Next Article in Journal
Zooplankton Indicators of Ecological Functioning Along an Urbanisation Gradient
Previous Article in Journal
Medicinal and Aromatic Plant Diversity of Greece: Biodiversity Knowledge, Ethnobotany and Sustainable Use—A Short Review
Previous Article in Special Issue
The Effects of Light Environment on Adult Odonate Communities in Disturbed and Intact Forest: The Importance of Small-Scale Effects
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 18
Issue 1
10.3390/d18010057
diversity-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Distribution and Potential Dispersal Corridors of Two Onychodactylus Species in the Republic of Korea

by
Young-Guk Kim
1,†,
Hahyun Nam
1,†,
Jaejin Park
2,
Jiho Park
1 and
Daesik Park
2,*
1
Interdisciplinary Program in Earth Environmental System Science & Engineering, Kangwon National University, Chuncheon 24341, Gangwon State, Republic of Korea
2
Department of Science Education, Kangwon National University, Chuncheon 24341, Gangwon State, Republic of Korea
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Diversity 2026, 18(1), 57; https://doi.org/10.3390/d18010057
Submission received: 27 December 2025 / Revised: 15 January 2026 / Accepted: 21 January 2026 / Published: 22 January 2026
(This article belongs to the Collection Disturbance Ecology: A New Emerging Field in Basic/Applied Ecology and Conservation)
Browse Figures
Versions Notes

Abstract

Accurate information regarding species boundaries is essential for ecological research and conservation planning. This information is particularly difficult to obtain but essential for cryptic amphibian species. The distribution and potential dispersal corridors of two cryptic salamander species, the Korean clawed (Onychodactylus koreanus) and the Yangsan clawed (O. sillanus) salamanders, were investigated using integrated approaches for high-resolution species distribution modeling (SDM), genetic species identification, and habitat connectivity analysis. The SDM results showed high habitat suitability in mid- and high-mountainous areas, but very low suitability in riverine areas for both species. Genetic species identification of the 25 populations delimited the distribution boundary between the two species along the Nakdong and Geumho rivers. Dispersal corridors of the two species commonly involved a detour around the major rivers and produced only one possible dispersal route, where both species moved into the opposite species’ habitat along the east side of the mountainous areas of the Geumho River. The findings not only clarify the distribution range of two cryptic Onychodactylus species in the Republic of Korea but also highlight the importance of the unique dispersal route for studying species interactions and maintaining ecological connectivity.

1. Introduction

Understanding the spatial distribution of a species and the environmental factors shaping its range are central themes in biogeography and macroecology [1,2]. Information on the connectivity between subpopulations, resilience to population fluctuations, and the possibility of long-term viability is key to establishing effective species conservation strategies [3,4,5]. With the advancement of molecular phylogenetic studies, the identification of so-called cryptic species—lineages that are genetically distinct but morphologically indistinguishable—has become increasingly frequent across various animal groups [6,7,8,9,10]. When new species are described based on limited regional surveys rather than comprehensive geographic sampling, distributional boundaries with existing species are often ambiguous [11,12]. Obtaining accurate distributional and ecological data for specific populations is challenging when morphological discrimination is difficult and the distribution boundaries are unclear. Such ambiguity increases the risk of species misidentification during field surveys, leading to various problems in species conservation and management [13,14,15]. Therefore, research aimed at accurately defining the distribution boundaries of species is essential to support sound decision-making in species management.
Traditionally, the geographical boundaries of a species’ distribution have been determined by connecting the outermost points of individual occurrence [16]. Recently, species distribution models (SDMs) have been used to determine the distribution range of species [17,18,19]. These models are powerful tools for predicting potential habitats for species and have been widely used in studies on various taxa [20,21,22]. Despite the widespread adoption of these modeling techniques, many studies have conducted modeling with environmental variables at coarse resolutions ranging from 1 km (30 arc-seconds) to tens of kilometers, resulting in unclear distribution boundaries [17,23]. Coarse-resolution modeling often leads to the underestimation or distortion of a species’ potential habitat [24,25,26]. Therefore, high-resolution distribution modeling is necessary to precisely determine the ecological requirements and distributional characteristics of species and to accurately predict dispersal corridors between subpopulations [27,28,29,30,31]. Specifically, high-resolution modeling is imperative to determine the precise distribution boundaries of amphibians whose habitats are shrinking due to climate change and are being destroyed by human activities [32,33].
The clawed salamander of the genus Onychodactylus lacks lungs and relies on cutaneous respiration; therefore, it inhabits mountainous streams with high dissolved oxygen and low water temperatures [34,35]. Their long larval period of over two years and the low dispersal ability of metamorphosed individuals makes them vulnerable to local stream environment alterations and habitat fragmentation [34,36]. The Onychodactylus species inhabiting the Korean Peninsula have low genetic diversity [37] and are vulnerable to climate change [32], necessitating rapid conservation and management. Two Onychodactylus species, the Korean clawed salamander (O. koreanus) [35], which inhabits most of the Korean Peninsula, and the Yangsan clawed salamander (O. sillanus), which is found in the southeastern part of the Korean Peninsula, are cryptic species [33]. Due to the difficulty in distinguishing them morphologically, the distribution boundary between the two species is unclear, increasing the possibility of errors in field surveys, and delaying the establishment of effective conservation measures.
In the current study, the distribution and potential dispersal corridors of two cryptic species, the Korean clawed salamander (O. koreanus) and Yangsan clawed salamander (O. sillanus) was investigated using integrated approaches of high-resolution SDM, genetic species identification, and habitat connectivity analysis. The results provide an important basis for future research on their biogeographic and evolutionary relationships and establish key baseline data for conservation planning and habitat management strategies.

2. Materials and Methods

2.1. Study Area

Onychodactylus koreanus is distributed in most of the Republic of Korea, whereas O. sillanus is found only in the southeastern part of the country [33,35]. The previously known range of O. sillanus is smaller than that of O. koreanus and is geographically enclosed by the latter species [33]. Therefore, the spatial extent of the current study encompassed the entire distribution range of O. sillanus, as well as the surrounding populations of O. koreanus, from 34.4–36.8° N latitude and 127.3–130° E longitude (Figure 1). Geographically, the study area included the Taebaek and Sobaek mountain ranges, which serve as primary mountain habitats for these species. The study area also encompasses major rivers, including the Nakdong, Geumho, and Hyeongsan rivers (Figure 1). Mapping and spatial data visualization were conducted using QGIS software (ver. 3.34.15; QGIS Development Team, 2025) [38].

2.2. Preparing Data for Species Distribution Modeling

2.2.1. Online Database

The occurrence data of the two species used in the analyses were primarily collected from the results of the fourth (2006–2013) and fifth (2014–2022) National Natural Environment Surveys by the Korean Ministry of Environment (NES; accessed on 16 April via the Ecobank platform) [39] and the Natural Resources Survey by the Korea National Park Service from 2010 to 2022 (Table S1). Additional occurrence data were obtained from the Global Biodiversity Information Facility (GBIF. Available online: https://doi.org/10.15468/dl.k6zedm (accessed on 15 April 2025)). Initially, 2602 records of O. koreanus and 27 records of O. sillanus were collected. From this dataset, only the occurrence points within the study area, determined using the terra package in R (v.4.5.1), were extracted [40,41]. If the occurrence points are concentrated within specific regions or pixels, the models can overfitted and reduce the predictive performance [42]. To avoid spatial autocorrelation and ensure environmental representativeness, a spatial thinning process was conducted using the spThin package (ver. 0.2.0) with a 1 km grid [43]. A total of 425 records for O. koreanus and 22 records for O. sillanus were obtained (Figure 1; Table S1).

