ArticlePDF Available

Abstract and Figures

For wildlife to be protected, it must first be known, as species that are not yet formally described are not the target of conservation attention, independently of threat levels. While habitat degradation has consistently increased across the last decades in the Republic of Korea, taxonomic and conservation efforts are lagging. For instance, a clade of Onychodactylus clawed salamanders from the extreme southeast of the Korean Peninsula is known to have diverged circa 6.82 million years ago from its sister species O. koreanus, and despite the candidate species status of this lineage, its extremely restricted range is under intense anthropogenic pressure. Here, through the use of genetics, morphometrics and landscape modelling, we confirm the species status of the southeast Korean Onychodactylus population and formally describe it as Onychodactylus sillanus sp. nov. We then proceed to determine threats, habitat loss and risk of extinction based on climatic models under different Representative Concentration Pathways and following the IUCN Red List categories and criteria. We highlight a decrease in the extent of occurrence between 87.6% and 97.3 % within the next three generations based on several climate change scenarios, a decline high enough for the species to be listed as Critically Endangered based on the category A3 of the IUCN Red List of species. Our results will enable the development of protection programs and legitimise citizen activities protecting the population. A conservation action plan is a priority to coordinate the activities linked to the protection of the species.
Range, suitable habitat, and phylogenetic relationships of Onychodactylus sillanus sp. nov. A: Georeferenced occurrence points of Onychodactylus koreanus and Onychodactylus sillanus sp. nov. Collected occurrence points are based on our survey data as well as the Global Biodiversity Information Facility (GBIF), VertNet, EcoBank platform, and voucher specimens deposited in the Ewha Woman’s University Natural History Museum (EWNHM). Genetic samples and specimens originate from a selection of sites, and all datapoints were used to generate ecological niche models after spatial distance thinning. The same general data collection methodology and resulting occurrence dataset for O. koreanus are provided in Shin et al. (2021). B: Estimated area of Onychodactylus sillanus sp. nov. presence based on Minimum Convex Polygon (MCP) and binary presence area of Maxent output. Note: Maxent binary output (792.7 km2) estimated significantly broader area of presence than MCP (258.3 km2). Maxent output largely fell within higher elevation area of South Gyeongsang Province. Map generated in QGIS v3.10. *: Maxent ranges are binary encoded for presence and absence based on the threshold determined in the Supplementary Materials. C: Maximum-likelihood consensus tree of the genus Onychodactylus derived from analysis of 1 153 bp cyt b, 654 bp COI, and 532 bp 16S rRNA gene fragments. Corresponding voucher specimen information, locality data, and GenBank accession numbers are given in Supplementary Table S3. Black circles indicate strongly supported nodes, empty circles indicate nodes with moderate support levels, no circles correspond to no node support. Numbers at tree nodes correspond to ML ultrafast-bootstrap (UFBS)/BI posterior probability (PP) support values. D, E: Onychodactylus sillanus sp. nov. in life: adult male from North Gyeongsang Province, Yangsan-si, Republic of Korea (D); larva from South Gyeongsang Province, Yangsan-si, Dong-myeon, Republic of Korea (E). F: Natural habitat of species in North Gyeongsang Province, Yangsan-si, Dong-myeon, Republic of Korea. Photos by Amaël Borzée (D) and Nikolay A. Poyarkov (E, F).
… 
Content may be subject to copyright.
Letter to the editor Open Access
Dwindlinginthemountains:Descriptionofacritically
endangeredandmicroendemicOnychodactylus
species(Amphibia,Hynobiidae)fromtheKorean
Peninsula
DEAR EDITOR,
Species that are not formally described are generally not
targetsforconservation,regardlessoftheirthreatenedstatus.
Whilehabitatdegradationhasincreasedoverthepastseveral
decades in the Republic of Korea, taxonomic and
conservation efforts continue to lag. For instance, a clade of
Onychodactylus clawed salamanders from the extreme
southeasterntipof the Korean Peninsula, whichdivergedca.
6.82millionyearsagofrom its sister species O. koreanus, is
under intense anthropogenic pressure due to its extremely
restricted range, despite its candidate species status. Here,
using genetics, morphometrics, and landscape modeling, we
confirmed the species status of the southeast Korean
Onychodactylus population, and formally described it as
Onychodactylus sillanussp. nov.Wealsodeterminedthreats,
habitat loss, and risk of extinction based on climatic models
under different Representative Concentration Pathways
(RCPs) and following the IUCN Red List categories and
criteria. Based on several climate change scenarios, we
estimated a decline in suitable habitat between 87.6% and
97.3% within the next three generations, sufficient to be
consideredCriticallyEndangeredaccordingto CategoryA3of
the IUCN Red List of Threatened Species. These findings
shouldhelpenablethedevelopmentofconservationprograms
and civic activities to protect the population. Conservation
actionplans are apriorityto coordinatetheactivities required
toprotectthisspecies.
Current anthropogenic activities continue to have a
tremendousimpactonglobalbiodiversity,leadingtotheworld’
s sixth mass extinction event (Ceballos et al., 2015).
Consequently,biodiversityisreachingatippingpointforfuture
self-sustainment (Rinawati et al., 2013), and conservation
efforts are critical for the continuity of natural evolutionary
patterns and the avoidance of anthropomorphic selection
(Otto, 2018). However, conservation first requires a formal
description of the species. This is especially true for highly
threatened amphibians (www.iucnredlist.org), where habitat
loss is a leading driver of species decline (Lindenmayer &
Fischer, 2013). Species protection relies on explicit and
recognized taxonomy, much of which is still under
consideration in East Asia (Kim et al., 2016; see
Supplementary Materials for examples and links between
taxonomyandconservation).
Despite recent taxonomic developments in amphibians
inhabitingtheRepublicofKorea(hereafterR.Korea),diversity
within the clawed salamander genus Onychodactylus
(Caudata,Hynobiidae)remainspoorlyunderstood(detailson
genus taxonomy are provided in the Supplementary
Materials).Basedongeneticdata,twopotentiallyundescribed
Onychodactylus species exist in R. Korea (Poyarkov et al.,
2012; Suk et al., 2018), one from southern Gyeongsang
Province (Poyarkov et al., 2012; Suk et al., 2018) and one
from Gangwon Province (Suk et al., 2018). The southern
Gyeongsang population is increasingly threatened by habitat
loss, threats shared with that of Hynobius yangi due to
overlapping habitats (Borzée & Min, 2021), and resulting in
legal challenges aiming at their protection
(https://www.nocutnews.co.kr/news/34985). With the
continuing decline in the genus Onychodactylus (Maslova et
Received:06 April 2022; Accepted: 26 July 2022; Online: 02 August
2022
Foundationitems:This workwas supportedbythe NationalResearch
FoundationofKorea (NRF) Grant fundedby the Korean Government
(NRF-2009-0067686, NRF-2016R1D1A1B03934071) to M.S.M., the
ForeignYouth TalentProgram (QN2021014013L)fromthe Ministryof
ScienceandTechnology toA.B.,andpartiallyby theRussianScience
Foundation(22-14-00037)toN.A.P.
This is an open-access article distributed under the terms of the
Creative Commons Attribution Non-Commercial License (http://
creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted
non-commercial use, distribution, and reproduction in any medium,
providedtheoriginalworkisproperlycited.
Copyright ©2022 Editorial Office of Zoological Research, Kunming
InstituteofZoology,ChineseAcademyofSciences
Received:06 April 2022; Accepted: 26 July 2022; Online: 02 August
2022
Foundationitems:This workwas supportedbythe NationalResearch
FoundationofKorea (NRF) Grant fundedby the Korean Government
(NRF-2009-0067686, NRF-2016R1D1A1B03934071) to M.S.M., the
ForeignYouth TalentProgram (QN2021014013L)fromthe Ministryof
ScienceandTechnology toA.B.,andpartiallyby theRussianScience
Foundation(22-14-00037)toN.A.P.
This is an open-access article distributed under the terms of the
Creative Commons Attribution Non-Commercial License (http://
creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted
non-commercial use, distribution, and reproduction in any medium,
providedtheoriginalworkisproperlycited.
Copyright ©2022 Editorial Office of Zoological Research, Kunming
InstituteofZoology,ChineseAcademyofSciences
Borzée et al. Zool. Res. 2022, 43(5): 750−755
https://doi.org/10.24272/j.issn.2095-8137.2022.048
al., 2018; Suk et al., 2018; see Supplementary Materials for
additionalinformation),theaimofthisstudywastodetermine
the taxonomic and threat status of Onychodactylus
populations in Yangsan and Miryang in southeastern R.
Korea.
We assessed the taxonomic status of the species using
morphological(SupplementaryTablesS1,S2)andmolecular
analyses (Supplementary Table S3). We then determined
niche differentiation for the focal clade and O. koreanus
through ecological niche modeling (ENM; Figure 1). Finally,
we modeled the impact of climate change and conducted
conservation assessment for the focal clade. All details on
study materials and methods are presented in the
SupplementaryMaterials.
Morphometricanalysesdemonstratedsignificantdifferences
between O. koreanus and the candidate Onychodactylus
species from Yangsan and Miryang. Analysis of variance
(ANOVA)usedtotestforvariationsbetween cladesbasedon
four principal components (PCs) was significant for PC3
(χ2=7.52; F(47,48)=8.74, P=0.005; Supplementary Table S2),
although only two variables significantly loaded into PC3:
length of front and back limbs. The larvae of the candidate
species had significantly longer front and back limbs than O.
koreanus(SupplementaryTable S4andFigureS1).However,
these variables overlapped between the two clades
(Supplementary Table S4). In addition, visual comparison of
the vomerine teeth indicated that the candidate species was
moresimilarto O. zhaoermii intoothshape,but O. koreanus
intoothnumber,althoughwenotethesmallsamplesizeused
for comparison (see Supplementary Figure S2). The
morphological differences between the two species may be
more generalized for adults as morphological divergences in
amphibiansaregenerally betterunderstoodforadultsthanfor
larvae.
For phylogenetic analysis, we obtained a 532 bp long
fragmentof16SrRNA,1 153bplongfragmentofcytb,and
654 bp long fragment of COI mitochondrial DNA (mtDNA)
genes.Thefinal concatenated alignment of mtDNAdatawas
2 339 bp,andincluded 1 728 conservativesites,601 variable
sites, and 397 parsimony informative sites within the ingroup
(Figure 1). Phylogenetic analyses employing Bayesian
inference (BI) and maximum-likelihood (ML) approaches
yieldednearly identical topologies,whichonly differedslightly
inassociationsatseveralpoorlysupportednodes(Figure1).
The topology of the matrilineal genealogies within
Onychodactylus was sufficiently resolved and largely
consistent with earlier phylogeny of the genus presented by
Poyarkovetal.(2012)andYoshikawa& Matsui(2013,2014).
Details on the phylogenetic relationships between non-focal
cladesofOnychodactylusarepresentedintheSupplementary
Materials.
WithintheO. koreanusspeciesgroupinhabitingtheKorean
Peninsula, three clades with unresolved topological
relationshipswererevealed(Figure1):(1)O. koreanussensu
stricto clade, which included populations from the southwest,
north,andnortheastofR.Korea(seeFigure1);(2)population
fromGangwonProvinceintheextremenortheastofR.Korea,
herein referred to as Onychodactylus sp. B; and (3)
populations from the extreme southeast of the Korean
Peninsula, corresponding to the candidate Onychodactylus
species from Yangsan and Miryang. Uncorrected genetic P-
distancescalculatedfromthecytbgenefragmentsupported
the species level of all three clades within the group
(Supplementary Table S5; details in Supplementary
Materials). The split between the candidate clade and O.
koreanus likely occurred due to continued tectonic activity
followingtheformationoftheYangsanfault(Suketal., 2018),
alowlandareathatdoesnotfittheecologicalrequirementsof
eitherspecies(seeSupplementaryMaterialsfordiscussionon
cladeorigin).
Ouranalyses indicatedcleardifferences inpotentialhabitat
suitability between the candidate species and O. koreanus.
