.2007.04.015 Nguyen L-T, Schmidt HA, von Haeseler A, Bui MQ (2015) IQ-TREE: A fast and effective stochastic algorithm for estimating maximum likelihood phylogenies. Molecular Biol- ogy and Evolution 32(1): 268-274. https://doi.org/10.1093/molbev/msu300 Palumbi SR (1996) Nucleic acids II: The polymerase chain reaction. In Hillis DM, et al. (Eds) Molecular systematic (2™ edn.). Sinauer Associates, Sunderland, MA, 205-247. Pham TV, Le Duc O, Leprince B, Bordes C, Ducotterd C, Luu VQ, Le AT, Nguyen HM, Lo OV, Ha VN, Tran QB, Luiselli L (2023a) An assessment of turtle communities in Bach Ma National Park, Vietnam. Nature Conservation Research 8: 72-80. https://doi. org/10.24189/ncr.2023.016 Pham TV, Le Duc O, Leprince B, Bordes C, Ducotterd C, Luu VQ, Ravor S, Tran BQ, Luiselli L (2023b) A preliminary assessment of chelonian diversity in the montane forests of two areas in northern Vietnam. The Herpetological Bulletin 164(2023): 7-12. https:// doi.org/10.33256/hb1 64.712 Pham TV, Le Duc O, Leprince B, Bodres C, Zuklin T, Ducotterd C, Luu VQ, Lo OV, Nguyen AT, Fa JE, Luiselli L (2020) Unexpected high forest turtle diversity in hill forests in north- ern Vietnam. Biodiversity and Conservation 29: 4019-4033. https://doi.org/10.1007/ $10531-020-02061-y Phillips SJ, Anderson RP, Schapire RR (2006) Maximum entropy modeling of species geographic distributions. Ecological Modelling 190(3-4): 231-259. https://doi. org/10.1016/j.ecolmodel.2005.03.026 Phillips SJ, Dudik M, Schapire RE (2017) Maxent software for modeling species nich- es and distributions (Version 3.4.1). http://biodiversityinformatics.amnh.org/open_ source/maxent/ Rhodin AGJ, Stanford CB, van Dijk PP, Eisemberg C, Luiselli L, Mittermeier RA, Hudson R, Horne BD, Goode EV, Kuchling G, Walde A, Baard EHW, Berry KH, Bertolero A, Blanck TEG, Bour R, Buhimann KA, Cayot LJ, Collett S, Currylow A, Das I, Diagne T, Ennen JR, Forero-Medina G, Frankel MG, Fritz U, Garcia G, Gibbons JW, Gibbons PM, Gong S, Guntoro J, Hofmeyr MD, Iverson JB, Kiester AR, Lau M, Lawson DP Lovich JE, Moll E, Paez VP Palomo-Ramos R, Platt K, Platt SG, Pritchard PCH, Quinn HR, Rahman SC, Randrianjafizanaka ST, Schaffer J, Selman W, Shaffer HB, Sharma DSK, Shi H, Singh S, Spencer R, Stannard K, Sutcliffe S, Thomson S, Vogt RC (2018) Global conservation status of turtles and tortoises (order Testudines). Chelonian Conservation and Biolo- gy 17(2): 135-161. https://doi.org/10.2744/CCB-1348.1 Ronquist F, Teslenko M, van der Mark P Ayres DL, Darling A, Hohna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61(3): 539-542. https://doi.org/10.1093/sysbio/sys029 Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 81 Minh Duc Le et al.: Conservation priorities for the Spotted Softshell Turtle Shi H-T, Parham JF, Zhiyong F, Meiling H, Feng Y (2008) Evidence of the massive scale of turtle farming in China. Oryx 42(1): 147-150. https://doi.org/10.1017/ $0030605308000562 Stuckas H, Fritz U (2011) Identity of Pelodiscus sinensis revealed by DNA sequences of an approximately 180-year-old type specimen and a taxonomic reappraisal of Pelodiscus species (Testudines: Trionychidae). Journal of Zoological Systemat- ics and Evolutionary Research 49(4): 335-339. https://doi.org/10.1111/j.1439- 0469.2011.00632.x Swets JA (1988) Measuring the accuracy of diagnostic systems. Science 240(4857): 1285-1293. https://doi.org/10.1126/science.3287615 Tang Y (1997) Research on a new species of Pelodiscus, Trionychidae in China. Zoolog- ical Research 18: 13-17 TTWG [Turtle Taxonomy Working Group] (2021) Turtles of the world. Annotated check- list and atlas of taxonomy, synonymy, distribution, and conservation status (9* edn.). Chelonian Research Monographs 8: 1-472. https://doi.org/10.3854/crm.8.checklist. atlas.v9.2021 Yang P Tang Y, Ding L, Guo X, Wang Y (2011) Validity of Pelodiscus parviformis (Testu- dines: Trionychidae) inferred from molecular and morphological analysis. Asian Her- petological Research 2(1): 21-29. https://doi.org/10.3724/SP.J.1245.2011.00021 Zhou G, Zhang X, Fang Z (1991) Bulletin of a new species of Trionyx. Acta Scientiarum Naturalium Universitatis Normalis Hunanensis 14: 379-382. [In Chinese with English abstract] Ziegler T, Nguyen TT, Ong AV, Pham CT, Nguyen TQ (2020) In search of the Spotted Soft- shell Turtle in Vietnam: An implementation of the One Plan Approach. WAZA News 2427. Supplementary material 1 Supplementary data Authors: Minh Duc Le, Dennis Rodder, Tao Thien Nguyen, Cuong The Pham, Truong Quang Nguyen, An Vinh Ong, Timothy E. M. Mccormack, Thang Tai Nguyen, Mai Huyen Le, Hanh Thi Ngo, Thomas Ziegler Data type: docx Explanation note: table $1. Samples used in this study. table S2. Relative importance of protected areas in China, Laos and Vietnam for Pelodiscus variegatus in terms of climatic conditions and land-cover. The Table is sorted according to the suitable area in each Reserve. figure $1. Full phylogram based on the Bayesian analysis. Numbers above and below branches of selected nodes are Bayesian posterior probabilities and ML ultrafast bootstrap values, respectively. figure S2. Summary of the receiver operating characteristic curve of 100 Maxent models for P variegatus trained with climatic variables. figure S3. Summary of the receiver operating characteristic curve of 100 Maxent models for P. variegatus trained with land-cover variables. Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/natureconservation.55.114746.suppl1 Nature Conservation 55: 67-82 (2024), DOI: 10.3897/natureconservation.55.114746 82