2.2.2. Environmental Variables

To construct the species distribution models, eight environmental variables were established based on previous studies [26,31,44,45]. The topographic variables were derived at a spatial resolution of 30 m (1 arc-second) using a digital elevation model (DEM) provided by the United States Geological Survey [46]. The five indices were elevation; slope; aspect; topographic roughness index (TRI), which quantifies terrain complexity; and topographic position index (TPI), which characterizes local landforms such as ridges and valleys [47]. The normalized difference vegetation index (NDVI) was included for shelter availability and moisture retention in amphibians [48]. The NDVI data were obtained from the Environmental Big Data Platform (https://www.bigdata-environment.kr, accessed on 9 October 2025) and derived from the 2019 Landsat 8 satellite imagery (bands 4 and 5) provided by the USGS and NASA.
For the two climate variables, annual mean temperature (AMT) and annual precipitation (AP) were calculated at a 1 km (30 arc-seconds) resolution using the data from the Korean Meteorological Administration (KMA; https://data.kma.go.kr/, accessed on 9 October 2025) from 1991–2020. To capture the local temperature and precipitation patterns associated with the mountainous regions of the study area, the resolution of the climate variables was downscaled to match the 30 m (1 arc-second) topographic variables. Using a spatial downscaling technique based on Random Forest (RF), the relationship between high-resolution topographic variables (e.g., elevation, slope, and aspect) as predictors and climate variables was determined. The climate variables obtained from the KMA were then matched to 30 m resolution topographic variables [49].

2.3. Species Distribution Modeling

2.3.1. Model Construction and Parameter Tuning

The maximum entropy (MaxEnt) algorithm was used to predict the potential distribution of the two Onychodactylus species using the dismo package (ver. 1.3-16) in R (ver. 4.5.1) [50]. The MaxEnt model uses a regression-based maximum entropy approach and high distribution prediction outputs can be expected even with relatively small occurrence data. Therefore, it is a widely used species distribution modeling technique not only for amphibians but also for various taxa [18]. Random sampling of the distribution data of both species was used, with 80% used as training data and 20% as testing data [18]. We applied the 80% data for training, considering the small sample size of O. sillanus. The species distribution model was constructed by using five combinations of mathematical transformation functions (feature combinations; FC: L, LQ, LQH, LQHP, LQHPT; L = linear, Q = quadratic, H = hinge, P = product, T = threshold) that describe the relationship between environmental variables and the probability of species occurrence, and four regularization multipliers (RM = 0.5, 1.0, 1.5, 2.0) that prevent the model from overfitting the training data [51], resulting in a total of 20 models (Figure S1).

2.3.2. Model Selection and Evaluation

The performance of the constructed species distribution model was evaluated using the area under the curve (AUC) and true skill statistic (TSS) metrics. The AUC was evaluated as ‘excellent’ when 0.9–1.0, ‘good’ when 0.8–0.9, ‘fair’ when 0.7–0.8, and ‘poor’ when 0.6 or less [52]. The optimal model for each species was selected based on the combination that yielded the highest AUC and TSS values. To prevent model overfitting and verify the generalization performance, a 10% omission rate (OR10) was used as a secondary indicator. Reliability was assessed by comparing the observed omission rate, calculated using the 20% testing data, against the theoretical expectation of 0.1 [53]. The relative influence of the eight environmental variables on the selected optimal models was evaluated using percentage contribution and permutation importance [18]. The information usefulness and independence of each variable were verified using the jackknife test [18]. Response curves were examined to visualize the changes in species distribution probability along individual variable shifts [18,54]. To visualize habitat suitability patterns, the continuous habitat suitability index (HSI) was reclassified into five categories at 0.2 intervals using the equal interval method [55]. The areas with an HSI value of 0.8 or higher were considered as suitable habitat areas [55].

2.4. Field Survey and Genetic Species Identification

Based on known distribution data [32,33,56], field surveys were conducted and individuals at 25 sites where O. koreanus and O. sillanus were likely to coexist were sampled to determine the distribution range of each species based on genetic species identification (Table S2). Given that the two species are cryptic with difficult morphological distinctions [33,35], genetically identifying them in areas of potential co-occurrence could increase the reliability of the results from the SDM. At each site, an average of 6.2 (±3.6 SD, n = 25, range = 1–10) larvae were captured and the geographic coordinates of the site were recorded (Figure 1). A 2–3 mm section from the tail tip was obtained and preserved in 95% ethanol. After sampling, all individuals were immediately released at the capture site.
Genomic DNA was extracted from the tissue using a DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) by following the manufacturer’s protocol. The concentration of the extracted DNA was measured using a NanoDrop Lite Plus Ultraviolet-visible (UV-Vis) spectrophotometer (NDL-PLUS-PR, Thermo Fisher Scientific Inc., Waltham, MA, USA). The mitochondrial Cytochrome b (Cytb) gene (1141 bp) of O. koreanus and O. sillanus was amplified from the extracted DNA using the primers 70 F and RN2 [57], with two internal primers, 520 F [57] and OFIR (5′-CCT GTT GGG TTA TTT GAG CC-3′), following a previous phylogenetic study [37].
The polymerase chain reaction (PCR) was conducted in a total volume of 20 µL, containing 10 µL of TOPsimple PCR PreMIX-nTaq (Enzynomics, Daejeon, Republic of Korea), 1 µL of genomic DNA (2 ng/µL), 1 µL each of forward and reverse primers (250 nM), and 7 µL of distilled water. Thermal cycling conditions of the PCR were 94 °C for 4 min, followed by 32 cycles of 94 °C for 30 s, 50 °C for 50 s, and 72 °C for 80 s, with a final extension at 72 °C for 7 min [37]. The PCR products were sent to NBIT (Chuncheon, Republic of Korea) for sequencing. The sequencing results were aligned and refined using the MUSCLE plugin [58] in Geneious Prime v2022.0.2 (https://www.geneious.com), resulting in obtaining a complete Cytb gene sequence (1141 bp) from all samples. For species identification, sequences were queried against the NCBI BLAST (ver. 2.17.0) database, based on the highest identity value among sequences with a query cover of 99% or more.
All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Kangwon National University (KW-240422-1). The samples were collected with the appropriate permissions from local cities and counties (Yeongyang-gun, Cheongsong-gun, Chilgok-gun, Yeongcheon-si, Pohang-si, Gyeongju-si, Sangju-si, Gimcheon-si, and Cheongdo-gun, North Gyeongsang Province, Hapcheon-gun, Uiryeong-gun, and Yangsan-si, North Gyeongsang Province, Gurye-gun and Gwangyang-si, South Jeolla Province, Dalseong-gun, Daegu Metropolitan City, Sasang-gu and Gijang-gun, Busan Metropolitan City, Republic of Korea) and the Korean National Park Research Institute.

2.5. Identification of Potential Dispersal Corridors

2.5.1. Niche Overlap Analysis

Based on environmental variables constructed at a 30 m resolution, the ecological niches of the two species were compared in the environmental space [59]. To define the environmental space, a principal component analysis (PCA) of the environmental variables extracted using the ENMTools package (ver. 1.1.5) in R [60] was conducted. A two-dimensional environmental space was established using the first two PCs (PC1 and PC2) [59]. Subsequently, the occurrence data of O. koreanus (n = 425) and O. sillanus (n = 22) were projected onto this PC space to visualize their distribution patterns [59].
To quantify the degree of ecological niche overlap between the two species, Schoener’s D [61] and Warren’s I [62] were calculated. Both indices range from 0 (no overlap) to 1 (complete overlap), with values closer to 0 indicating greater ecological niche heterogeneity and values closer to 1 indicating greater similarity. The statistical significance of niche overlap was assessed using the niche identity test and symmetric background test using the ENMTools package (ver. 1.1.5) in R [59,62]. All statistical tests were performed with 1000 iterations and the null hypothesis was rejected if the observed overlap value fell outside the confidence interval of the null distribution.