Based on evaluation metrics, the (ENMs) for O. koreanus
(AUC=0.886±0.02; TSS=0.626±0.03) and the candidate
species (AUC=0.819±0.07; TSS=0.549±0.18) showed good
predictiveability.The lower AUC forthecandidate species is
likelylinked tothe lowersample size.Theadditionaljackknife
testappliedtothecandidate speciesalso indicatedsignificant
predictive ability of the ENM, despite the small sample size
(P=0.03). The contributions of environmental variables in the
ENMs differed between O. koreanus and the candidate
species, with slope showing the highest contribution for O.
koreanusandmean diurnal range (bio2) showingthehighest
contribution for the candidate species (Supplementary Table
S6;detailsinSupplementaryMaterials).Inaddition,theENMs
for both species were congruent with known distribution
(Supplementary Figure S3; details in Supplementary
Materials).Ofnote, while the ENM for O. koreanuspredicted
high habitat suitability within the range of the candidate
species, without occurring in the area, the ENM for the
candidate species provided only weak predictions within the
O. koreanusrange(SupplementaryFigureS3).
Niche identity analysis indicated that while the ENMs
generated for O. koreanus and the candidate species were
somewhat similar (Schoener’s D=0.30; I=0.59), they were
significantlydifferent(P<0.05). Range break analysis showed
no significant differences in the ENMs between O. koreanus
andthecandidatespecies(PD=0.3663;PI=0.3663). However,
comparison between O. koreanus and the “ribbon clade” on
theYangsanFault showed that theENMsofthe two species
differed significantly based on the I statistic (PD=0.059;
PI=0.039), while the ENMs of the candidate species and
“ribbon clade” differed significantly based on the D statistic
(PD=0.039; PI=0.079). Therefore, the presence of unsuitable
habitatbetween theranges ofO. koreanusandthecandidate
species likely acts as a barrier to dispersal, resulting in the
isolation of both clades (see Supplementary Materials for
additionaldiscussionon“ribbonclade”).
Taxonomic account
Following concordant lines of evidence based on molecular
and morphometric analyses of species-level divergence, we
hypothesize that the Yangsan-Miryang population of
Onychodactylus represents a discretely diagnosable lineage.
Its species status should be urgently recognized, as formally
described below. The species description presented herein
relies on individuals already in collections as the species is
threatened and sacrificing additional individuals for formal
Zoological Research 43(5): 750−755, 2022 751
752 www.zoores.ac.cn
description contravenes the conservation practices
recommendedinthiswork.
Onychodactylus sillanus
sp. nov. Min, Borzée, &
Poyarkov
SupplementaryFiguresS2,S4–S6andTablesS4,S7
Holotype: CGRB15897 (field ID MMS6682) deposited in the
Conservation Genome Resources Bank for Korean Wildlife
(CGRB), subadult (metamorph, likely female) specimen
collectedbyMi-Sook Min and ChangHoonLee on 13 March
2014 from a montane stream at Sasong-ri, Dong-myeon,
Yangsan, Republic of Korea (N35.309659°, E129.067086°;
110ma.s.l.)(SupplementaryFigureS5).
Paratypes:Three specimens,includingCGRB15898 (fieldID
MMS6688); juvenile (larval) specimen collected by Mi-Sook
Minand ChangHoon Leeon 24March 2014fromamontane
stream at Sasong-ri, Dong-myeon, Yangsan, Republic of
Korea (N35.306060°, E129.074238°; 100 m a.s.l.;
SupplementaryFigureS2);CGRB15906(field IDMMS-7207),
subadult (metamorph, likely male) specimen collected by Mi-
SookMin andHaejunBaek on15May 2015froma montane
stream at Sasong-ri, Dong-myeon, Yangsan, Republic of
Korea (N5.31275°, E129.06892°; 110 m a.s.l.; CGRB 15907
(field ID MMS 7270 / NAP-05131), subadult (likely male)
specimencollectedbyMi-Sook Min and Nikolay A. Poyarkov
on 5 May 2015 from a montane stream at Unmun Mountain,
Unmun-myeon, Unmunsan-gil, Cheongdo-gun, North
Gyeongsang Province, Republic of Korea (N35.643892°,
E129.018449°;730ma.s.l.;SupplementaryFigureS6).
Diagnosis: A slender, medium-sized hynobiid salamander
and member of the genus Onychodactylus based on a
combination of the following morphological features: lungs
absent; black claw-like keratinous structures present on both
fore- and hindlimbs in larvae and breeding adults; tail longer
than sum of head and body lengths, tail in adults almost
cylindricalatbase, slightly compressedlaterallyatdistal end;
vomerineteethin transverse row of shortarch-shapedseries
in contact with each other; larvae with skinfolds on posterior
edgesofbothfore-andhindlimbs;breedingmaleswithdermal
flapsonposterioredgesofhindlimbs;other typicalfeaturesof
genus. The new species can be diagnosed from other
members of the genus by the following combination of adult
characters: 11–12 costal grooves; vomerine teeth in two
comparativelyshallow, slightlycurved serieswith18–22teeth
ineach,incontactmedially;outerbranches ofvomerinetooth
series slightly longer than inner branches and outer ends of
series located more posteriorly than anterior ends; dark
ground dorsum, head and tail (slate-black to brown) with
numerous,medium-sized(size<SVL/20)yellowishto reddish-
orange confluent elongated spots and ocelli, ventral side
purplish gray; light dorsal band always absent; juveniles with
darkventraltrunk,largeyellowishblotchesondorsumandtail.
Measurements and counts of holotype: SVL: 54.8; TL:
51.8;GA: 26.5;IC: 6.5;CW:6.3; CGBL:1; HLL:15;HL:12.0
HW:8.5;EL:3.4;IN:5.0;ON:2.7;IO:3.2;IC:6.5;OR:5.2;1-
FL:1.6;2-FL:2.2;3-FL:3.3;4-FL:2.0;1-TL:1.6;2-TL:2.1;3-
TL: 3.1; 4-TL: 3.0; 5-TL: 2.7; VTL: 1.9; VTW: 4.5; MTH: 4.7;
MTW:4.0;MAXTH:7.1;CGN(L):11;CGN(R):11;VTN(L):21;
VTN(R): 22. Please refer to the Supplementary Materials for
measurements and counts of type series, morphological
measurements of referred specimens, detailed description of
holotypecolorationand morphology,andvariationandnatural
historyofnewspecies.
Etymology: The specific name sillanus” is a toponymic
adjective in the nominative singular, masculine gender,
referringtothehistoricalKoreankingdomofSilla(57BC–935
AC) located on the southeastern parts of the Korean
Peninsula, coinciding with the geographic distribution of the
newspecies.
Common names: We suggest “Yangsan Clawed
Salamander”asthecommon name in English and “Yangsan
GgorichireDorongnyong”(KkorichireDorongnyong)(양산꼬리
치레도롱뇽)asthecommon name in Korean, in reference to
itsdistribution.
Comparisons:WithintheO. koreanusgroup,Onychodactylus
sillanussp. nov. is most closely related to O. zhaoermii,O.
koreanus,andundescribedclade Onychodactylus sp. B from
GangwonProvinceinR.Korea.Itformsamonophyleticgroup
with the latter two species and comparisons of the new
species with O. zhaoermii and O. koreanus are the most
pertinent.ThenewspeciesdiffersfromO. koreanus bymodal
costal grooves 11 (vs. higher CGN count and modal costal
grooves13 (12–13)forboth sexesinO. koreanus);vomerine
Figure 1 Range, suitable habitat, and phylogenetic relationships of
Onychodactylus sillanus
sp. nov.
A:Georeferenced occurrencepointsofOnychodactylus koreanus andOnychodactylus sillanussp. nov.Collectedoccurrencepointsare basedon
oursurveydata aswellas theGlobalBiodiversity InformationFacility(GBIF),VertNet,EcoBankplatform,andvoucherspecimensdepositedinthe
Ewha Woman’s University Natural History Museum (EWNHM). Genetic samples and specimens originate from a selection of sites, and all
datapoints were used to generate ecological niche models after spatial distance thinning. The same general data collection methodology and
resultingoccurrencedataset for O. koreanusare provided inShinet al. (2021). B:Estimated area of Onychodactylussillanussp. nov. presence
based on Minimum Convex Polygon (MCP) and binary presence area of Maxent output. Note: Maxent binary output (792.7 km2) estimated
significantlybroaderareaofpresencethanMCP(258.3km2).MaxentoutputlargelyfellwithinhigherelevationareaofSouthGyeongsangProvince.
Map generated in QGIS v3.10. *: Maxent ranges are binary encoded for presence and absence based on the threshold determined in the
SupplementaryMaterials.C:Maximum-likelihoodconsensustreeofthegenusOnychodactylusderivedfromanalysisof1 153bpcytb,654bpCOI,
and532bp 16SrRNAgene fragments.Correspondingvoucher specimeninformation,locality data,andGenBank accessionnumbersare givenin
Supplementary Table S3. Black circles indicate strongly supported nodes, empty circles indicate nodes with moderate support levels, no circles
correspondtononodesupport.NumbersattreenodescorrespondtoMLultrafast-bootstrap(UFBS)/BIposteriorprobability(PP)supportvalues.D,
E: Onychodactylus sillanus sp. nov. in life: adult male from North Gyeongsang Province, Yangsan-si, Republic of Korea (D); larva from South
GyeongsangProvince,Yangsan-si,Dong-myeon,Republic ofKorea(E).F:Natural habitatofspeciesinNorth GyeongsangProvince,Yangsan-si,
Dong-myeon,RepublicofKorea.PhotosbyAmaëlBorzée(D)andNikolayA.Poyarkov(E,F).
Zoological Research 43(5): 750−755, 2022 753
teeth 18–22 (mean 20) in each series (vs. VTN count 10–18
(mean 16.4±2.21) in O. koreanus); inner branch of vomerine
tooth arch notably curved forward at medial end (vs. inner
branchof vomerinetootharch notcurved atmedialend inO.
koreanus); light markings on dorsum forming confluent
blotchesandspots(vs. light markings on dorsum usually not
confluent and not forming reticulate pattern in O. koreanus);
and front and back limbs in larvae FLL/SVL 0.09–0.13
(1.11±0.01 mean±SD) and HLL/SVL 0.10–0.16 (0.12±0.01)
(vs. FLL/SVL 0.07–0.12 (0.09±0.01) and HLL/SVL 0.07–0.13
(0.11±0.01)inO. koreanus).
Onychodactylus sillanussp. nov. further differs from O.
zhaoermiiby vomerine teeth 18–22(mean20) in eachseries
(vs. vomerine teeth usually less than 16 per series (mean
13.8±1.3) in O. zhaoermii); and light markings on dorsum
formingconfluentblotchesandspots,withlightocelli inadults
(see Figure 1) (vs. light markings on dorsum forming dense
reticulated pattern, with light ocelli absent in O. zhaoermii).
Onychodactylussillanussp. nov.differsfromothercongeners
based on a combination of the following morphological
attributes:average number ofcostalgrooves 11; slightlybent
vomerinetoothseries,withhighernumberofvomerineteethin
each series (mean 20); typical coloration of juveniles and
adults (Supplementary Figure S7), dark-brown ground color
withnumeroussmall,roundandbrightgoldentoorangespots,
andnodistinctdorsalband.
Current distribution and climate change: Chang et al.
(2009) provided the first record of the new species from
UnmunMountainas O. fischeri.Modeling placedtheextent of
occurrence (EOO) of Onychodactylussillanussp. nov. at
792.7 km2 based on the Maxent binary map and 258.3 km2
based on the minimum convex polygon (MCP). This small
rangeispredictedtoshrinkfurtherasallbinarymapsfromthe
forecasted ENMs showed a drastic reduction in suitable
habitat for Onychodactylussillanussp. nov. (Supplementary
Figure S8). The models for 2050 predicted 32.1 km2 of
suitablehabitatundertheRCP4.5 scenario and 21.6 km2 of
suitable habitat under the RCP 8.5 scenario (Figure 1).
ModelsbasedonlatertimeframesandhigherRCPspredicted
even greater decreases in suitable habitat (Supplementary
Materials).
Conservation assessment: According to the international
Protected Planet database (https://www.protectedplanet.net/),
suitablehabitatforOnychodactylussillanussp. nov.overlaps
with several protected areas of significance. The species is
alsopresentorexpectedtobepresent infiveprotectedareas
in R. Korea (see list in Supplementary Materials). The
projected population size reductions based on EOO declines
andclimate changepredictions for2050 wellmatchtheIUCN
RedListCategory A3(c). TheRCP4.5 scenario resulted ina
decline of 87.6% (MCP) and 95.9% (Maxent binary) and the
RCP 8.5 scenario resulted in a decline of 91.6% (MCP) and
97.3% (Maxent binary). Under such scenarios, the species
would reach the status of Critically Endangered under
CategoryA3(c).BasedonEOO,thecurrentgeographicrange
of Onychodactylussillanussp. nov. was estimated at
792.7 km2 (Maxent binary) and 258.3 km2 (MCP). Thus, we
recommend Onychodactylussillanussp. nov.beclassifiedas
CriticallyEndangeredunder Category A3(c)oftheIUCN Red
List. Our findings highlight the need for a coordinated
conservationactionplancoveringallareaswherethespecies
isknowntooccur.