2.5.2. Habitat Connectivity Evaluation

Dispersal corridors were identified by analyzing habitat connectivity in the potential habitats predicted by the two species distribution models. First, the potentially suitable habitats of the two species were inversely transformed to create resistance surfaces (RS), representing the movement costs for each species [63]. The calculated RS values were from 0 to 1, with lower values indicating areas where movement was easy and higher values indicating areas where movement was restricted [63,64]. For the dispersal corridor analysis, 25 genetically verified points were used as focal nodes.
The least-cost path (LCP) and cost-weighted distance (CWD) between each pair of nodes were calculated using the gdistance package (ver. 1.6.5) in R [63,65]. After converting the resistance surface into a transition matrix representing the probability of movement between adjacent pixels, pathways with the minimum accumulated cost between the start and end points of the nodes were derived [63]. After deriving the corridors, their quality (corridor quality, CQ = CWD/LCP) was evaluated [66]. A smaller CQ value indicated a more efficient route, whereas a larger CQ value indicated a route with a higher topographical resistance and lower movement efficiency.
The specific pathways with low CQ values, particularly in the potential areas where O. koreanus and O. sillanus could move into the habitat of the opposite species were selected and visualized, given the results from the species distribution models. When selecting these specific pathways from the total corridors identified from the LCP analysis, cases where the pathways crossed the rivers that would act as a dispersal barrier were excluded, based on a previous study of Ambystoma californiense [67]. Whether the two species could meet and compete at specific areas on the pathway was evaluated.

3. Results

3.1. Species Distribution Modeling

The species distribution models of the two Onychodactylus species showed good predictive performance, with an AUC of 0.877 (±0.017) and TSS of 0.610 (±0.021) for O. koreanus and an AUC of 0.844 (±0.027) and TSS of 0.555 (±0.036) for O. sillanus. The mean and range of the eight variables across the study area and at the occurrence points of O. koreanus and O. sillanus were presented (Table 1). In the model of O. koreanus, elevation contributed 52.3%, TPI contributed 19.5%, and slope contributed 15.4% to the model. The contributions of the remaining variables were less than 5.2% (Table S3). For O. sillanus, elevation, slope, TPI, AMT, TR, NDV, and aspect contributed 33.1%, 17.8%, 15.4%, 10.7%, 9.0%, 7.0%, and 6.9% to the model, respectively. The AP was low at 0.1% (Table S4).
The response curves of eight environmental variables indicated that both species commonly exhibited the highest occurrence probabilities in deep valley terrain, with TPI values between −15 and −10 and in regions where AP was <1300 mm. Onychodactylus koreanus showed a high probability of occurrence in areas characterized by high altitude (>1000 m); gentle, north-facing slopes of 5–10°; and low NDVI values (<0.1). In contrast, O. sillanus was most likely to occur in the areas of mid-elevation (250–500 m); moderate TRI of 2–6; southeast- to southwest-facing slopes of 100–250°; high NDVI values (>0.7); and AMP <11 °C (Figure 2).
The habitat suitability of both species was low in the basins surrounding the Nakdong, Geumho, and Hyeongsan rivers, but high in mountainous regions (Figure 3). The most suitable habitats for O. koreanus were found in the mountains of Juwang and Myeonbong within the Taebaek mountain range, as well as in the mountains of Deogyu, Gaya, and Jiri within the Sobaek mountain range (Figure 3A). The habitats within the Sobaek mountain range demonstrated high spatial connectivity (Figure 3A). Additionally, such habitats were found at high elevations in Biseul, Gaji, Cheonseong, and Geumjeong, where O. sillanus was reported. Highly suitable habitats for O. sillanus were widely distributed across the Taebaek and Sobaek mountain ranges (Figure 3B). These habitats were fragmented and widely distributed across mid-elevations in the Songni, Deogyu, Gaya, and Jiri mountains. Within the areas where O. sillanus was reported, such habitats were found in the high-elevation regions of the mountains of Unmun, Gaji, and Cheonseong (Figure 3B).

3.2. Field Survey and Genetic Species Identification

Of the 25 populations whose species were determined using Cytb sequences, 16 were O. koreanus and 9 were O. sillanus (Figure 1; Table S2). There were no coexisting populations. Onychodactylus sillanus was found only in areas east of the Nakdong River and in the southern parts of the Geumho River. Only O. koreanus was detected in the remaining areas.

3.3. Identification of Potential Dispersal Corridors

The PCA yielded three PCs with eigenvalues >1. PC1 explained 38.68% of the total variance with slope, TRI, elevation, and NDVI as the main variables. PC2 explained 14.38% of the variance with AP and AMT, and PC3 explained 13.36% of the variance with TPI and slope (Table 2). In the environmental space, using PC1 and PC2, the environmental niche areas of the two species did not differ (Figure S2). Neither the symmetric background test (D = 0.411 and I = 0.626) nor the niche identity test (D = 0.353 and I = 0.575) indicated differences between the two species (p > 0.05; Figure S2).
The primary dispersal corridors of the two species on the resistance surfaces were identified as the pathways connecting the mountains within the Taebaek and Sobaek mountain ranges, and they did not cross major waterways, such as the Nakdong and Geumho rivers (Figure 4). The mean corridor quality indices (CQ) for the entire corridor of O. koreanus and O. sillanus were similar at 0.32 and 0.35, respectively. The dispersal corridors of O. sillanus were more complex in the mountains of Deogyu, Gaya, Jiri, and Juwang than those of O. koreanus. Based on the distribution modeling and genetic species identification, O. koreanus could move into the O. sillanus habitat only along the east side mountain parts of the Geumho River, from mountains Palgong and Bohyeon to Unmun and Biseul, with a CQ value of 0.397 (±0.0096; Figure 4C). Onychodactylus sillanus could also move into O. koreanus habitat using only the same route from mountains Biseul, Nam, Gaji, and Unmun to Bohyeon, with a CQ value of 0.293 (±0.047; Figure 4D).

4. Discussion

Using genetic species identification, high-resolution species distribution modeling, and habitat connectivity analysis, this study indicated that the cryptic species, the Korean clawed salamander (O. koreanus) and the Yangsan clawed salamander (O. sillanus), have distribution boundaries along the Nakdong and Geumho rivers. The habitat of O. koreanus was concentrated in high-elevation mountains, whereas O. sillanus was widely distributed and fragmented in low- and mid-elevation mountains. The only route where the two species moved into the opposite species’ habitat was in the mountainous areas between the Geumho and Hyeongsan rivers. The results clearly delineate the distribution range of the two cryptic species.
High-resolution SDM facilitates more accurate prediction of species distribution. High-resolution SDM incorporates complex environmental heterogeneity and reflects local climatic patterns, providing more detailed species distribution characteristics [26,27,28]. For example, the improved resolution of the model for the yellow-spotted newt (Neurergus derjugini) has indicated that the micro-topographic structure of valleys, often overlooked by coarse models, plays a key role in maintaining connectivity between habitats [31]. In salamanders, SDM has been used to elucidate ecological isolation mechanisms in Plethodon albagula [68] and to predict potential habitats and connectivity in Chinese Onychodactylus species [69]. In the Republic of Korea, SDM of O. koreanus has been conducted twice, with spatial resolutions of 1 km and 2 km, respectively [56,69]. Compared to those studies, the current high-resolution analysis has increased the spatial resolution of habitat areas, elucidating the connectivity between habitats. Furthermore, areas where the two species could not inhabit, with an HSI of 0–0.2, were clearly identified, highlighting the boundaries between suitable and unsuitable habitats.
Despite the spatial niche overlap, each species exhibited specific habitat characteristics. The major habitats of O. koreanus were located in high-elevation mountainous areas within the Taebaek and Sobaek mountain ranges. Habitat connectivity was particularly strong within the Sobaek mountain range. This characteristic is consistent with known habitat characteristics, such as high mountain habitat, cold water streams, and dense forest [34]. Even if climate change shifts populations northeastward and toward high-elevation areas [56,70], these mountain ranges may provide climatic refugia for O. koreanus. A suitable habitat for O. koreanus was identified at high-elevation areas in the Biseul, Cheonseong, and Geumjeong mountains within the distribution range of O. sillanus, whose identity was genetically determined. The possible presence of O. koreanus on the tops of mountains within these areas should be questioned in the future. Unlike O. koreanus, suitable habitats for O. sillanus were found throughout the study area, but highly suitable habitats appeared to be fragmented. To the best of our knowledge, no previous study on the SDM of O. sillanus has been conducted. Because of the low habitat suitability in high-elevation mountains, population connectivity was low, unlike in O. koreanus. The probability of occurrence of O. sillanus was high at low- and mid-altitude elevations with flat terrain. Such areas appeared on specific slopes at specific elevations within a single mountain, resulting in fragmented suitable habitats. A similar pattern of fragmented distribution was reported in yellow-spotted newts (Neurergus derjugini) [28]. Despite these characteristics, suitable habitats for O. sillanus were found in the habitats of O. koreanus, whose identity was genetically determined. These results suggest that natural barriers and interspecific competition may limit the northward and westward dispersal of O. sillanus or the southward and eastward dispersal of O. koreanus [71,72].
The major river systems function as key dispersal barriers for O. koreanus and O. sillanus. Mountain ranges and rivers are well-known barriers that induce species isolation and allopatric speciation in diverse taxa, including mammals [73], reptiles [74], and amphibians [67,75]. For example, river systems are closely related to the formation of genetic lineages of spiny-bellied frogs (Quasipaa boulenger) [76]. River systems limit the dispersal and migration of the natterjack toad (Bufo calamita) and California tiger salamander (Ambystoma californiense) [67,75]. The current SDM results indicated that habitat suitability of both species was very low along major rivers, such as the Nakdong, Geumho, and Hyeongsan rivers. Potential dispersal corridors for both species avoided the major rivers. The genetic species determination in the 25 populations also supports the distribution ranges of the two species being clearly separated by the Nakdong and Geumho rivers. Major rivers therefore function as dispersal and isolation barriers for O. koreanus and O. sillanus, preventing their eastward and westward dispersal, respectively. Phylogenetically, O. koreanus and O. sillanus diverged approximately 6.82 million years ago (Late Miocene) [33,37,67,75]. The timeline is consistent with tectonic events on the Korean Peninsula, where the eastern part of the Korean Peninsula was uplifted, the Taebaek mountain range formed, and the major river systems flowing to the southwest were established [67,75,77,78]. Tectonic events and the development of the Nakdong river system may have driven the differentiation of the two species and subsequently isolated O. koreanus and O. sillanus in the east/west and north/south parts of the Nakdong and Geumho rivers, respectively. The studies on the divergence of the genus Onychodactylus on the Korean Peninsula using mitochondrial and nuclear markers could give more clear insights on the historical boundary establishments of them.
The overlapping dispersal route between the two species provides a key area for studying their potential interaction. Dispersal corridor analysis using LCP has been used to assess habitat connectivity, gene flow, and the impact of habitat disturbance [79,80]. For example, corridors have been identified as facilitating genetic exchanges in the eastern tiger salamander (Ambystoma tigrinum), which has lower landscape resistance than physical distance [81]. In the current study, after the corridor analysis, routes with low CQ values were extracted from the potential corridors present between the Geumho and Hyeongsan rivers. This area was identified as a unique area in which O. koreanus and O. sillanus could move into each other’s distribution areas through southward and northward dispersal, respectively. A unique dispersal route along mountainous areas at an elevation of 500 m was found between the Geumho and Hyeongsan rivers. The discovery of this unique dispersal route has several implications for future research. It is possible that O. koreanus prevents the northward dispersal of O. sillanus through interspecific spatial competition. The possibility of coexistence and hybridization between the two species along this route can be investigated. Furthermore, intricate historical migration pathways can be further explored utilizing more sophisticated analytical tools such as Migration-N [82]. Considering the distribution range shifts in amphibians following climate change [83,84], the possible southward or northward dispersal of species could also be detected along this route. To understand their distribution and conserve populations despite threats from climate change, attention should be given to the identified dispersal route.

5. Conclusions

The distribution patterns of two cryptic O. koreanus and O. sillanus species on the Korean Peninsula were determined using 30 m high-resolution SDM. Based on the SDM results and genetic species identification in the field populations, the distribution range and boundaries of the two species were determined. The Nakdong and Geumho rivers function as dispersal and isolation barriers and a unique dispersal route between the two species along the east side of the mountainous areas of the Geumho River was identified. The results ameliorate species identification ambiguity in field surveys and facilitate the study of contact zones and possible hybridization of the dispersal route. The study provides an example of how uncertainty in distribution boundaries between cryptic species can be resolved.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d18010057/s1, Figure S1: Feature classes and regularization multipliers used to build the species distribution models of the two Onychodactylus species in the study area; Figure S2: The result of the niche overlap analysis: (A) the quantified niche overlap between Onychodactylus koreanus (blue) and O. sillanus (red) in the two-dimensional environmental space defined by eight environmental variables, (B) the results of niche identity test (left column) and symmetric background test (right column) with Schoener’s D and Warren’s I indices; Table S1: The species occurrence data of two Onychodactylus species in the study area used for species distribution modeling; Table S2: Geographic coordinates and GenBank accession numbers of two Onychodactylus species samples collected in the study area; Table S3: Percent contribution and permutation importance of eight environmental variables in the final species distribution model (SDM) of Onychodactylus koreanus selected from 20 models based on area under the curve (AUC) and true skill statistic (TSS) indices; Table S4: Percent contribution and permutation importance of eight environmental variables in the final species distribution model (SDM) of Onychodactylus sillanus selected from 20 models based on area under the curve (AUC) and true skill statistic (TSS) indices.

Author Contributions

Conceptualization, Y.-G.K. and D.P.; methodology, Y.-G.K., H.N. and D.P.; formal analysis, Y.-G.K. and H.N.; investigation, H.N., J.P. (Jaejin Park), J.P. (Jiho Park) and D.P.; resources, D.P.; data curation, Y.-G.K. and H.N.; writing—original draft preparation, Y.-G.K., H.N. and D.P.; writing—review and editing, Y.-G.K., J.P. (Jaejin Park) and D.P.; visualization, Y.-G.K.; supervision, J.P. (Jaejin Park) and D.P.; project administration, Y.-G.K.; funding acquisition, Y.-G.K. and D.P. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by a grant from the National Institute of Biological Resources (NIBR, NIBRE202501) and the National Research Foundation of Korea (RS-2024-00346579), funded by the Ministry of Environment (MOE) of the Republic of Korea.

Institutional Review Board Statement

The animal study protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of the Kangwon National University (KW-240422-1 and 20 June 2024).

Data Availability Statement

We included all the data in Supplementary Materials. However, some of the data are available upon reasonable request to the corresponding author following the data policy of the Korean National Park Service.