In the current study, based on morphometric, genetic, and
ecologicalnichemodeling,wedescribeanewOnychodactylus
species (Onychodactylussillanussp. nov) restricted to the
southeasternmargin of theKorean Peninsula. Inaddition, we
link species description to threat assessments based on the
IUCN Red List of Threatened Species. We suggest the
species be listed as Critically Endangered, in view of
projections of future areas based on climate change
scenarios,orasEndangered,basedonthesizeofthecurrent
EOO. Without conservation action, the newly described
speciesisfacinganuncertainfuture.
NOMENCLATURAL ACTS REGISTRATION
The electronic version of this article in portable document
format represents a published work according to the
InternationalCommissiononZoologicalNomenclature(ICZN),
andhence thenew namescontained intheelectronicversion
are effectively published under that Code from the electronic
edition alone (see Articles 8.5–8.6 of the Code). This
published work and the nomenclatural acts it contains have
beenregisteredinZooBank,the onlineregistration systemfor
the ICZN. The ZooBank LSIDs (Life Science Identifiers) can
be resolved and the associated information can be viewed
throughany standard webbrowserby appendingtheLSID to
theprefixhttp://zoobank.org/.
PublicationLSID:
urn:lsid:zoobank.org:pub:66EEAC0C-2421-43B2-84D2-
BF11684EFEE5
Onychodactylus sillanusLSID:
urn:lsid:zoobank.org:act:C62AAA19-197A-4B6C-BA6C-
AD231FDA0874
SCIENTIFIC FIELD SURVEY PERMISSION INFORMATION
Allobservationsandexperimentsconductedinthisstudywere
performedinaccordancewiththeethicalrecommendationsof
the College of Biology and the Environment at Nanjing
ForestryUniversity.
SUPPLEMENTARY DATA
Supplementarydatatothisarticlecanbefoundonline.
COMPETING INTERESTS
Theauthorsdeclarethattheyhavenocompetinginterests.
AUTHORS’ CONTRIBUTIONS
A.B.andM.S.M.designedthestudy.A.B.,Y.S.,N.A.P.,J.Y.J.,
H.J.B., C.H.L., J.A., Y.J.H. and M.S.M. conducted field work
and data analysis. A.B., Y.S., N.A.P., and M.S.M. wrote the
manuscript.Allauthorsreadandapprovedthe finalversionof
themanuscript.
ACKNOWLEDGEMENTS
WethankTaeHoKim, DongYounKim,MinSeob Kim,andIl
754 www.zoores.ac.cn
Hoon Kim for assistance in the field. Nikolay A. Poyarkov
kindly thanks K. Sarkisian and J.H. Yoon for support and
assistance. The authors thank all anonymous reviewers for
theirusefulcommentsonanearlierdraftofthemanuscript.
AmaëlBorzée1,#,*,YucheolShin1,2,NikolayA.Poyarkov3,
JongYoonJeon4,HaeJunBaek4,ChangHoonLee4,
JunghwaAn5,YoonJeeHong5,Mi-SookMin4,#,*
1Laboratory of Animal Behaviour and Conservation, College of
Biology and the Environment, Nanjing Forestry University,
Nanjing, Jiangsu 210037, China
2Department of Biological Sciences, School of Natural Science,
Kangwon National University, Chuncheon 24341, Republic of
Korea
3Department of Vertebrate Zoology, Biological Faculty, M.V.
Lomonosov Moscow State University, Moscow 119234, Russia
4Research Institute for Veterinary Science, College of Veterinary
Medicine, Seoul National University, Seoul 08826, Republic of
Korea
5Animal Resources Division, National Institute for Biological
Resources, Incheon 22689, Republic of Korea
#Authors contributed equally to this work
*Correspondingauthors,E-mail:amaelborzee@gmail.com;
minbio@yahoo.co.kr
REFERENCES
BorzéeA,MinMS.2021.Disentanglingtheimpactsof speciation,sympatry
and the island effect on the morphology of seven Hynobius sp.
salamanders.Animals,11(1):187.
Ceballos G, Ehrlich PR, Barnosky AD, García A, Pringle RM, Palmer TM.
2015. Accelerated modern human–induced species losses: entering the
sixthmassextinction.Science Advances,1(5):e1400253.
Chang MH, Kim BS, Park SG, Kim TW, Oh HS. 2009. Herpetofauna in
mountainunmun.Korean Journal of Herpetology,1:53−60.
Kim DY, Kim D, Park SK, Lee H, Suk HY, Min MS. 2016. Isolation and
characterization of novel microsatellite loci in the wrinkled frog. Genetics
and Molecular Research,15(1):gmr.15017183.
Lindenmayer DB, Fischer J. 2013. Habitat Fragmentation and Landscape
Change: An Ecological and Conservation Synthesis. Washington: Island
Press.
Maslova IV, Portnyagina EY, Sokolova DA, Vorobieva PA, Akulenko MV,
PortnyaginAS,etal. 2018.Distributionofrare andendangeredamphibians
and reptiles in Primorsky Krai (Far East, Russia). Nature Conservation
Research,3(S1):61−72.
OttoSP.2018. Adaptation, speciation andextinction in the Anthropocene.
Proceedings of the Royal Society B,285(1891):20182047.
PoyarkovJrJN,CheJ,MinMS,Kuro-O M,YanF,LiC,etal.2012.Review
of the systematics, morphology and distribution of Asian Clawed
Salamanders, genus Onychodactylus (Amphibia, Caudata: Hynobiidae),
withthedescriptionoffournewspecies.Zootaxa,3465(1):1−106.
Rinawati F, Stein K, Lindner A. 2013. Climate change impacts on
biodiversity—thesettingofalingeringglobalcrisis.Diversity,5(1):114−123.
ShinY,Min MS, Borzée A.2021. Driven to the edge:Species distribution
modelingofa ClawedSalamander(Hynobiidae: Onychodactylus koreanus)
predictsrange shifts and drasticdecrease of suitable habitatsin response
toclimatechange.Ecology and Evolution,11(21):14669−14688.
Suk HY, Lee MY, Bae HG, Lee SJ, Poyarkov Jr N, Lee H, et al. 2018.
Phylogenetic structure and ancestry of Korean clawed salamander,
Onychodactylus koreanus (Caudata: Hynobiidae). Mitochondrial DNA Part
A,29(4):650−658.
Yoshikawa N, Matsui M. 2013. A new salamander of the genus
Onychodactylus from Tsukuba Mountains, Eastern Honshu, Japan
(Amphibia,Caudata,Hynobiidae).Current Herpetology,32(1):9−25.
Yoshikawa N, Matsui M. 2014. Two new salamanders of the genus
Onychodactylus from eastern Honshu, Japan (Amphibia, Caudata,
Hynobiidae).Zootaxa,3866(1):53−78.
Zoological Research 43(5): 750−755, 2022 755
1
Supplementary Materials
Introduction
The principal threat to species as a result of human activities is the change in
landscapes and resulting habitat loss (Lindenmayer & Fischer, 2013). This is especially
important for species with narrow ranges, and therefore more susceptible to minor variations
and stochastic events (Manne & Pimm, 2001). As a result of human activities, between 900
and 130,000 species became extinct during the last five centuries (www.iucnredlist.org;
Régnier et al., 2015), in connection with human activities, and numerous undescribed species
also become extinct every year (Tedesco et al., 2014). These silent extinctions are likely to
have a worse impact on biodiversity than estimated as ecological contribution to the
environment are poorly known in undescribed species, and virtually unknown for
undescribed species (Brodie et al., 2014; e.g., Rix et al., 2017). Indeed, it is not possible to
care about what we do not know, and this relationship between taxonomy, knowledge and
conservation highlights a need for the urgent taxonomic assessment of all clades). However,
conservation effort do generally result in improvements in species conservation statuses, for
instance reducing bird extinction rates by 40% (Monroe et al., 2019) and preventing the
extinction of dozens of vertebrates in recent times (Bolam et al., 2021). For amphibians as
well, conservation activities can result in the decrease of threat, and for instance the Oaxaca
Treefrog (Sarcohyla celata) was transferred from Critically Endangered to Near Threatened
in 2021 following its protection and rigorous habitat management by local communities
(Neam & Borzée, 2021).
Clawed salamanders are a peculiar group of lungless salamanders endemic to
northeast Asia and a sister lineage to the rest of the family Hynobiidae (Zhang et al., 2006).
These salamanders are habitat specialists adapted to life and reproduction in flowing
mountain streams or subterranean environments, and their distribution is generally restricted
to montane areas of the Russian Far East, northeastern China, the Korean Peninsula and the
Japanese Archipelago (Honshu and Shikoku islands; Kuzmin, 1995; Poyarkov et al., 2012).
The genus Onychodactylus was traditionally considered to contain only two species, O.
japonicus (Houttuyn) from Japan and O. fischeri (Boulenger) from the northeastern Asian
mainland (Kuzmin, 1995; Sato, 1943). However, recent phylogenetic and integrative
taxonomic studies (Poyarkov et al., 2012; Yoshikawa et al., 2008, 2013; Yoshikawa &
Matsui, 2013, 2014, 2022) revealed a previously unknown diversity within Onychodactylus
and added eight new species to the genus, by describing six new species from Japan (O.
fuscus Yoshikawa & Matsui, O. intermedius Yoshikawa & Matsui, O. kinneburi Yoshikawa,
Matsui, Tanabe & Okayama, O. nipponoborealis Kuro-O, Poyarkov & Vieites, O.
pyrrhonotus Yoshikawa & Matsui, and O. tsukubaensis Yoshikawa & Matsui), two new
species from China (O. zhangyapingi Che, Poyarkov & Yan, and O. zhaoermii Che,
Poyarkov & Yan), and one new species from the Korean Peninsula (O. koreanus Min,
Poyarkov & Vieites). Therefore, there are currently four recognized nominal species of
Onychodactylus occurring in the mainland Asia: O. fischeri (Primorye, Russian Far East), O.
zhangyapingi (Jilin Province, China), O. zhaoermii (Jilin and Liaoning provinces, China and
also expected to occur in D.P.R. Korea; Borzée et al., 2021) and O. koreanus (Korean
Peninsula; Shin et al., 2021b).
The genus Onychodactylus is generally declining (Maslova et al., 2018; Suk et al., 2018), and
in R. Korea O. koreanus was listed as threatened in 1989 under the Protection of Wild Fauna
and Flora Act of the Korean Ministry of Environment (as O. fischeri). The species was
however excluded from the list in 1998 in view of its large population size and large
2
distributional range, which largely resulted from the identification of O. koreanus as O.
fischeri as the species was not described at that time.
Materials and Methods
Introduction about the species
Within the three Onychodactylus clades occurring in R. Korea, haplotypes are not
abundantly shared between geographically close populations (Suk et al., 2018), indicating
weak gene flow, and weak dispersal abilities of Onychodactylus, despite the known migration
patterns in fall (Shin et al., 2020b). Onychodactylus species are strict ecological specialists
(Kuzmin, 1995; Suk et al., 2018) and are generally found in comparatively higher elevation
and forested areas near cold mountain streams (Hong, 2017; Kim et al., 2019; Poyarkov et al.,
2012). Spawning in the candidate species from Yangsan has not been observed, likely due to
the subterranean breeding behaviour typical for Onychodactylus species, making it difficult
for eggs to be detected (Kuzmin, 1995; Park, 2005).
General sampling
Sampling was conducted in several phases. The first section was conducted in 2002
2015 and all voucher specimens, except one from Unmun Mountain (see Poyarkov et al.,
2012), were re-used in the present paper for the genetic analysis but not for the morphological
comparisons. The collected specimens were stored in 70 % alcohol until 2021, when they
were measured for the morphometric analysis. In addition, field surveys were conducted in
November and December 2020 for morphological analyses and to collect datapoints for
modeling. The surveys were in the form of road cruising in wet evenings in areas where the
focal clade was expected to be occurring, on 17, 19, 20 and 27 November, and on 7 and 8
December 2020. For each individual, GPS coordinates were recorded, some adult individuals
were measured and released for the taxonomic assessment, but no individual was collected.