Acknowledgments

We thank Jongsun Kim, Min-Woo Park, Narae Joo, Jaebum Jeong, Heesoo Lee for their field assistance and the Korean National Park Service for allowing to use the distribution data.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AMTAnnual mean temperature
APAnnual precipitation
AUCArea under the curve
CQCorridor quality
CWDCost-weighted distance
DEMDigital elevation model
FCFeature combination
HHinge
HISHabitat suitability index
KMAKorea meteorological administration
KNPSKorea national park service
LLinear
LCPLeast cost path
MaxEntMaximum entropy
NDVINormalized difference vegetation index
NESNational ecosystem survey
OR1010% omission rate
PProduct
PCPrincipal component
PCAPrincipal component analysis
QQuadratic
RFRandom forest
SDMSpecies distribution model
TThreshold
TPITopographic position index
TRITopographic roughness index
TSSTrue skill statistic
USGSUnited States geological survey

References

  1. Brown, J.H. Macroecology; University of Chicago Press: Chicago, IL, USA, 1995; ISBN 0-226-07615-6. [Google Scholar]
  2. Lomolino, M.V. Biogeography: A Very Short Introduction; Oxford University Press: New York, NY, USA, 2020; ISBN 978-0-198-85006-9. [Google Scholar]
  3. Hanski, I. Metapopulation dynamics. Nature 1998, 396, 41–49. [Google Scholar] [CrossRef]
  4. Moilanen, A.; Hanski, I. On the use of connectivity measures in spatial ecology. Oikos 2001, 95, 147–151. [Google Scholar] [CrossRef]
  5. Crooks, K.R.; Sanjayan, M. Connectivity Conservation; Cambridge University Press: New York, NY, USA, 2006; ISBN 978-0-521-67381-5. [Google Scholar]
  6. Kusuma, W.E.; Ratmuangkhwang, S.; Kumazawa, Y. Molecular phylogeny and historical biogeography of the Indonesian freshwater fish Rasbora lateristriata species complex (Actinopterygii: Cyprinidae): Cryptic species and west-to-east divergences. Mol. Phylogenet. Evol. 2016, 105, 212–223. [Google Scholar] [CrossRef] [PubMed]
  7. Hoang, C.V.; Nguyen, T.T.; Ninh, H.T.; Luong, A.M.; Pham, C.T.; Nguyen, T.Q.; Orlov, N.L.; Chen, Y.; Wang, B.; Ziegler, T.; et al. Two new cryptic species of Microhyla Tschudi, 1838 (Amphibia, Anura, Microhylidae) related to the M. heymonsi group from central Vietnam. ZooKeys 2021, 1036, 47–74. [Google Scholar] [CrossRef]
  8. Guo, P.; Wang, P.; Lyu, B.; Liu, Q.; Zheng, J.; Fu, C.; Wu, Y.; Shu, G.; Hou, S. Molecular phylogeny reveals cryptic diversity in Sibynophis from China (Serpentes: Sibynophiidae). Front. Ecol. Evol. 2023, 13, e10367. [Google Scholar] [CrossRef]
  9. Borzée, A.; Min, M.S. Disentangling the impacts of speciation, sympatry and the island effect on the morphology of seven Hynobius sp. salamanders. Animals 2021, 11, 187. [Google Scholar] [CrossRef]
  10. Niwa, K.; Nishikawa, K.; Matsui, M.; Kanamori, S.; Kuro-O, M. Taxonomic reassessment of salamanders (genus Hynobius) from Tsushima Islands, Japan, with a resurrection of Hynobius tagoi Dunn, 1923 (Amphibia: Caudata). Zootaxa 2023, 5339, 201–236. [Google Scholar] [CrossRef]
  11. Bergsten, J.; Bilton, D.T.; Fujisawa, T.; Elliott, M.; Monaghan, M.T.; Balke, M.; Hendrich, L.; Geijer, J.; Herrmann, J.; Foster, G.N.; et al. The effect of geographical scale of sampling on DNA barcoding. Syst. Biol. 2012, 61, 851–869. [Google Scholar] [CrossRef]
  12. Funk, W.C.; Caminer, M.; Ron, S.R. High levels of cryptic species diversity uncovered in Amazonian frogs. Proc. R. Soc. B Biol. Sci. 2012, 279, 1806–1814. [Google Scholar] [CrossRef]
  13. Bickford, D.; Lohman, D.J.; Sodhi, N.S.; Ng, P.K.L.; Meier, R.; Winker, K.; Ingram, K.K.; Das, I. Cryptic species as a window on diversity and conservation. Trends Ecol. Evol. 2007, 22, 148–155. [Google Scholar] [CrossRef]
  14. Bortolus, A. Error cascades in the biological sciences: The unwanted consequences of using bad taxonomy in ecology. Ambio 2008, 37, 114–118. [Google Scholar] [CrossRef] [PubMed]
  15. Angulo, A.; Icochea, J. Cryptic species complexes, widespread species and conservation: Lessons from Amazonian frogs of the Leptodactylus marmoratus group (Anura: Leptodactylidae). Syst. Biodivers. 2010, 8, 357–370. [Google Scholar] [CrossRef]
  16. Gaston, K.J. The Structure and Dynamics of Geographic Ranges; Oxford University Press: New York, NY, USA, 2003; ISBN 019-8-52641-5. [Google Scholar]
  17. Guisan, A.; Thuiller, W. Predicting species distribution: Offering more than simple habitat models. Ecol. Lett. 2005, 8, 993–1009. [Google Scholar] [CrossRef] [PubMed]
  18. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  19. Elith, J.; Graham, C.H.; Anderson, R.P.; Dudík, M.; Ferrier, S.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; Leathwick, J.R.; Lehmann, A.; et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 2006, 29, 129–151. [Google Scholar] [CrossRef]
  20. Raxworthy, C.J.; Martinez-Meyer, E.; Horning, N.; Nussbaum, R.A.; Schneider, G.E.; Ortega-Huerta, M.A.; Peterson, A.T. Predicting distributions of known and unknown reptile species in Madagascar. Nature 2003, 426, 837–841. [Google Scholar] [CrossRef]
  21. Kozak, K.H.; Wiens, J.J. Does niche conservatism promote speciation? A case study in North American salamanders. Evolution 2007, 60, 2604–2621. [Google Scholar] [CrossRef]
  22. Rondinini, C.; Di Marco, M.; Chiozza, F.; Santulli, G.; Baisero, D.; Visconti, P.; Hoffmann, M.; Schipper, J.; Stuart, S.N.; Tognelli, M.F.; et al. Global habitat suitability models of terrestrial mammals. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2011, 366, 2633–2641. [Google Scholar] [CrossRef]
  23. Fick, S.E.; Hijmans, R.J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 2017, 37, 4302–4315. [Google Scholar] [CrossRef]
  24. Trivedi, M.R.; Berry, P.M.; Morecroft, M.D.; Dawson, T.P. Spatial scale affects bioclimate model projections of climate change impacts on mountain plants. Glob. Change Biol. 2008, 14, 1089–1103. [Google Scholar] [CrossRef]
  25. Randin, C.F.; Engler, R.; Normand, S.; Zappa, M.; Zimmermann, N.E.; Pearman, P.B.; Vittoz, P.; Thuiller, W.; Guisan, A. Climate change and plant distribution: Local models predict high-elevation persistence. Glob. Change Biol. 2009, 15, 1557–1569. [Google Scholar] [CrossRef]
  26. Lembrechts, J.J.; Lenoir, J.; Roth, N.; Hattab, T.; Milbau, A.; Haider, S.; Pellissier, L.; Pauchard, A.; Backes, A.R.; Dimarco, R.D.; et al. Comparing temperature data sources for use in species distribution models: From in-situ logging to remote sensing. Glob. Ecol. Biogeogr. 2019, 28, 1578–1596. [Google Scholar] [CrossRef]
  27. Franklin, J.; Davis, F.W.; Ikegami, M.; Syphard, A.D.; Flint, L.E.; Flint, A.L.; Hannah, L. Modeling plant species distributions under future climates: How fine scale do climate projections need to be? Glob. Change Biol. 2013, 19, 473–483. [Google Scholar] [CrossRef] [PubMed]
  28. Storlie, C.J.; Phillips, B.L.; VanDerWal, J.; Williams, S.E. Improved spatial estimates of climate predict patchier species distributions. Divers. Distrib. 2013, 19, 1106–1113. [Google Scholar] [CrossRef]
  29. Zeilhofer, P.; Cezar, A.; Tôrres, N.M.; de Almeida Jácomo, A.T.; Silveira, L. Jaguar Panthera onca habitat modeling in landscapes facing high land-use transformation pressure—Findings from Mato Grosso, Brazil. Biotropica 2014, 46, 98–105. [Google Scholar] [CrossRef]
  30. Connor, T.; Hull, V.; Viña, A.; Shortridge, A.; Tang, Y.; Zhang, J.; Wang, F.; Liu, J. Effects of grain size and niche breadth on species distribution modeling. Ecography 2018, 41, 1270–1282. [Google Scholar] [CrossRef]
  31. Karami, P.; Tavakoli, S.; Esmaeili, M. Fine-scale habitat suitability and connectivity analysis for the core populations of yellow-spotted mountain pond-breeding newt (Neurergus derjugini) in the west of Iran and east of Iraq. Glob. Ecol. Conserv. 2023, 43, e02429. [Google Scholar] [CrossRef]
  32. Shin, Y.; Min, M.S.; Borzée, A. 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. Ecol. Evol. 2021, 11, 14669–14688. [Google Scholar] [CrossRef]
  33. Borzée, A.; Shin, Y.; Poyarkov, N.A.; Jeon, J.Y.; Baek, H.J.; Lee, C.H.; An, J.; Hong, Y.J.; Min, M.S. Dwindling in the mountains: Description of a critically endangered and microendemic Onychodactylus species (Amphibia, Hynobiidae) from the Korean Peninsula. Zool. Res. 2022, 43, 750–755. [Google Scholar] [CrossRef]
  34. Kuzmin, S.L. The Clawed Salamanders of Asia: Genus Onychodactylus: Biology, Distribution, and Conservation; Westarp Wissenschaften: Magdeburg, Germany, 1995; ISBN 978-3-894-32425-4. [Google Scholar]
  35. Poyarkov, N.A.; Che, J.; Min, M.S.; Kuro-o, M.; Yan, F.; Li, C.; Iizuka, K.; Vieites, D.R. Review of the systematics, morphology and distribution of Asian clawed salamanders, genus Onychodactylus (Amphibia, Caudata: Hynobiidae), with the description of four new species. Zootaxa 2012, 3465, 1–106. [Google Scholar] [CrossRef]
  36. Lee, J.H.; Ra, N.Y.; Eom, J.; Park, D. Population dynamics of the long-tailed clawed salamander larva, Onychodactylus fischeri, and its age structure in Korea. J. Ecol. Environ. 2008, 31, 31–36. [Google Scholar] [CrossRef]
  37. Suk, H.Y.; Lee, M.Y.; Bae, H.G.; Lee, S.J.; Poyarkov, N.A.; Lee, H.; Min, M.S. Phylogenetic structure and ancestry of Korean clawed salamander, Onychodactylus koreanus (Caudata: Hynobiidae). Mitochondrial DNA Part A DNA Mapp. Seq. Anal. 2018, 29, 650–658. [Google Scholar] [CrossRef]
  38. QGIS.org. QGIS, version 3.40.14; QGIS Association: Bern, Switzerland, 2025. [Google Scholar]
  39. Kim, H.W.; Yoon, S.; Kim, M.; Shin, M.; Yoon, H.; Kim, K. EcoBank: A flexible database platform for sharing ecological data. Biodivers. Data J. 2021, 9, e61866. [Google Scholar] [CrossRef] [PubMed]
  40. Hijmans, R.J. Package ‘Terra’. 2024. Available online: https://cran.r-project.org/web/packages/terra/index.html (accessed on 9 October 2025).
  41. R Core Team. R: A Language and Environment for Statistical Computing; R Core Team: Vienna, Austria, 2025. [Google Scholar]
  42. Boria, R.A.; Olson, L.E.; Goodman, S.M.; Anderson, R.P. Spatial filtering to reduce sampling bias can improve the performance of ecological niche models. Ecol. Model. 2014, 275, 73–77. [Google Scholar] [CrossRef]
  43. Aiello-Lammens, M.E.; Boria, R.A.; Radosavljevic, A.; Vilela, B.; Anderson, R.P. spThin: An R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 2015, 38, 541–545. [Google Scholar] [CrossRef]
  44. Park, D.; Jeong, H.; Park, J.; Park, I.K. Distribution and habitat assessments of the slender racer, Orientocoluber spinalis, for the registration of nationally endangered species in the Republic of Korea. Sci. Rep. 2023, 13, 12025. [Google Scholar] [CrossRef]
  45. Park, I.K.; Shin, Y.; Baek, H.J.; Kim, J.; Kim, D.I.; Seok, M.; Oh, Y.; Park, D. Establishment potential across South Korea for two gecko species, Gekko japonicus and G. swinhonis, adapted to different climates. NeoBiota 2024, 93, 39–62. [Google Scholar] [CrossRef]
  46. U.S. Geological Survey. USGS 3D Elevation Program Digital Elevation Model. Available online: https://elevation.nationalmap.gov/arcgis/rest/services/3DEPElevation/ImageServer (accessed on 9 October 2025).
  47. Amatulli, G.; Domisch, S.; Tuanmu, M.N.; Parmentier, B.; Ranipeta, A.; Malczyk, J.; Jetz, W. A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. Sci. Data 2018, 5, 180040. [Google Scholar] [CrossRef]
  48. Pettorelli, N.; Vik, J.O.; Mysterud, A.; Gaillard, J.M.; Tucker, C.J.; Stenseth, N.C. Using the satellite-derived NDVI to assess ecological responses to environmental change. Trends Ecol. Evol. 2005, 20, 503–510. [Google Scholar] [CrossRef]
  49. Hengl, T.; Walsh, M.G.; Sanderman, J.; Wheeler, I.; Harrison, S.P.; Prentice, I.C. Global mapping of potential natural vegetation: An assessment of machine learning algorithms for estimating land potential. PeerJ 2018, 6, e5457. [Google Scholar] [CrossRef]
  50. Hijmans, R.J.; Phillips, S.; Leathwick, J.; Elith, J. Dismo: Species Distribution Modeling. R Package Version 1.3-15. 2024. Available online: https://cran.r-project.org/web/packages/dismo/index.html (accessed on 9 October 2025).
  51. Phillips, S.J.; Dudík, M. Modeling of species distributions with Maxent: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  52. Swets, J.A. Measuring the accuracy of diagnostic systems. Science 1988, 240, 1285–1293. [Google Scholar] [CrossRef] [PubMed]
  53. Radosavljevic, A.; Anderson, R.P. Making better Maxent models of species distributions: Complexity, overfitting and evaluation. J. Biogeogr. 2014, 41, 629–643. [Google Scholar] [CrossRef]
  54. Yost, A.C.; Petersen, S.L.; Gregg, M.; Miller, R. Predictive modeling and mapping sage grouse (Centrocercus urophasianus) nesting habitat using Maximum Entropy and a long-term dataset from Southern Oregon. Ecol. Inform. 2008, 3, 375–386. [Google Scholar] [CrossRef]