We also acquired buccal swabs (cotton-tipped swab; 16H22, MedicalWire; Corsham, UK)
from three adult individuals, two from Yangsan (35.448333 °N; 128.97055 °E; 340 m a.s.l.)
and one from Miryang (35.466442 °N; 128.929128 °E; 280 m a.s.l. (Figure 1); to ascertain
the identity of the populations. Specimens collected earlier were subsequently deposited in in
the Conservation Genome Resources Bank for Korean Wildlife (CGRB), College of
Veterinary Medicine, Seoul National University, R. Korea.
Ethical approval
All applicable international, national, and/or institutional guidelines for the care and
use of animals were strictly followed. All animal sample collection protocols complied with
the current laws of the Republic of Korea and the People’s Republic of China. All
observations and experiments conducted in this study are in agreement with the ethical
recommendations of the College of Biology and the Environment at Nanjing Forestry
University. We did not collect adult individuals as voucher specimen, in line with the IUCN
recommendation on research involving species at risk of extinction (IUCN, 1989).
Morphometrics
All individuals included in the morphometric analysis were similarly preserved in
70% alcohol since sampling, and we therefore did not consider shrinking because of
dehydration as a problem (Shu et al., 2017). The field identification number of the larvae
measured were the following mms1351 to mms1358, mms2585; mms2590 to mm2609;
mms3211 to mms3212; mms3616 to mms3621, mms4339; mms5675 to mms5681; mms6683
to mms6684; mms6686 to mms6691. All larvae were estimated to be in their second year of
development based on size, and they all had reached the same developmental stage, matching
with the stage 28 in other caudata species (Hurney et al., 2015). The individual vouchers
3
mms2594, mms2599, mms2601 and mms2609 were not included in the analyses. However,
the data collected here cannot be compared with live individuals or specimens preserved and
stored in other preservatives. A total of 49 larvae were included in the morphometric analysis,
such as n= 26 for O. koreanus and n= 23 for the candidate species from Yangsan and
Miryang. The data were collected with digital callipers (1108–150, Insize; Suzhou, China) to
the nearest 0.1 mm, following the methodological recommendations for salamanders in
general (Nishikawa et al., 2007) and specifically to the genus (Poyarkov et al., 2012). Each
individual was measured three times by the same observer, all samples were measured by the
same observer over the course of a single day to ensure accuracy, and the results were
averaged for each individual for further analyses. The morphometric variables collected for
larval specimens were: SVL—snout-vent-length; TL—tail length; GA—dextral gleno–
acetabular distance (minimum distance between axilla and groin measured on the right side
of the straightened body); CW—chest width (minimum distance between left and right
axillae); FLL—dextral forelimb length (length of the straightened right forelimb measured
from axilla to tip of the longest finger of forelimb); HLL—dextral hindlimb length (length of
the straightened right hindlimb from groin to tip of the longest toe of hindlimb); HL—head
length (from tip of snout to the gular fold at mouth angle); HW—head width (as maximum
width of head); EL—dextral eye length (minimum distance from the anterior corner of the
right eye to the posterior corner of the right eye); IN—internarial distance; ON—dextral
orbitonarial distance (minimum distance between the right external nares and the anterior
corner of the eye on the same side of the head); IO—interorbital distance (minimum distance
between the upper eyelids) and IC—intercanthal distance (measured as minimum distance
between anterior corners of the eyes).
For the subsequent holotype and paratype descriptions, in line with the description of
Onychodactylus koreanus (Poyarkov et al., 2012), we also recorded OR—orbitorostral
distance; 1-FL–4-FL—first to fourth finger lengths; 1-TL–5-TL—first to fifth toe lengths;
VTL—vomerine tooth series depth; VTW—vomerine tooth series width; MTH—tail height
in the middle of the tail; MTW—tail width in the middle of the tail; MAXTH—maximal tail
height; and the following meristic characters CGN —costal grooves number both dextrally
and sinistrally (number of costal grooves between the forelimbs and hindlimbs, excluding
axillary and inguinal grooves, following Misawa, 1989); VTN—vomerine teeth number in
the left and right tooth series; and CGBL—number of costal between adpressed limbs. We
also measured these characters for two adults and seven larvae for the species description.
The individuals were placed in a 20 % benzocaine bath for 1 min to anesthetize them while
acquiring adequate morphological measurements. As we are not aware of any recommended
protocol, the individuals were not totally anesthetized, but unable to right themselves for 15
to 20 min and therefore slow enough for adequate morphological measurement. All
measurements were conducted on a wet towel at the point of capture, and individuals were
released on the spot once alert again.
4
Supplementary Table S1. Correlation table for all morphological variables for
Onychodactylus sillanus sp. nov. specimens. We conducted a Pearson’s correlation test for all
morphological variables included in the analyses (n= 49). Each variable was measured three
times and averaged, and data are presented in the form of a ratio with the focal variables
divided by SVL
GA
(right)
CW
FFL
(right)
HLL
(right)
HL
HW
EL
(right)
IN
IO
IC
TL
R
-0.577
-0.496
-0.413
0.057
-0.243
-0.424
-0.551
-0.152
-0.214
-0.458
p
< 0.001
< 0.001
0.003
0.695
0.093
0.002
< 0.001
0.298
0.14
0.001
GA (right)
R
-
0.591
0.268
0.021
0.066
0.52
0.34
0.084
0.155
0.178
p
-
< 0.001
0.062
0.886
0.65
< 0.001
0.017
0.567
0.288
0.222
CW
R
-
-
0.328
-0.022
0.209
0.726
0.261
0.014
0.198
0.132
p
-
-
0.021
0.88
0.149
< 0.001
0.071
0.924
0.172
0.367
FFL (right)
R
-
-
-
0.506
0.148
0.208
0.058
0.226
0.338
0.193
p
-
-
-
< 0.001
0.31
0.152
0.691
0.118
0.018
0.185
HLL (right)
R
-
-
-
-
0.112
-0.141
-0.272
0.172
0.094
-0.054
p
-
-
-
-
0.443
0.335
0.058
0.237
0.52
0.715
HL
R
-
-
-
-
-
0.281
0.22
0.427
0.354
0.573
p
-
-
-
-
-
0.051
0.128
0.002
0.012
< 0.001
HW
R
-
-
-
-
-
-
0.317
0.108
0.201
0.239
p
-
-
-
-
-
-
0.027
0.462
0.165
0.098
EL (right)
R
-
-
-
-
-
-
-
0.154
-0.003
0.304
p
-
-
-
-
-
-
-
0.29
0.983
0.034
IN
R
-
-
-
-
-
-
-
-
0.466
0.259
p
-
-
-
-
-
-
-
-
0.001
0.073
ON (right)
R
-
-
-
-
-
-
-
-
-0.025
0.037
p
-
-
-
-
-
-
-
-
0.866
0.802
IO
R
-
-
-
-
-
-
-
-
-
0.312
p
-
-
-
-
-
-
-
-
-
0.029
5
Following established protocols (Borzée & Min, 2021) and in line with the general
literature on morphological analyses (Hayek et al., 2001; Lleonart et al., 2000), we used a
ratio to correct for body size in relation with individual growth. To calculate the ratio, we
divided each variable by the SVL value of the same individual. To harness as much of the
variation possible expressed by the 12 morphological measurements collected for the 49
individuals, and because 20 measurements were correlated (Pearson correlation;
Supplementary Table S1), we opted for a factor reduction analysis. We selected a principal
component analysis (PCA) and tested the resulting principal components (PCs) for significant
difference between the two clades. Factor extraction for the PCA was conducted under a
varimax rotation with Kaiser normalisation. The rotations converged in six iterations and
variables were selected as loading into a PC if > 0.50, ensuring that each variable loaded into
one of the PCs (Supplementary Table S2). Once the PCs were extracted, we tested for
significant differences between the two clades using a one-way ANOVA with PCs as
dependent variables and clades as independent variables.
Supplementary Table S2. Component matrix with the loading factors of each variable into
the PCs. The factors were extracted with a principal component analysis under a varimax
rotation with Kaiser normalisation. Rotations converged in 6 iterations. The representative
PCs are in bold. The value used as the cut-off was 0.47, and loading variables are in bold.
The PCs were selected if the eigenvalues were >1, based on varimax rotations. The total
loading cumulative percentage is in bold. Each PC explained between 8.45 and 30.73% of
variation, and PC3 was significantly different between the two clades.
PC1
PC2
PC3
PC4
Rotated Component
Tail length
-0.64
-0.39
0.03
0.38
Gleno-acetabular
distance
0.79
0.05
0.08
-0.18
Chest width
0.88
0.01
0.11
-0.05
Forelimb length (right)
0.36
0.25
0.71
-0.06
Hindlimb length (right)
-0.12
0.04
0.87
-0.10
Head length
0.12
0.79
0.01
0.11
Head width
0.85
0.15
-0.09
0.15
Eye length (right)
0.37
0.36
-0.42
-0.50
Internarial distance
-0.01
0.70
0.23
0.19
Orbitonarial distance
(right)
-0.04
0.25
-0.19
0.79
Interorbital distance
0.16
0.57
0.35
0.09
Intercanthal distance
0.13
0.76
-0.13
-0.25
Variance
Eigenvalues
3.69
1.93
1.57
1.01
% of Variance
30.73
16.11
13.07
8.45
ANOVA
Mean square
0.25
0.39
7.52
0.06
df
47,48
47,48
47,48
47,48
F
0.25
0.39
8.74
0.06
p
0.621
0.536
0.005
0.816
6
In addition, we performed a visual comparison of the shape and number of vomerine
teeth between the candidate species from Yangsan and Miryang and the two most closely
phylogenetically related Onychodactylus species: O. koreanus and O. zhaoermii, based on the
graphical representation from Poyarkov et al. (2012). We did not conduct any statistical
analysis regarding the number of vomerine teeth due to the low sample size (n= 3) for the
candidate species.
Phylogenetics
The phylogenetic analysis was conducted using the samples already reported in
Poyarkov et al. (2012) and Suk et al. (2018) for the three mitochondrial DNA markers,
including cytochrome b(Cyt b), 1st subunit of cytochrome oxidase c(COI), and a partial
sequence of 16S rRNA fragment. We also added the four newly collected samples (two from
Yangsan, one from Unmun mountain, and one from Miryang), thus ascertaining the genetic
assignment of the population in Miryang, and the monophyly with the samples from Yangsan.
The DNA was extracted from the buccal swabs using the DNeasy Blood and Tissue kit
(Qiagen, Hilden, Germany) following the manufacturer’s protocol. Details on the PCR
reactions followed Poyarkov et al. (2012), with the same primers and PCR conditions for 16S
rRNA and COI as reported by Poyarkov et al. (2012); for amplification of Cyt bgene
fragment we used the primers Onycho_Cytb_up70F (forward: 5′-
CATAGCAAGTAAAGCTAAATAAATCAT-3′; Yoshikawa et al., 2008) and
salamander_Cytb_RN2 (reverse: 5′-YTYTCAATCTTKGGYTTACAAGACC-3′; Matsui et
al., 2008). Quality check and trimming of sequences were conducted using Geneious Prime
2019.1.3 (https://www.genei ous.com).
Nucleotide sequences were initially aligned in MAFFT v.6 (Katoh et al., 2002) with
default parameters, and then checked by eye and slightly adjusted in BioEdit v.7.0.5.2 (Hall,
1999). Genetic distances and sequence statistics were calculated using MEGA-X (Kumar et
al., 2018) and DnaSP v. 5 (Librado & Rozas, 2009). We deposited the resulting sequences in
GenBank (accession numbers MZ404936; MZ404937 and MZ404938 Supplementary Table
3). We added GenBank data for all other ten currently recognized Onychodactylus species, a
candidate species from Gangwon, Korea (Onychodactylus sp. A), as well as seven species of
hynobiid salamanders (Hynobius chinensis,H. arisanensis,Pachyhynobius shangchengensis,
Ranodon sibiricus,Batrachuperus yenyuanensis,Paradactylodon persicus and
Salamandrella keyserlingii; Supplementary Table 3) which were used as outgroups
(following the phylogenetic results of Chen et al., 2015; Weisrock et al., 2013; Zheng et al.,
2011, 2012). Altogether, the final alignment included sequences from 79 hynobiid
representatives, including six specimens of O. zhaoermii, 40 specimens of O. koreanus, one
specimen of Onychodactylus sp. A from Gangwon, and 16 specimens of the candidate
Onychodactylus species from Yangsan and Miryang (Supplementary Table 3).
We concatenated the sequence data of 16S rRNA, COI and Cyt bfor a final
alignment of 2357 bp. The best fit substitution models were determined with PartitionFinder
v. 2.1.1 (Lanfear et al., 2017), using the Akaike Information Criterion and the Bayesian
Information Criterion for the Maximum Likelihood (ML) and Bayesian Inference (BI)
analyses, respectively. The 16S rRNA gene fragment was treated as a single partition,
whereas Cyt band COI were partitioned by codon position. For ML and BI analyses, the best
partitioning scheme and models of nucleotide substitutions were: 16S rRNA (GTR+I+G), 1st
position of Cyt b(TrN), 2nd position of Cyt b(HKY+I), 3rd position of Cyt b(HKY), 1st
position of COI (SYM+I), 2nd position of COI (F81+I), and 3rd position of COI (HKY+G).
7
Supplementary Table S3. Sequences and voucher specimens of Onychodactylus and outgroup hynobiid taxa used in this study.
No.
Sample ID
Species
Region
Locality
Cyt b
16S rRNA
COI
1
JQ710885
Hynobius chinensis
China
Hubei
JQ710885
JQ710885
JQ710885
2
EF462213
Hynobius arisanensis
Taiwan, Alishan
EF462213
EF462213
EF462213
3
MK890366
Pachyhynobius shangchengensis
China
Henan
MK890366
MK890366
MK890366
4
AJ419960
Ranodon sibiricus
China
Xinjiang
AJ419960
AJ419960
AJ419960
5
DQ333818
Batrachuperus yenyuanensis
China
Sichuan
DQ333818
DQ333818
DQ333818
6
MK737945
Paradactylodon persicus
Iran
Gorgan
MK737945
MK737945
MK737945
7
JX508762
Salamandrella keyserlingii
China
Heilongjiang
JX508762
JX508762
JX508762
8
NP-OF2-1
Onychodactylus fischeri
Russia
Primorskiy Krai, Tigrovoy
JX157971
JX158086
JX158208
9
KJ627181
Onychodactylus zhangyapingi
China
Jilin, Linjiang
KJ627181
KJ627181
KJ627181
10
MKO-2-1
Onychodactylus nipponoborealis
Japan
Aomori, Hirosaki
JX158074
JX158197
JX158320
11
NPH-109-1
Onychodactylus japonicus
Japan
Kanagawa, Hakone
JX158050
JX158171
JX158293
12
KI-17
Onychodactylus kinneburi
Japan
Ehime, Ishizuchi-san
JX158283
JX158284
JX158285
13
OJ112
Onychodactylus pyrrhonotus
Japan
Kyoto, Nantan-shi
AB452922
14
OJ50
Onychodactylus tsukubaensis
Japan
Ibaraki, Tsukuba-san
AB452860
15
KUHE48284
Onychodactylus fuscus
Japan
Fukushima, Tadami
AB923989
16
KUHE47455
Onychodactylus intermedius
Japan
Fukushima, Koriyama
AB924003
17
KJ627182
Onychodactylus zhaoermii
China
Changbai-Shan
KJ627182
KJ627182
KJ627182
18
KIZ-YPX10504
Onychodactylus zhaoermii
China
Changbai-Shan
JX158235
JX157998
JX158116
19
KIZ-YPX10505
Onychodactylus zhaoermii
China
Liaoning, Qinchengzi
JX158236
JX157999
JX158117
20
KIZ-YPX10506
Onychodactylus zhaoermii
China
Liaoning, Qinchengzi
JX158237
JX158000
JX158118
21
CBS-2010-18
Onychodactylus zhaoermii
China
Liaoning, Kuandian
JX158234
JX157997
JX158114
22
CBS-2010-19
Onychodactylus zhaoermii
China
Liaoning, Kuandian
JX158115
23
PDS A-1
Onychodactylus koreanus
Korea
North Chungcheong, Naeam-myeon
JX158261
JX158024
JX158143
24
PDS A-2
Onychodactylus koreanus
Korea
North Chungcheong, Naeam-myeon
JX158262
JX158025
JX158142
25
PDS A-3
Onychodactylus koreanus
Korea
North Chungcheong, Naeam-myeon
JX158263
JX158026
JX158144
26
PDS A-4
Onychodactylus koreanus
Korea
North Chungcheong, Naeam-myeon
JX158264
JX158027
JX158141
27
PDS A-5
Onychodactylus koreanus
Korea
North Chungcheong, Naeam-myeon
JX158265
JX158028
JX158145
8
28
PDS B-1
Onychodactylus koreanus
Korea
Gangwon, Nae-myeon
JX158247
JX158010
JX158132
29
PDS B-2
Onychodactylus koreanus
Korea
Gangwon, Nae-myeon
JX158248
JX158011
JX158131
30
PDS B-3
Onychodactylus koreanus
Korea
Gangwon, Nae-myeon
JX158249
JX158012
JX158130
31
PDS B-4
Onychodactylus koreanus
Korea
Gangwon, Nae-myeon
JX158250
JX158013
JX158129
32
PDS B-5
Onychodactylus koreanus
Korea
Gangwon, Nae-myeon
JX158251
JX158014
JX158128
33
PDS C-1
Onychodactylus koreanus
Korea
Gangwon, Seo-myeon
JX158238
JX158001
JX158120
34
PDS C-2
Onychodactylus koreanus
Korea
Gangwon, Seo-myeon
JX158239
JX158002
JX158119
35
PDS C-3
Onychodactylus koreanus
Korea
Gangwon, Seo-myeon
JX158240
JX158003
JX158121
36
PDS C-4
Onychodactylus koreanus
Korea
Gangwon, Seo-myeon
JX158241
JX158004
JX158122
37
PDS D-1
Onychodactylus koreanus
Korea
Gangwon, Dongnae-myeon
JX158242
JX158005
JX158123
38
PDS D-2
Onychodactylus koreanus
Korea
Gangwon, Dongnae-myeon
JX158243
JX158006
JX158126
39
PDS D-3
Onychodactylus koreanus
Korea
Gangwon, Dongnae-myeon
JX158244
JX158007
JX158124
40
PDS D-4
Onychodactylus koreanus
Korea
Gangwon, Dongnae-myeon
JX158245
JX158008
JX158125
41
IK20
Onychodactylus koreanus
Korea
North Jeolla, Naejangsan NP
JX158272
JX158033
JX158155
42
H059
Onychodactylus koreanus
Korea
North Jeolla, Naejangsan NP
JX158273
JX158034
JX158156
43
H095
Onychodactylus koreanus
Korea
Gangwon, Chiaksan NP, Gillaji
JX158252
JX158015
JX158133
44
H096
Onychodactylus koreanus
Korea
Gangwon, Chiaksan NP, Gillaji
JX158253
JX158016
45
MMS-258
Onychodactylus koreanus
Korea
Gangwon, Singi-myeon
JX158259
JX158022
JX158139
46
MMS-259
Onychodactylus koreanus
Korea
Gangwon, Singi-myeon
JX158260
JX158023
JX158140
47
MMS-141
Onychodactylus koreanus
Korea
Gangwon, Chiaksan NP, Chiak-san
JX158254
JX158017
JX158134
48
MMS-142
Onychodactylus koreanus
Korea
Gangwon, Chiaksan NP, Chiak-san
JX158246
JX158009
JX158127
49
H031
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158275
JX158035
JX158157
50
H032
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158276
JX158036
JX158158
51
H082
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158277
JX158037
JX158160
52
IK71
Onychodactylus koreanus
Korea
North Jeolla, Naedong
JX158271
JX158031
53
IK70
Onychodactylus koreanus
Korea
North Jeolla, Naedong
JX158270
JX158032
JX158154
54
H083
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158278
JX158038
JX158161
55
H084
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158279
JX158039
JX158159
56
MMS-109
Onychodactylus koreanus
Korea
South Gyeongsang, Jirisan NP
JX158280
JX158040
JX158162
9
57
MMS-104
Onychodactylus koreanus
Korea
Gangwon, Bongpyeong-myeon
JX158255
JX158018
JX158135
58
MMS-105
Onychodactylus koreanus
Korea
Gangwon, Bongpyeong-myeon
JX158256
JX158019
JX158136
59
MMS-242
Onychodactylus koreanus
Korea
Gangwon, Seokbyeong-san
JX158257
JX158020
JX158137
60
MMS-243
Onychodactylus koreanus
Korea
Gangwon, Seokbyeong-san
JX158258
JX158021
JX158138
61
MMS-244
Onychodactylus koreanus
Korea
North Jeolla, Jucheon-Myeon
JX158268
JX158029
JX158153
62
MMS-245
Onychodactylus koreanus
Korea
North Jeolla, Jucheon-Myeon
JX158269
JX158030
JX158152
63
MMS1753
Onychodactylus sp. A
Korea
Gangwon, Yangyang-gun
KX590686
64
MMS-257
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si, Dong-myeon
JX158281
JX158041
JX158163
65
MMS-444
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si, Dong-myeon
JX158282
JX158042
JX158164
66
MMS 7270 /
NAP-05131
Onychodactylus sillanus sp. nov.
Korea
North Gyeongsang, Unmunsan
OL771443
OL830810
OL792790
67
MMS10917
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si, Dong-myeon
MZ404936
68
MMS10918
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si, Dong-myeon
MZ404937
69
MMS10919
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Myriang-si
MZ404938
70
MMS443
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590671
71
MMS1350
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590672
72
MMS1333
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590673
73
MMS1334
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590674
74
MMS1335
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590675
75
MMS1331
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590676
76
MMS1343
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590677
77
MMS1349
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590678
78
MMS1344
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590679
79
MMS1339
Onychodactylus sillanus sp. nov.
Korea
South Gyeongsang, Yangsan-si
KX590680
10
We conducted the ML analysis in the IQ-TREE webserver
(http://iqtree.cibiv.univie.ac.at/). We employed 1,000 bootstrap pseudoreplicates via the
ultrafast bootstrap (UFBS; Hoang et al., 2018) approximation algorithm, and nodes having
ML UFBS values of 95 and above were a-priori considered highly supported, while the nodes
with values of 90–94 were considered well-supported, and the nodes with values of 70–89
were considered as tendencies. BI trees were generated using MrBayes v. 3.1.2 (Ronquist &
Huelsenbeck, 2003). The BI analyses were performed using three heated and one cold
Metropolis Coupled Markov Chain Monte Carlo chains for 200 million generations, with
sampling every 100 generations. We checked the convergence of the runs and that the
effective sample sizes were all above 200 by exploring the likelihood plots using TRACER v.
1.6 (Rambaut & Drummond, 2013). The initial 10% of trees were discarded as burn-in. After
burn-in, trees of two independent runs were combined in a single majority consensus
topology. Confidence in BI tree topology was assessed using the posterior probability (BI PP;
Huelsenbeck & Ronquist, 2001). We a priori considered BI PP 0.95 or greater as significant
support (Leaché & Reeder, 2002). For divergence time estimates within Onychodactylus we
relied on the earlier results by Suk et al. (2018).
Ecological niche modeling
We implemented ecological niche models (ENMs) to provide additional evidence of
divergence between O. koreanus and the candidate species from extreme south-eastern R.
Korea (Pyron & Burbrink, 2009; Rissler & Apodaca, 2007). To generate ENMs, we first
collected georeferenced occurrence points for O. koreanus and the candidate species from our
own survey data, the Global Biodiversity Information Facility (GBIF; accessed 7 October
2020; https://doi.org/10.15468/dl.khx67d), VertNet (searched under O. fischeri; accessed 22
September 2020), voucher specimens deposited in the Ewha Woman’s University Natural
History Museum (Shin et al., 2020a), and from the survey dataset of the National Institute of
Ecology accessed through the EcoBank platform (Kim et al., 2021). In total, we collected 921
occurrence points for O. koreanus and 28 occurrence points for the candidate species (Figure
1). Species assignment was based on the results of the genetic analyses. For environmental
variables, we used 19 bioclimatic and elevation layers downloaded from WorldClim 2.1
(https://www.worldclim.org; Fick & Hijmans, 2017), slope, coniferous, broadleaf and mixed
forest layers downloaded from EarthEnv (https://www.earthenv.org; Amatulli et al., 2018;
Tuanmu & Jetz, 2014), and Euclidean distance to water bodies derived from the elevation
layer in ArcMap 10.8.1 (ESRI, Redlands, CA). All environmental layers were in 1-km spatial
resolution. The same general data collection methodology and resulting occurrence dataset
for O. koreanus are also provided in Shin et al. (2021b).
Our methods for selecting environmental variables broadly follow Rissler & Apodaca
(2007). To select non-collinear environmental variables, we first sampled 50,000 random
points across the Korean Peninsula using the dismo package (Hijmans et al., 2021) in R
version 3.6.3 (R Core Team, 2020). Next, we extracted environmental raster values from
these random points and conducted a Spearman’s test in the R package ntbox (Osorio-Olvera
et al., 2020). Excluding highly correlated (coefficient > 0.7) variables, our environmental
layers were reduced to the following: annual mean temperature (bio1), mean diurnal range
(bio2), isothermality (bio3), annual precipitation (bio12), precipitation of wettest months
(bio13), precipitation seasonality (bio15), broadleaf forest cover, needleleaf forest cover,
mixed forest cover, slope, and distance to nearest water bodies. Next, we spatially thinned our
occurrence dataset to reduce spatial clustering of occurrence points. Using the spThin
package in R (Aiello-Lammens et al., 2015) and applying a 1-km thinning distance parameter,
we retained 187 unique occurrence points for O. koreanus and seven unique occurrence
11
points for the candidate species.This thinning resulted in the points used for the analysis to
be at least 1 km away from each other.
Using the reduced sets of environmental and occurrence data, we implemented ENMs
separately for O. koreanus and the candidate species in the software Maxent (version 3.4.1;
https://biodiversityinformatics.amnh.org/open_source/maxent). We used Maxent because of
the algorithm’s ability to explicitly account for the spatial bias of species occurrence points
(via a “bias file”; Shin et al., 2021a). Another benefit of Maxent is that the algorithm can be
implemented with small sample size (Pearson et al., 2007). While the number of independent
datapoints for the candidate species is small (n= 7), a previous study has demonstrated
predictive ability of Maxent models when implemented with sample size greater than n 5
(Pearson et al., 2007).
For O. koreanus, we specifically tuned the Maxent model parameters using the R
package ENMeval (Muscarella et al., 2014). Furthermore, using a published dataset used to
generate ENMs of O. koreanus (Shin et al., 2021b; http://dx.doi.org/10.17632/yc3nw4d9f2.2),
we employed target group background sampling to correct for spatial bias of O. koreanus
occurrence points between R. Korea and the Democratic People’s Republic of Korea
(hereafter D.P.R. Korea). From 48 candidate models run in ENMeval, we selected model
parameters with the Hinge feature and a regularization multiplier of 2 based on the AICc
value (Muscarella et al., 2014). In addition to these settings, we enabled random seed,
sampled 10,000 target group background points, set maximum iteration to 500, used 30% of
data for model testing, and applied Maximum Training Sensitivity plus Specificity threshold
rule (Shin et al., 2021b). Using these parameters, the ENM for O. koreanus was calibrated on
the entire extent of mainland Korean Peninsula (see Shin et al., 2021b for justification of
modelling extent).
For the candidate species we implemented Maxent using random seed, auto features, a
default regularization multiplier (= 1), 10,000 random background points sampled within a
20-km radius buffer around each occurrence point, testing data percentage of 20, and
maximum iterations of 500, and applied Minimum Training Presence threshold rule. To
generate the ENM for the candidate species we first calibrated the model on the
environmental layers clipped by a 20-km circular buffer around each occurrence point. This
model was then projected across the entire extent of mainland Korean Peninsula.
For both clades, we run 10 bootstrap replicates of the ENMs, and we averaged the
results of 10 replicate runs for downstream analyses. We evaluated model performance using
the area under the Receiver Operating Characteristic curve (AUC), True Skill Statistic (TSS;
Allouche et al., 2006), and the plotted distribution. Regarding the ENM for the candidate
species we implemented an additional model validation with a jackknife test as described in
Pearson et al. (2007) due to the small sample size used for model construction.
Niche differentiation
We implemented two tests to investigate niche differentiation between O. koreanus and
the candidate species. First, we examined whether the ENMs generated for each clade were
identical using niche identity test. We run 100 Maxent replicates using the same set of
occurrences and the environmental dataset used for niche modeling. We used Schoener’s D
and Istatistics to summarize the test results.
Next, we conducted the ribboning analysis as described in Glor & Warren (2011). This
analysis tests for the presence of unsuitable habitat separating two suitable regions. The
mountains inhabited by each O. koreanus and the candidate species are separated by a low
elevation plain called the Yangsan Fault unsuitable for the species (Kyung, 2010), with a
known impact on phylogeographic structures (Bae & Suk, 2015). Thus we considered the
ribboning analysis to be most appropriate to test niche differentiation between the two
12
Onychodactylus among the three types of range breaking analyses introduced in Glor &
Warren (2011). To implement the ribboning analysis, we first defined the region of
unsuitable habitat (“ribbon”) between the ranges of O. koreanus and the candidate species by
joining five outermost occurrence points of O. koreanus and two outermost occurrence points
of the candidate species in a minimum convex polygon (MCP). Next, we sampled 97 random
points (half the number of O. koreanus and the candidate species occurrences combined)
within this MCP, representing a third “clade” inhabiting a zone of habitat unsuitable to both
Onychodactylus clades. Therefore, we generated separate ENMs for O. koreanus, the
candidate species, and the “ribbon clade” in 100 Maxent replicates and used Schoener’s D
and Istatistics to summarize the test results. Both niche identity and ribboning tests were
implemented using the R package ENMTools (Warren et al., 2021).
Climate change
As with other members of the genus Onychodactylus, the candidate species is likely
prone to anthropogenic climate change (Borzée et al., 2019a; Kuzmin, 1995; Shin et al.,
2021b; Suk et al., 2018). Thus, we projected the ENM of current habitat suitability to two
representative concentration pathways (RCPs) corresponding to intermediate (RCP 4.5) and
extreme (RCP 8.5) climate change scenarios for the years 2050 and 2070. Following a
previous study, we selected the CCSM4 climate model under the Coupled Model
Intercomparison Project Phase 5 (CMIP5) downloaded from WorldClim 2.1 (Borzée et al.,
2019a; Shin et al., 2021b). For model projections, only the bioclimatic variables were
changed according to each RCP and time frame, and topographic and vegetation layers
remained constant for the lack of corresponding climate change scenarios for these variables.
Using the projected output models and the corresponding minimum training presence
threshold values, we also calculated the total area of suitable habitat under each climate
change scenario. We conducted model projections and area calculations within the 20-km
circular buffer surrounding all known occurrence points of the candidate species.
Conservation assessment
We conducted a threat assessment following the IUCN Red List categories and criteria due
their robustness (Maes et al., 2015) and the representation of threats at the global scale (Mace
et al., 2008). The IUCN Red List classifies species into three threatened categories:
Vulnerable, Endangered or Critically Endangered. Evaluations are conducted against
quantitative thresholds for five criteria that determine whether a species is at risk of
extinction: A, population size reduction; B, geographic range size; C, small population size
and decline; D, very small population and/or restricted distribution; and E, quantitative
analysis of extinction risk. A species can be assessed for any or all of the five criteria
simultaneously and if it does not meet one of these criteria it is placed into one of the non-
threatened IUCN categories (Least Concern and Near Threatened). Here, we follow the
IUCN Red List categories and criteria (IUCN Species Survival Commission, 2012) to suggest
an assessment for the candidate Onychodactylus species.
All threats, habitats, uses and trades are presented following the IUCN Red List criteria and
categories (www.iucnredlist.org/resources/redlistguidelines), and in the order specified. The
focal clade here is impacted by residential and commercial developments in parts of its range
(similarly to Hynobius yangi; Borzée & Min, 2021), but it is not known to be used or traded.
The clade is found along cold forested streams with high humidity and dense leaf litter (Hong,
2017). In addition, the clade is also under generalized threats such as habitat degradation and
climate change resulting in the warming and deoxygenation of waters at the country level
(Borzée, 2020; Jane et al., 2021), a critical set of variables for salamanders (Borzée et al.,
2019a; Mahler & Bourgeais, 2013; Zhang et al., 2020). Global warming was called to be
13
limited to a 1.5 °C increase above preindustrial levels by the Paris Climate Agreements
(Cop21, 2015), a difficult but insufficient request (Mundaca et al., 2019) under which the
environment is still expected to be severely impacted, and especially water resources (IPCC,
2014, 2018). Even if the targets specified by the Paris Climate Agreements are reached, air
temperature in northeast Asia is still expected to increase by 2.7 °C (peaking at 7.0 °C under
a 4 °C scenario if the Paris Climate Agreements are not held; Xu et al., 2017), and humidity
by 3.3% (13% under a 4 °C scenario; Xu et al., 2017).
Population size reduction (category A)
The population size reduction is measured over the longer of 10 years or 3 generations,
here 21 to 27 years based on the development time of the genus (Lee & Park, 2016), and a
longevity expected to be around 17 years for a closely related species (Smirina, 1994). When
the number of individuals is not known, the population reduction can be projected through a
decline in area of occupancy (AOO), extent of occurrence (EOO) and/or habitat quality
(Category A3(c)), here we used the values provided by the projection for the impact of
climate change on the habitat.
Geographic range (category B)
The geographic range is expressed in terms of AOO or EOO, and here we rely on the EOO
as we do not have data for the AOO. First, we converted the continuous probability map from
the Maxent output model into a binary presence/absence map. For this purpose, we used the
model calibrated on the buffered extent to prevent the inclusion of suitable area outside the
known range of the candidate species. We used the Minimum Training Presence threshold
calculated in Maxent as the cut-off for binary conversion. Next, we generated a minimum
convex polygon (MCP) connecting all known occurrence points of the candidate species (n=
27) and calculated the area of MCP using the R package ConR (Dauby, 2020).
Categories C, D and E
The criteria for small population size and decline, very small or restricted populations and
quantitative analyses rely on knowing the number of individuals. However, this information
is not available for this candidate species and these categories cannot be used in this study.
Results
Morphometrics
The morphometric analyses demonstrated significant differences between
Onychodactylus koreanus and the candidate Onychodactylus species from Yangsan and
Miryang (Supplementary Figure S1; Supplementary Table S4). The visual comparison of the
vomerine teeth showed that the candidate species is more similar to O. zhaoermii in shape but
the number of teeth is similar to O. koreanus, noting the small sample size used for
comparisons (Supplementary Figure S2).
Phylogenetic relationships
Onychodactylus fischeri from the Russian Far East was suggested as the sister
lineage to all remaining species of Onychodactylus with strong support (98/1.0, hereafter
node support values are given for ML UFBS/BI PP, respectively; Figure 1). Within the latter
group, all lineages of Onychodactylus from Japan (including O. japonicus,O.
nipponoborealis,O. tsukubaensis,O. fuscus,O. intermedius,O. kinneburi and O.
pyrrhonotus) grouped into a significantly supported monophylum (93/0.91; Figure 1). The
remaining mainland populations of Onychodactylus, at the exception of O. fischeri and O.
zhangyapingi from Jilin Province of northeast China, formed a strongly supported clade
hereafter referred to as O. koreanus species group (98/0.97; Figure 1). The phylogenetic
position of O. zhangyapingi is poorly supported, as this species showed a tendency to group
14
with the Japanese Onychodactylus clade, though with insignificant support; a similar
topology was also reported by Poyarkov et al. (2012).
Supplementary Figure S1. Morphometric comparison of limb length between
Onychodactylus koreanus and Onychodactylus sillanus sp. nov. PC3 was the only
significantly different PC between the two clades, represented by the length of front- and
hindlimbs. Both front- and hindlimbs are significantly longer in Onychodactylus sillanus sp.
nov. than in O. koreanus.
Supplementary Table S4. Descriptive statistics for morphological variables measured on
larvae of Onychodactylus koreanus and Onychodactylus sillanus sp. nov.The data is not
controlled for SVL as used for the analyses.
Onychodactylus sillanus sp. nov.
Onychodactylus koreanus
mean
SD
min
max
mean
SD
min
max
SVL
38.48
3.84
31.68
45.04
35.65
4.23
27.49
40.86
TL
33.02
4.07
26.18
39.98
31.51
5.96
18.88
39.24
GA (right)
19.57
2.23
14.68
23.01
18.33
2.46
12.87
22.57
CW
5.71
0.96
3.79
7.19
5.57
0.71
4.34
7.26
FFL (right)
7.97
1.32
5.48
9.96
6.55
0.92
5.18
8.02
HLL (right)
8.74
1.58
6.43
11.04
7.59
1.33
4.02
9.44
HL
8.19
1.11
6.10
10.32
8.04
0.79
5.47
9.17
HW
7.38
0.86
5.85
8.50
6.99
0.89
5.32
9.16
EL (right)
2.51
0.33
1.87
3.13
2.38
0.25
1.91
3.06
IN
3.56
0.43
2.36
4.30
3.20
0.53
2.12
4.45
ON(right)
1.55
0.24
1.15
2.09
1.35
0.41
0.63
2.26
IO
2.16
0.43
1.17
2.84
1.93
0.34
1.20
2.42
IC
4.16
0.35
3.67
4.87
3.84
0.47
3.02
4.51
Within the O. koreanus species group our analyses revealed four well-supported
clades (100/1.0) of putatively species-level differentiation in agreement with the earlier
15
results of Suk et al. (2018). The O. zhaoermii populations from Liaoning in northeast China
were suggested as a sister lineage to all remaining members of O. koreanus species group
inhabiting the Korean Peninsula (95/0.91).
The uncorrected genetic p-distance calculated from the Cyt bgene fragment between
the candidate Onychodactylus species from Yangsan and Miryang and other congeners varied
from p= 13.74 (with O. fischeri) to p= 6.49 (with O. koreanus) and p= 6.59 (with
Onychodactylus sp. A from Gangwon; see Supplementary Table S5). The divergence time
between these lineages was estimated by Suk et al. (2018) as ca. 6.82 mya (5.38 8.20).
Despite the presence of several well-supported subclades within O. koreanus, the overall
topological relationship remains unresolved (Figure 1).
Supplementary Figure S2.Schematic drawing and details of vomeral teeth in palatal view
of open mouth. The shape of the vomeral tooth row in Onychodactylus sillanus sp. nov. is
similar to O. zhaoermii but number of teeth is similar to O. koreanus (noting that n= 3 for O.
sillanus sp. nov.). (A) Onychodactylus koreanus from Poyarkov et al., 2012. (B)
Onychodactylus zhaoermii from Poyarkov et al., 2012. (C) Onychodactylus sillanus sp. nov.,
the head shape is modified from Poyarkov et al., 2012 for graphic consistency. (D+E)
Onychodactylus sillanus sp. nov. pictures when swabbing. (F) Holotype Onychodactylus
sillanus sp. nov.
16
Supplementary Table S5. Uncorrected p-distance (percentage) between the sequences of Cyt bmtDNA gene (below the diagonal), and
intraspecific genetic p-distance (above the diagonal) of Onychodactylus species included in phylogenetic analyses.
Species
1
2
3
4
5
6
7
8
9
10
11
12
13
1
O. fischeri
0.91
1.01
0.98
1.01
0.98
1.07
1.00
1.06
0.90
0.82
0.94
0.87
2
O. zhangyapingi
15.20
0.83
0.78
0.84
0.87
0.83
0.89
0.89
0.91
0.89
0.94
0.81
3
O. nipponoborealis
14.41
9.05
0.70
0.78
0.81
0.69
0.75
0.74
0.85
0.78
0.90
0.85
4
O. japonicus
14.59
9.05
6.15
0.69
0.75
0.73
0.74
0.72
0.86
0.77
0.86
0.84
5
O. kinneburi
14.67
8.96
7.56
6.77
0.64
0.73
0.84
0.77
0.86
0.81
0.95
0.86
6
O. pyrrhonotus
14.24
10.90
8.26
8.26
5.27
0.88
0.86
0.83
0.91
0.79
0.90
0.88
7
O. tsukubaensis
14.50
9.05
6.41
6.85
7.91
8.96
0.62
0.72
0.77
0.76
0.84
0.74
8
O. intermedius
14.94
10.28
7.12
8.35
8.70
9.49
5.54
0.60
0.91
0.85
0.96
0.85
9
O. fuscus
14.76
9.75
6.59
7.12
7.38
8.96
5.45
4.22
0.82
0.87
0.93
0.84
10
O. zhaoermii
15.01
11.60
9.30
10.44
9.83
10.79
9.83
11.06
10.44
0.21
0.75
0.74
0.79
11
O. koreanus
14.48
12.32
9.16
10.67
10.47
10.93
9.89
11.38
10.62
8.63
1.94
0.66
0.64
12
Onychodactylus sp. A
14.85
11.25
9.67
10.81
10.98
11.60
10.46
11.60
10.98
8.80
7.65
0.69
13
O. sillanus sp. nov.
13.74
10.19
8.53
9.24
9.39
9.74
8.75
10.34
10.10
6.96
6.49
6.59
0.17
17
Ecological niche modeling
In the ENM of O. koreanus, slope had the highest contribution, followed by annual
mean temperature (bio1), annual precipitation (bio12), mixed forest cover, needleleaf forest
cover, precipitation of wettest months (bio13), distance to nearest water bodies, broadleaf
forest cover, mean diurnal range (bio2), isothermality (bio3), and precipitation seasonality
(bio15; Supplementary Table S6). In contrast, the ENM for the candidate species showed that
the mean diurnal range (bio2) had the highest contribution, followed by needleleaf forest
cover, slope, distance to nearest water bodies, mixed forest cover, broadleaf forest cover,
annual precipitation (bio12), annual mean temperature (bio1), isothermality (bio3),
precipitation seasonality (bio15), and precipitation of wettest months (bio13; Supplementary
Table S6).
Supplementary Table S6. Contribution of environmental variables used to generate
ecological niche models of Onychodactylus koreanus and Onychodactylus sillanus sp. nov.
The contribution of each variable to niche models is considerably different between species.
The asterisks indicate variables with highest contribution in each species.
Variables
Contribution (%)
Onychodactylus koreanus
Onychodactylus sillanus sp. nov.
Annual mean temperature (bio1)
24.2
1.4
Mean diurnal range (bio2)
0.2
35.0*
Isothermality (bio3)
0.1
1.0
Annual precipitation (bio12)
7.8
3.2
Precipitation of wettest months (bio13)
2.1
0
Precipitation seasonality (bio15)
0.1
0.3
Broadleaf forest cover
0.6
3.7
Needleleaf forest cover
5
31.5
Mixed forest cover
5.8
5.1
Slope
53.3*
12.5
Distance to nearest water bodies
0.8
6.2
The ENMs for both species were also congruent with the known occurrence points. The
predicted habitat suitability of O. koreanus largely followed the main mountain ranges and
associated mountain chains in R. Korea and extends northward into central D.P.R. Korea. On
the other hand, the predicted area of high suitability for the candidate species was mostly
restricted to the area around Yangsan and Miryang (Supplementary Figure S3). When
projected across the Korean Peninsula, the ENM for the candidate species predicted
additional areas of intermediate to high habitat suitability (0.4 ~ 1.0) on the southwestern and
eastern coasts of R. Korea, as well as on the northeastern and western coasts of D.P.R. Korea
(Supplementary Figure S3).
18
Supplementary Figure S3. Comparison of Maxent habitat suitability models between
Onychodactylus koreanus and Onychodactylus sillanus sp. nov. Note the significant
differences in predicted habitat suitability between the two species across the Korean
Peninsula. While some overlap of predicted habitat suitability can be found, the actual known
ranges of O. koreanus and Onychodactylus sillanus sp. nov. do not overlap.
Genus Onychodactylus Tschudi, 1838
Onychodactylus sillanus sp. nov. Min, Borzée & Poyarkov
This is the supplementary data to the publication http://zoobank.org/
urn:lsid:zoobank.org:act:C62AAA19-197A-4B6C-BA6C-AD231FDA0874 where the species
is officially described.
Three specimens were designated as paratypes, including CGRB15898 (field ID MMS6688;
Supplementary Figure S4)
Measurements and counts of the type series. Measurements and counts of the type series
are presented in Supplementary Table S4 & S7.
19
Supplementary Figure S4. Onychodactylus sillanus sp. nov.paratype (CGRB15898;
Yangsan; 35.306°N, 129.074°E) in preservative. (A) General dorsal view of the paratype; (B)
general ventral view of the paratype. SVL = 82.67 mm. Note the presence of black keratinous
claws specific to the species.
20
Supplementary Table S7. Measurements (in mm) and counts of the type series of
Onychodactylus sillanus sp. nov. For character abbreviations see Materials and Methods.
Specimen ID
CGN(L)
CGN(R)
VTN(L)
VTN(R)
SVL
TL
GA
IC
CGRB15897
11
11
21
22
54.8
51.8
26.5
6.5
CGRB15906
11
11
18
19
51.0
53.2
27.0
5.7
CGRB 15907
12
12
20
20
54.8
56.9
27.9
5.7
Specimen ID
CW
CGBL
MAXTH
MTH
MTW
1-FL
2-FL
VTW
CGRB15897
6.3
-1
7.1
4.7
4.0
1.6
2.2
4.5
CGRB15906
7.6
-1
-
5.1
4.6
1.7
2.0
-
CGRB 15907
7.5
-1
4.3
4.3
3.8
1.8
2.5
4.6
Specimen ID
HLL
HL
HW
EL
IN
ON
IO
IC
OR
CGRB15897
15.0
12.0
8.5
3.4
5.0
2.7
3.2
6.5
5.2
CGRB15906
14.5
12.3
8.0
3.0
4.7
2.1
3.0
5.7
3.9
CGRB 15907
18.5
14.3
8.6
3.8
5.0
2.0
2.7
5.7
4.5
Specimen ID
4-FL
1-TL
2-TL
3-TL
4-TL
5-TL
VTL
VTW
CGRB15897
2.0
1.6
2.1
3.1
3.0
2.7
1.9
4.5
CGRB15906
1.9
1.5
2.0
3.0
2.8
2.6
-
-
CGRB 15907
2.1
1.9
2.6
3.7
3.3
2.1
1.9
4.6
Additional measurements for the holotype CGRB15897 include (all in mm): OR: 5.2; 1-FL:
1.6; 2-FL: 2.2; 3-FL: 3.3; 4-FL: 2.0; 1-TL: 1.6; 2-TL: 2.1; 3-TL: 3.1; 4-TL: 3.0; 5-TL: 2.7;
VTL: 1.9; VTW: 4.5; MTH: 4.7; MTW: 4.0; MAXTH: 7.1. Additional counts for the
holotype CGRB15897 include: CGN(L): 11; CGN(R): 11; CGBL: -1; VTN(L): 21; VTN(R):
22. Additional measurements (all in mm) and counts for CGRB15906 (field ID MMS-7207)
include: OR: 3.9; 1-FL: 1.7; 2-FL: 2.0; 3-FL: 2.6; 4-FL: 1.9; 1-TL: 1.5; 2-TL: 2.0; 3-TL: 3.0;
4-TL: 2.8; 5-TL: 2.6; MTH: 5.1; MTW: 4.6; MAXTH: 5.1; CGN(L): 11; CGN(R): 11; CGBL:
-1; VTN(L): 18; VTN(R): 19. Additional measurements (all in mm) and counts for CGRB
15907 (field ID MMS 7270 / NAP-05131) OR: 4.5; 1-FL: 1.8; 2-FL: 2.5; 3-FL: 3.1; 4-FL:
2.1; 1-TL: 1.9; 2-TL: 2.6; 3-TL: 3.7; 4-TL: 3.3; 5-TL: 2.1; VTL: 1.9; VTW: 4.6; MTH: 4.3;
MTW: 3.8; MAXTH: 4.3. Additional counts include: CGN(L): 12; CGN(R): 12; CGBL: -1;
VTN(L): 20; VTN(R): 20.
Additional morphological measurements of referred specimens. We also measured the
characters given in Supplementary Table S4 for two adults (one male and one female), and
six juveniles (larvae), collected from Miryang, Republic of Korea (35.4489°N, 128.9708 °E;
110 m a.s.l.), which were subsequently released upon taking the measurements. Male (in mm):
SVL: 82.18, TL: 102.3, GA right: 41.5, CW: 9.2, FFL right: 17.0, FFL left: 17.2, HLL right:
22.8, HLL left: 22.3, HL: 15.2, HW: 12.3, EL right: 5.3, EL left: 5.4, IN: 6.7, ON right: 3.3,
ON left: 3.7, IO: 4.4, OR right: 5.8, OR left: 6.1, IC: 7.9, CGN right: 11, CGN left: 11.
Female (in mm): SVL: 75.29, TL: 89.7, GA right: 36.9, CW: 7.6, FFL right: 15.9, FFL left:
15.4, HLL right: 17.0, HLL left: 16.7, HL: 14.3, HW: 11.4, EL right: 4.7, EL left: 4.6, IN: 5.8,
ON right: 2.7, ON left: 2.8, IO: 4.0, OR right: 5.3, OR left: 4.6, IC: 6.0, CGN right: 11, CGN
left: 11. Juveniles (n= 6; mean ± SD): SVL: 89.3 ± 8.04, TL: 41.91 ± 3.78, GA right: 18.53 ±
9.25, CW: 5.51 ± 0.57; FFL right: 12.02 ± 1.39, FFL left: 11.93 ± 1.33, HLL right: 13.06 ±
1.99, HLL left: 13.12 ± 1.86, HL: 11.52 ± 0.64, HW: 8.43 ± 0.52, EL right: 3.54 ± 0.18, EL
left: 3.68 ± 0.2, IN: 4.58 ± 0.43, ON right: 2.13 ± 0.36, ON left: 2.12 ± 0.4, IO: 3.11 ± 0.2,
OR right: 4.2 ± 0.7, OR left: 3.88 ± 0.68, IC: 4.96 ± 0.17, CGN right (median): 11 ± 1, CGN
left (median): 11 ± 1.
Description of the holotype (Supplementary Figure S5). A subadult specimen (presumably,
female) fully metamorphosed, in a poor state of preservation (toes on right hindlimb decayed
21
and almost absent; ca. 1 cm of the tail tip damaged; forelimbs partially dehydrated), fixed and
preserved in 75% ethanol, with a SVL of 54.8 mm (measured on the preserved specimen;
Supplementary Figure S5).
Head flattened, comparatively long and narrow, much longer than wide (HL/HW
ratio 1.41), head noticeably wider than neck (Supplementary Figure S5C). Tongue large,
elliptical, distinctly widening towards its anterior free end. Snout long (OR/HL ratio 0.43);
snout rounded in lateral view (Supplementary Figure S5E), obtusely rounded in dorsal view
(Supplementary Figure S5C). Nostrils small, with dorsolateral orientation, not protuberant,
widely spaced and distant from snout tip (IN/HL ratio 0.42). Eyes protuberant in dorsal and
lateral views (Supplementary Figure S5C, E), eye diameter shorter than snout length and the
internarial distance (EL/OR ratio 0.65; EL/IN ratio 0.68). Orbits close to each other,
interorbital distance short (IO/HL ratio 0.26; EL/IC ratio 0.52). Eyelids present, well-
developed; labial folds absent; gular fold distinct. Parotoid glands absent. Distinct
longitudinal postorbital groove running from the posterior corner of eye towards the angle of
the mouth; postorbital groove almost the same length as the eye; a distinct transverse groove
running upwards from the mouth angle to join the postorbital groove, approximately two
times shorter than latter. A distinct longitudinal groove running posteriorly of the mouth
angle, separating the parotoid area from the swollen elongated subparotoidal protuberance,
and further continuing to the gular fold. The subparotoidal protuberance distinct, elongated,
oval-shaped, swollen. Vomerine teeth in two transverse notably bended shallow arch-shaped
series, forming a «^^»-shaped figure (Supplementary Figure S5H); the medial ends of the
inner branches of the left and right vomerine tooth series contacting each other; 21 vomerine
teeth in the left vomerine tooth series and 20 vomerine teeth in the right series; the inner
branches slightly curving anteriorly; the outer branches notably longer than the inner
branches; their lateral ends located more posteriorly than the ends of the inner branches
(Supplementary Figure S5H).
Body elongated, cylindrical and slender; chest rather narrow (CW/SVL ratio 0.12).
The skin on dorsum and venter smooth. A middorsal groove well-developed, extending from
the base of tail to the base of the head. Eleven costal grooves present both on the right and
left sides of the body. Cloaca as a short, elongated slit.
Limbs slender, hindlimbs slightly longer and more robust than the forelimbs; when
forelimb and hind limb are adpressed towards each other against the flank, the digit tips do
not meet; a distinct gap more than 1 costal segment present between toes and fingers of the
adpressed limbs; FLL/GA ratio 0.53; HLL/GA ratio 0.57. Palmar or tarsal tubercles on palms
or feet absent. Four fingers and five toes lacking digital webbing; in order of decreasing
length, the relative lengths of the fingers: F1 < F4 < F2 < F3; relative length of toes: T1 < T2
< T5 < T4 < T3. Tips of fingers and toes rounded, black cornified claws present on all fingers
(Supplementary Figure S5F, G), not discernible on toes (since the hindlimbs are partially
decayed). Hindlimbs with a fleshy skinfold running along the posterior edge of the hindlimb.
Tail short and its tip partially damaged (TL/SVL ratio 1.54). Tail cylindrical in transverse
section at the anterior third of its length; turning to laterally compressed at the middle of its
length. Caudal fin absent. Tail tip slightly tapered.
Coloration of the holotype (Supplementary Figure S5). Coloration of the holotype in life
unknown; after seven years in preservative dorsum dark-brown with light beige or yellowish
spots on the dorsal surfaces of head, trunk, limbs and tail (Supplementary Figure S5A).
Ventrally uniform light greyish-brown with indistinct greyish mottling on chest and trunk and
tail flanks (Supplementary Figure S5B). The principal pattern of the original coloration does
not change and is clearly recognizable.
22
Supplementary Figure S5. Holotype of Onychodactylus sillanus sp. nov. (CGRB15897,
subadult) from South Gyeongsang Province, Yangsan-si, Dong-myeon, Republic of Korea, in
preservative. A: Dorsal view; B: Ventral view; C: Head, dorsal view; D: Head, ventral view;
E: Head, lateral view; F: Opisthenar view of left hand; G: Volar view of left hand; H: Palatal
region showing the vomerine tooth series; J: Ventral view of cloacal area. Photos by Nikolay
A. Poyarkov.
23
Supplementary Figure S6. Paratype of Onychodactylus sillanus sp. nov. (CGRB 15907,
subadult) from North Gyeongsang Province, Unmun Mountain, Republic of Korea, in life. A:
Dorsal view; B: Ventral view; C: Head, dorsal view; D: Head, ventral view; E: Head, lateral
view; F: Volar view of left hand; G: Plantar view of left foot; H: Ventral view of cloacal area.
Photos by Nikolay A. Poyarkov.
24
Variations (Supplementary Figures S1 S2 & S6 S7). The paratypes resemble the
holotype very closely in all aspects of coloration and pattern. Meristic and mensural
variations are presented in Supplementary Table S7. Specimen CGRB 15907 is in a good
state of preservation and has the following life coloration (Supplementary Figure S6).
Dorsum dark slate-black, varying to dark-lavender-brown at the flanks of body and tail
(Supplementary Figure S6A). Flanks of tail and body distinctly lighter than dorsum, dark
lavender-gray. Costal grooves darker than costal segments. Head laterally reddish-brown
turning lavender-gray ventrally (Supplementary Figure S6E). Ventral surfaces bright violet-
grey or lavender-grey (Supplementary Figure S6B). Ventral surfaces of the belly slightly
translucent. Limbs dorsally dark reddish-brown, ventrally lavender gray to dark gray. Dorsal
sides of the body, head and tail covered with numerous bright-orange to golden elongated
spots, forming a denser reticulated pattern on dorsal surfaces of head, getting more
longitudinally elongated at mid-trunk and further to the dorsal surfaces of the tail. Light
markings vary in size. On body and head flanks the light spots become less intense in
coloration; they merge with the light ventral coloration along the belly margins; light
markings absent on the ventral side of the body, tail or head. Iris brownish with dark copper
speckles dorsally and ventrally (Supplementary Figure S6E). The pupil is horizontal and oval
shaped.
Natural history. Biology of the new species is poorly studied. As in other Onychodactylus
species, O. sillanus sp. nov. occurs in moist, cool and shady places in evergreen and mixed
montane forests in the mountains of Yangsan and Miryang in the south-eastern region of the
Korean Peninsula, where it was recorded from 83 to 871 m a.s.l. (299.3 ± 179; Mean ± SD).
The new species inhabits clean, fast-running streams in mountain areas, in association with
pine trees or deciduous forests. Adults were recorded in the upstream on elevation from 280
to 800 m a.s.l.; whereas larvae and metamorphs were observed also in lower elevations
(Figure 1 and Supplementary Figure S7) hiding under stones or tree-logs in the streambed.
Supplementary Figure S7. Onychodactylus sillanus sp. nov. in life. A: adult male from
North Gyeongsang Province, Yangsan-si, Republic of Korea, in life (not collected); B:
subadult paratype (CGRB 15907) from North Gyeongsang Province, Unmun Mountain,
Republic of Korea, in life. A: metamorph from North Gyeongsang Province, Yangsan-si,
Dong-myeon, Republic of Korea, in life (not collected); B: larva from North Gyeongsang
Province, Yangsan-si, Dong-myeon, Republic of Korea, in life (not collected),
Adults and juveniles were found at night, active on large rock boulders on a steam bed,
always in proximity of water. The species seems to favour wet warm night of very late
autumn to disperse along streams, or on roads on rainy days. Larvae can be found during both
25
daytime and night time, throughout most of the year, at the exception of the coldest months.
In Yangsan area, O. sillanus sp. nov. is found in sympatry with Hynobius yangi,Glandirana
emeljanovi and Rana uenoi, with larvae potentially feeding on other amphibian larvae and
freshwater invertebrates. Like in other species of Onychodactylus, breeding in the new
species most likely takes place in streams or slow-moving subterranean water bodies;
however, the details of its reproductive biology remain unknown.
Current distribution and climate change. The models for 2070 predicted 24.4 km2of
suitable habitat under the RCP 4.5 and 37.6 km2of suitable habitat under the RCP 8.5 (Figure
1). Under the RCP 4.5 scenario, these reductions in suitable habitat translate into 87.6%
(MCP) and a 95.9% (Maxent binary) reduction by 2050, and 90.5% (MCP) and 96.9%
(Maxent binary) reduction by 2070. Under the RCP 8.5 scenario, the reductions reach 91.6%
(MCP) and 97.3% (Maxent binary) by 2050 and 85.5% (MCP) and 95.3% (Maxent binary)
by 2070.Modelling placed the EOO of O. sillanus sp. nov. at 792.7 km2based on the Maxent
binary map and 258.3 km2based on the MCP (Supplementary Figure S8).
26
Supplementary Figure S8. Potential impacts of future climate change on the suitable habitat
of Onychodactylus sillanus sp. nov. The ecological niche models projected on to four climate
change scenarios under two timeframes all predicted significant and drastic contraction of
suitable habitat area. Depending on the scenario and time frame, as much as 97% reduction of
suitable habitat area was predicted. *The Maxent ranges is binary encoded for presence and
absence based on the threshold determined in the text.
Conservation assessment. Based on the international database protected planet
(https://www.protectedplanet.net/), the habitat suitability of O. sillanus sp. nov. overlaps with
several protected area of significance. The species is also present in the five following
protected areas, or within its vicinity and most likely to be found in the area: Hoedong Water
Source Protection Area (www.protectedplanet.net/555622430), Gyeongnam Yangsan Habuk-
myeon 1 Wildlife Protection Area (www.protectedplanet.net/555625924), Gyeongnam
Yangsan Habuk-myeon 2 Wildlife Protection Area (www.protectedplanet.net/555625925),
Cheong-do Unmunsan Ecosystem and Landscape Conservation Area
(www.protectedplanet.net/555558241) and Unmunsan County Park
(www.protectedplanet.net/555571343).
The preliminary assessments provided by the ConR package put O. sillanus sp. nov. in the
Endangered category based on B1a and B2a criteria.
Discussion
By describing the species with a focus on conservation we aim at highlighting the
general extinction denial surrounding the current 6th mass extinction (Lees et al., 2020). This
global problem is also prevalent in the Republic of Korea (Lee & Miller-Rushing, 2014),
despite the potential for conservation being currently at its highest (Borzée et al., 2019b).
This stems from the availability of monetary resources following the rapid development of
the nation (Kim, 2005) coupled with the general interest for a betterment of conservation
activities in the nation when supported by popular support, as seen with Bottlenose dolphins
(Kim & Tatar, 2018) and Asiatic black bears (Andersen et al., 2021). However, a high
number of extinctions is foreseeable in the near future (NIBR, 2014), with numerous species
in need of urgent actions (Borzée et al., 2018; Hong et al., 2007; Kim et al., 2013; Zöckler et
al., 2010). Onychodactylus sillanus sp. nov. is currently the source of legal actions and citizen
actions to protect its habitat from urbanisation and groundwater development (Choi, 2020a,
2020b; Kim, 2021). Action plan for amphibians generally include a list of main threats and