  55. Yang, X.Q.; Kushwaha, S.P.S.; Saran, S.; Xu, J.; Roy, P.S. Maxent modeling for predicting the potential distribution of medicinal plant, Justicia adhatoda L. in Lesser Himalayan foothills. Ecol. Eng. 2013, 51, 83–87. [Google Scholar] [CrossRef]
  56. Lee, S.Y.; Choi, S.-Y.; Bae, Y.-S.; Suh, J.-H.; Jang, H.J.; Do, M.S. Distribution prediction of Korean clawed salamander (Onychodactylus koreanus) according to the climate change. Korean J. Environ. Ecol. 2021, 35, 480–489. (In Korean) [Google Scholar] [CrossRef]
  57. Yoshikawa, N.; Matsui, M.; Nishikawa, K.; Kim, J.B.; Kryukov, A. Phylogenetic relationships and biogeography of the Japanese clawed salamander, Onychodactylus japonicus (Amphibia: Caudata: Hynobiidae), and its congener inferred from the mitochondrial cytochrome b gene. Mol. Phylogenet. Evol. 2008, 49, 249–259. [Google Scholar] [CrossRef]
  58. Edgar, R.C. MUSCLE: Multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004, 32, 1792–1797. [Google Scholar] [CrossRef]
  59. Broennimann, O.; Fitzpatrick, M.C.; Pearman, P.B.; Petitpierre, B.; Pellissier, L.; Yoccoz, N.G.; Thuiller, W.; Fortin, M.J.; Randin, C.; Zimmermann, N.E.; et al. Measuring ecological niche overlap from occurrence and spatial environmental data. Glob. Ecol. Biogeogr. 2012, 21, 481–497. [Google Scholar] [CrossRef]
  60. Warren, D.L.; Matzke, N.J.; Cardillo, M.; Baumgartner, J.B.; Beaumont, L.J.; Turelli, M.; Glor, R.E.; Huron, N.A.; Simões, M.; Iglesias, T.L.; et al. ENMTools 1.0: An R package for comparative ecological biogeography. Ecography 2021, 44, 504–511. [Google Scholar] [CrossRef]
  61. Schoener, T.W. The anolis lizards of Bimini: Resource partitioning in a complex fauna. Ecology 1968, 49, 704–726. [Google Scholar] [CrossRef]
  62. Warren, D.L.; Glor, R.E.; Turelli, M. Environmental niche equivalency versus conservatism: Quantitative approaches to niche evolution. Evolution 2008, 62, 2868–2883. [Google Scholar] [CrossRef] [PubMed]
  63. van Etten, J.R. Package gdistance: Distances and routes on geographical grids. J. Stat. Softw. 2017, 76, 1–21. [Google Scholar] [CrossRef]
  64. Li, J.; Liang, L.; Chao, Y.; Wang, X.; Qiu, M.; Luo, P.; Zhu, Y. Construction of the ecological security pattern of Mu Us sandy land on the basis of the “Source − Resistance − Corridor” theory. Ecol. Indic. 2025, 171, 113162. [Google Scholar] [CrossRef]
  65. Adriaensen, F.; Chardon, J.P.; De Blust, G.; Swinnen, E.; Villalba, S.; Gulinck, H.; Matthysen, E. The application of ‘least-cost’ modelling as a functional landscape model. Landsc. Urban Plan. 2003, 64, 233–247. [Google Scholar] [CrossRef]
  66. Dutta, T.; Sharma, S.; McRae, B.H.; Roy, P.S.; DeFries, R. Connecting the dots: Mapping habitat connectivity for tigers in central India. Reg. Environ. Change 2016, 16, 53–67. [Google Scholar] [CrossRef]
  67. Wang, I.J.; Savage, W.K.; Shaffer, H.B. Landscape genetics and least-cost path analysis reveal unexpected dispersal routes in the California tiger salamander (Ambystoma californiense). Mol. Ecol. 2009, 18, 1365–1374. [Google Scholar] [CrossRef]
  68. Peterman, W.E.; Connette, G.M.; Semlitsch, R.D.; Eggert, L.S. Ecological resistance surfaces predict fine-scale genetic differentiation in a terrestrial woodland salamander. Mol. Ecol. 2014, 23, 2402–2413. [Google Scholar] [CrossRef]
  69. Peng, Y.; Li, Y.; Cao, G.; Li, H.; Shin, Y.; Piao, Z.; Perez, F.; Zhu, W.; Borzée, A. Estimation of habitat suitability and landscape connectivity for Liaoning and Jilin clawed salamanders (Hynobiidae: Onychodactylus) in the transboundary region between the People’s Republic of China and the Democratic People’s Republic of Korea. Glob. Ecol. Conserv. 2023, 48, e02694. [Google Scholar] [CrossRef]
  70. Shin, Y.; Borzée, A.; Park, D. Climatic data sources and limitations of ecological niche models impact the estimations of historical ranges and niche overlaps in distantly related Korean salamanders. BMC Ecol. Evol. 2025, 25, 105. [Google Scholar] [CrossRef]
  71. Soberón, J.; Peterson, A.T. Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodivers Inform. 2005, 2, 1–10. [Google Scholar] [CrossRef]
  72. Colwell, R.K.; Rangel, T.F. Hutchinson’s duality: The once and future niche. Proc. Natl. Acad. Sci. USA 2009, 106, 19651–19658. [Google Scholar] [CrossRef] [PubMed]
  73. Harcourt, A.H.; Wood, M.A. Rivers as barriers to primate distributions in Africa. Int. J. Primatol. 2012, 33, 168–183. [Google Scholar] [CrossRef]
  74. Jackson, N.D.; Austin, C.C. The combined effects of rivers and refugia generate extreme cryptic fragmentation within the common ground skink (Scincella lateralis). Evolution 2010, 64, 409–428. [Google Scholar] [CrossRef] [PubMed]
  75. Stevens, V.M.; Verkenne, C.; Vandewoestijne, S.; Wesselingh, R.A.; Baguette, M. Gene flow and functional connectivity in the natterjack toad. Mol. Ecol. 2006, 15, 2333–2344. [Google Scholar] [CrossRef]
  76. Yan, F.; Zhou, W.; Zhao, H.; Yuan, Z.; Wang, Y.; Jiang, K.; Jin, J.; Murphy, R.W.; Che, J.; Zhang, Y. Geological events play a larger role than Pleistocene climatic fluctuations in driving the genetic structure of Quasipaa boulengeri (Anura: Dicroglossidae). Mol. Ecol. 2012, 22, 1120–1133. [Google Scholar] [CrossRef]
  77. Chough, S.K.; Kwon, S.T.; Ree, J.H.; Choi, D.K. Tectonic and sedimentary evolution of the Korean peninsula: A review and new view. Earth Sci. Rev. 2000, 52, 175–235. [Google Scholar] [CrossRef]
  78. Son, M.; Song, C.W.; Kim, M.C.; Cheon, Y.; Jung, S.; Cho, H.; Kim, H.G.; Kim, J.S.; Sohn, Y.K. Miocene crustal deformation, basin development, and tectonic implication in the southeastern Korean Peninsula. J. Geol. Soc. Korea 2013, 49, 93–118. (In Korean) [Google Scholar] [CrossRef]
  79. Cushman, S.A.; McKelvey, K.S.; Hayden, J.; Schwartz, M.K. Gene flow in complex landscapes: Testing multiple hypotheses with causal modeling. Am. Nat. 2006, 168, 486–499. [Google Scholar] [CrossRef]
  80. McRae, B.H.; Dickson, B.G.; Keitt, T.H.; Shah, V.B. Using circuit theory to model connectivity in ecology, evolution, and conservation. Ecology 2008, 89, 2712–2724. [Google Scholar] [CrossRef]
  81. Spear, S.F.; Peterson, C.R.; Matocq, M.D.; Storfer, A. Landscape genetics of the blotched tiger salamander (Ambystoma tigrinum melanostictum). Mol. Ecol. 2005, 14, 2553–2564. [Google Scholar] [CrossRef]
  82. Beerli, P.; Mashayekhi, S.; Sadeghi, M.; Khodaei, M.; Shaw, K. Population genetic inference with Migrate. Curr. Protoc. Bioinform. 2019, 68, e87. [Google Scholar] [CrossRef]
  83. Araújo, M.B.; Thuiller, W.; Pearson, R.G. Climate warming and the decline of amphibians and reptiles in Europe. J. Biogeogr. 2006, 33, 1712–1728. [Google Scholar] [CrossRef]
  84. Das, B.; Das, A.; Baishya, K.K. Predicting impressions of climate change on the distribution of Crocodile newt (Tylototriton verrucosus), a rare amphibian in the Darjeeling Himalayan sub-region, India. Int. J. Ecol. Environ. Sci. 2023, 49, 17–29. [Google Scholar] [CrossRef]
Diversity 18 00057 g001
Figure 1. The study area and the occurrence locations in the southeastern parts of the Republic of Korea. The occurrence records collected from databases are in black dots, and the sampled populations are in blue (Onychodactylus koreanus) and red (O. sillanus) circles. Blue lines on the map indicate major rivers. Black triangles are major mountains within the Taebaek and Sobaek mountain ranges, indicated in brown.
Figure 1. The study area and the occurrence locations in the southeastern parts of the Republic of Korea. The occurrence records collected from databases are in black dots, and the sampled populations are in blue (Onychodactylus koreanus) and red (O. sillanus) circles. Blue lines on the map indicate major rivers. Black triangles are major mountains within the Taebaek and Sobaek mountain ranges, indicated in brown.
Diversity 18 00057 g001
Diversity 18 00057 g002
Figure 2. The response curves of eight environmental variables in the species distribution models of Onychodactylus koreanus (in blue) and O. sillanus (in red).
Figure 2. The response curves of eight environmental variables in the species distribution models of Onychodactylus koreanus (in blue) and O. sillanus (in red).
Diversity 18 00057 g002
Diversity 18 00057 g003
Figure 3. Habitat suitability distribution of Onychodactylus koreanus (A) and O. sillanus (B), generated by MaxEnt modeling in the southeastern parts of the Republic of Korea.
Figure 3. Habitat suitability distribution of Onychodactylus koreanus (A) and O. sillanus (B), generated by MaxEnt modeling in the southeastern parts of the Republic of Korea.
Diversity 18 00057 g003
Diversity 18 00057 g004
Figure 4. Potential dispersal corridors of Onychodactylus koreanus (in blue; (A)) and O. sillanus (in red; (B)), identified using least-cost path (LCP) analysis on the habitat suitability distribution surface of each species. A unique dispersal route for O. koreanus (in green; (C)) and O. sillanus (in brown; (D)) was identified in the areas where the two species had the possibility to meet and compete in mountainous areas between the Geumho and Hyeongsan rivers.
Figure 4. Potential dispersal corridors of Onychodactylus koreanus (in blue; (A)) and O. sillanus (in red; (B)), identified using least-cost path (LCP) analysis on the habitat suitability distribution surface of each species. A unique dispersal route for O. koreanus (in green; (C)) and O. sillanus (in brown; (D)) was identified in the areas where the two species had the possibility to meet and compete in mountainous areas between the Geumho and Hyeongsan rivers.
Diversity 18 00057 g004
Table 1. Environmental characteristics (mean ± SD, range) over the study area and at the occurrence points of two Onychodactylus species. TPI, topographic position index; TRI, topographic roughness index; NDVI, normalized difference vegetation index; AMT, annual mean temperature, °C; AP, annual precipitation, mm.
Table 1. Environmental characteristics (mean ± SD, range) over the study area and at the occurrence points of two Onychodactylus species. TPI, topographic position index; TRI, topographic roughness index; NDVI, normalized difference vegetation index; AMT, annual mean temperature, °C; AP, annual precipitation, mm.
VariableStudy AreaO. koreanusO. sillanus
Elevation (m)163.42 ± 212.27
(0.000–1891.946)		595.26 ± 293.22
(80.30–1519.15)		315.62 ± 151.82
(128.86–744.44)
Slope (°)9.53 ± 10.52
(0.000–68.793)		15.40 ± 8.07
(2.09–45.45)		14.34 ± 6.92
(1.35–34.93)
Aspect178.95 ± 102.54
(0.00–360.00)		180.50 ± 106.28
(0.34–359.28)		176.50 ± 89.97
(26.38–337.36)
TPI−3.17 × 10−5 ± 2.57
(−54.43–56.61)		−2.06 ± 3.73
(−17.60–20.44)		−1.66 ± 2.38
(−6.43–3.59)
TRI4.38 ± 4.50
(0.00–63.50)		7.70 ± 3.87
(0.97–24.63)		6.36 ± 2.72
(2.09–13.29)
NDVI0.73 ± 0.19
(−1.00–1.00)		0.83 ± 0.10
(−0.05–0.92)		0.77 ± 0.09
(0.49–0.88)
AMT12.61 ± 0.45
(10.01–14.32)		12.21 ± 0.43
(11.05–13.36)		12.29 ± 0.41
(11.12–13.14)
AP1285.33 ± 56.35
(1091.77–3070.31)		1338.48 ± 138.81
(1188.87–2146.79)		1269.04 ± 55.37
(1210.60–1520.31)
Table 2. The results of the principal component analysis of the eight environmental variables extracted from the study area in the Republic of Korea. The percentages indicate the amount of variation explained by each PC.
Table 2. The results of the principal component analysis of the eight environmental variables extracted from the study area in the Republic of Korea. The percentages indicate the amount of variation explained by each PC.
VariablePC1PC2PC3
Elevation−0.4560.4370.156
Slope−0.497−0.348−0.141
Aspect−0.0030.0240.262
Topographic position index−0.0460.255−0.339
Topographic roughness index−0.492−0.341−0.143
Normalized difference vegetation index−0.424−0.044−0.265
Annual mean temperature0.346−0.478−0.509
Annual precipitation0.0320.525−0.65
Eigenvalue3.0951.1511.069
Variance explained (%)38.6814.3813.36
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kim, Y.-G.; Nam, H.; Park, J.; Park, J.; Park, D. Distribution and Potential Dispersal Corridors of Two Onychodactylus Species in the Republic of Korea. Diversity 2026, 18, 57. https://doi.org/10.3390/d18010057

AMA Style

Kim Y-G, Nam H, Park J, Park J, Park D. Distribution and Potential Dispersal Corridors of Two Onychodactylus Species in the Republic of Korea. Diversity. 2026; 18(1):57. https://doi.org/10.3390/d18010057

Chicago/Turabian Style

Kim, Young-Guk, Hahyun Nam, Jaejin Park, Jiho Park, and Daesik Park. 2026. "Distribution and Potential Dispersal Corridors of Two Onychodactylus Species in the Republic of Korea" Diversity 18, no. 1: 57. https://doi.org/10.3390/d18010057

APA Style

Kim, Y.-G., Nam, H., Park, J., Park, J., & Park, D. (2026). Distribution and Potential Dispersal Corridors of Two Onychodactylus Species in the Republic of Korea. Diversity, 18(1), 57. https://doi.org/10.3390/d18010057

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop