Next Article in Journal
High-Throughput Sequencing-Based Assessment of Intestinal Parasitic Infections in Economically and Medicinally Valuable Captive Tokay Gecko (Gekko gecko) and Chinese Blue-Tailed Skink (Plestiodon chinensis)
Previous Article in Journal
In Vitro and In Vivo Evaluation of Autochthonous Probiotics and Their Effects on the Mucosal Health of Nile Tilapia (Oreochromis niloticus)
Previous Article in Special Issue
A New Species of Boulenophrys (Anura, Megophryidae) from Northern Jiangxi, China
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Animals
Volume 15
Issue 22
10.3390/ani15223297
animals-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

The Effects of Cold Tolerance on the Distribution of Two Extreme Altitude Lizard Species in the Qinghai–Tibetan Plateau

by
Xiaqiu Tao
1,*,
Yiyi Li
2 and
Jiasheng Li
3
1
Jiangsu Key Laboratory for Biodiversity and Biotechnology, College of Life Sciences, Nanjing Normal University, Nanjing 210023, China
2
Department of Computer Science, New Mexico State University, Las Cruces, NM 88003, USA
3
College of Life Sciences, China Jiliang University, Hangzhou 310018, China
*
Author to whom correspondence should be addressed.
Animals 2025, 15(22), 3297; https://doi.org/10.3390/ani15223297
Submission received: 17 October 2025 / Revised: 10 November 2025 / Accepted: 13 November 2025 / Published: 15 November 2025
(This article belongs to the Special Issue Evolution, Diversity, and Conservation of Herpetofauna: Second Edition)
Browse Figures
Versions Notes

Simple Summary

Predicting where species occur is crucial for their research, but traditional models often overestimate suitable areas because they rely only on general climate. This is especially problematic for animals in extreme environments like the Qinghai–Tibetan Plateau, where surviving the harsh winter is critical. Our study aimed to improve distribution maps for two extreme altitude lizard species by adding a key piece of physiological information: their ability to withstand cold stress during winter hibernation. We compared maps based on just climate data with new maps that also incorporated this cold-tolerance trait. Our results showed that the new, physiology-informed models were smaller and more accurate, removing areas where winters are too severe for the lizards to survive. This proves that including an animal’s physiological limits in models is essential for realistic predictions. Creating more accurate habitat maps is valuable because it helps conservationists focus their efforts on the areas that are truly vital for protecting unique species in a changing world.

Abstract

Species distribution models (SDMs) have been widely used to predict potentially suitable habitats for species. However, traditional SDMs have been criticized for ignoring the physiological processes by which species respond to their environment. Integrating physiological tolerance into the model is essential to improve the prediction accuracy of SDMs. Currently, this approach has not been applied in the study of Phrynocephalus erythrurus and Phrynocephalus theobaldi, which are part of the world’s highest reptiles and endemic to Qinghai–Tibetan Plateau. In this study, based on experiments, we found that the critical thermal minimum (CTmin) of the Phrynocephalus theobaldi was 0.85 °C. Further, we studied the effect of cold tolerance on the prediction of potential areas for these two reptile species. The high suitability area predicted by the SDMs incorporating cold tolerance data were 37.13% smaller than that predicted by the traditional SDMs. The difference between the two SDMs is primarily concentrated at the edges of the high suitability areas. The incorporation of cold tolerance data influenced the model’s predictions by effectively reducing the extent of edges of the high suitability areas. Our findings have theoretical significance for optimizing SDM predictions and provide a robust scientific basis for biodiversity conservation in the extreme-altitude ecosystems of the Qinghai–Tibetan Plateau.

1. Introduction

Accurately defining the geographic distribution of species is a fundamental goal in ecology and conservation, providing the basis for effective management and biodiversity assessment [1]. Species distribution models (SDMs) have become indispensable tools for this purpose, correlating species occurrence records with environmental variables to predict habitat suitability across landscapes [2].
However, traditional SDMs often ignore biological traits such as physiological limits and dispersal capacity, leading to uncertainty in predictions [3,4]. Physiological traits significantly influence species’ climate vulnerability [5]. These traits often vary with elevations to adapt to cold temperatures, low oxygen, and high solar radiation at high elevations [5,6]. High-elevation ectotherms, for example, must tolerate extreme cold [7], but prolonged winter conditions may limit their survival and reproduction [8,9]. Incorporating cold-stress adaptations into models is thus critical for predicting climate impacts on these species.
As ectotherms, reptiles are highly sensitive to ambient temperature because body temperature governs physiology, behavior, and population dynamics [10,11]. Tropical ectotherms are more vulnerable to warming than temperate ones [12,13]. High-elevation species may benefit from warming due to broader thermal tolerances [14], but upward migrations could intensify competition [15]. Many lizards face extinction risks from climate change [16]. For example, recent work on Bedriaga’s rock lizards reports warming-driven physiological disruption and heightened risk [17]. Yet high-elevation species remain understudied [18,19].
The Qinghai–Tibetan Plateau, the highest and harshest plateau globally [20], offers unique insights into high-elevation species’ responses to climate change. With an average elevation exceeding 4000 m, the extreme diurnal temperature shifts, low oxygen, and intense UV radiation have shaped specialized physiological and ecological adaptations in plateau-dwelling lizards. Two toad-headed lizards, Phrynocephalus theobaldi Blyth, 1863 (IUCN: Least Concern [21]) and Phrynocephalus erythrurus Zugmayer, 1909 (IUCN: Least Concern [22]), occurring on the Qinghai–Tibetan Plateau, provide an excellent model system for exploring these questions. These two lizards belong to the viviparous Phrynocephalus radiation on the Qinghai–Tibet Plateau. Phylogeographic studies place Phrynocephalus erythrurus into two Qiangtang lineages, northern and southern, shaped by the Tanggula and Nyainqentanglha ranges and regional drainage [23]. Distributed across a vast and environmentally challenging landscape, these two lizards have evolved specific adaptations to cope with extreme temperatures and hypoxia [6]. In this study, we focus on two species with differing altitudinal ranges, Phrynocephalus theobaldi and Phrynocephalus erythrurus. We aim to investigate how incorporating a key physiological trait—cold stress experienced during the overwintering period—improves predictions of their current suitable habitats.
We developed two sets of models for each species: a traditional SDM using only bioclimatic and topographic variables, and a physiology-informed SDM that integrates a “cold stress frequency” variable. We hypothesize that the physiology-informed model will provide a more constrained and ecologically realistic estimate of suitable habitat. Specifically, we predict that the traditional model will overestimate the species’ ranges by failing to exclude areas where severe winter conditions exceed their physiological tolerance for cold. This study seeks to provide a more accurate understanding of the current distributions of these extreme altitude specialists and highlight the importance of integrating physiology into conservation biogeography.

2. Materials and Methods

2.1. Data Acquisition

We compiled occurrence records of the two lizards from two sources: our own field observations (N = 62) and published literature (N = 34). In total, 96 raw records were collected. To reduce spatial autocorrelation, we applied the ENMTools package (v 1.1.5) [24] in R (v 4.3.1) [25] to filter records by retaining only one presence within each grid cell (30 arc-seconds, equal to 1 km), thereby ensuring spatial independence of the occurrence data. After spatial thinning, 86 occurrence points (Phrynocephalus theobaldi, 59; Phrynocephalus erythrurus, 27) were retained for subsequent analyses (Figure 1).

2.2. Predicting the CTmin and Cold Stress Frequency for Phrynocephalus

2.2.1. Measuring the CTmin in Phrynocephalus theobaldi

The Phrynocephalus theobaldi used in the experiment were collected in July 2020 from Gar County (elevation: 4316 m.a.s.l.) and Zhongba County (elevation: 4561 m.a.s.l.) in Tibet. All individuals were adults with a snout-vent length greater than 40 mm.
The CTmin values were determined experimentally by placing individual lizards in a temperature-controlled chamber, where the temperature was lowered from 10 °C [26] at a rate of approximately 0.5 °C per minute. The CTmin was recorded as the temperature at which a lizard lost its righting response and could no longer perform voluntary movements (e.g., blinking, crawling), with the critical criterion that the animal fully recovered after being returned to ambient temperatures.

2.2.2. Cold Stress Frequency During the Overwintering Period

We downloaded a dataset of 3-hourly soil temperatures for the year 2020 from the China Meteorological Data Service Center (https://data.cma.cn/, accessed on 10 October 2025). The dataset included measurements from 1950 national-level stations across the Qinghai–Tibet Plateau region of China, with a spatial resolution of 0.5°. We focused on soil temperatures at depths of 1–40 cm, as field investigations and previous literature indicate that most Phrynocephalus lizard nests on the Qinghai–Tibet Plateau are located at depths of approximately 30–50 cm [27,28]. Stations with incomplete data (i.e., missing temperature readings for any 3 h interval) were removed from the analysis. For each remaining station, soil temperature data were cropped to the overwintering period of the two lizard species, defined as 1 November to 31 March.
We defined a “cold stress event” as any 3 h interval in which the mean soil temperature (at 10–40 cm depth) fell below the critical thermal minimum (CTmin). Besides Phrynocephalus theobaldi, the CTmin values for Phrynocephalus erythrurus used were −1.58 °C [29]. Correspondingly, a “cold stress frequency” was defined as a single day in which at least one cold stress event occurred. Subsequently, for each location, we calculated the “cold stress frequency” as the total number of cold stress days during the overwintering period (from 00:00 on 1 November to 23:00 on 31 March). This follows the event–day framework of Sun et al. [30] that was developed for heat stress. This variable was then used as a physiological predictor in our species distribution models.

2.3. Traditional Environmental Variables

We selected a suite of bioclimatic and topographic variables to model the habitat suitability for the two lizard species. 19 variables (version 2.1) were downloaded from the WorldClim database (www.worldclim.org, accessed on 10 October 2025) at a 30-arc-second (~1 km) resolution.
Topographic variables, including elevation, slope, and aspect, were derived from a 1 km resolution digital elevation model (DEM) obtained from OpenTopography (https://www.opentopography.org, accessed on 10 October 2025) to capture terrain-induced heterogeneity typical of high-mountain systems.
To avoid overfitting caused by multicollinearity, we conducted pairwise Spearman correlation analysis among all variables. When two variables were highly correlated (|r| ≥ 0.8), the variable with the higher contribution rate was selected. After this screening process, the predictor variables used for model construction are listed in Table 1.

2.4. Species Distribution Modelling

We predicted the potential distribution of lizards using BIOMOD2 (v 4.2-6-2) package [31] in R (v 4.3.1). Based on the number of occurrence records, 10 sets of 100 random pseudo-absence points were generated, which is considered sufficient for most machine-learning algorithms except for GLM and GAM [32]. 8 modelling algorithms were applied: classification tree analysis (CTA), generalized additive models (GAM), generalized boosted regression trees (GBM), generalized linear models (GLM), multivariate adaptive regression splines (MARS), maximum entropy (MaxEnt), random forests (RF), and extreme gradient boosting (XGBoost).
For model calibration, 80% of the occurrence data were used for training and 20% for testing. Cross-validation was repeated 10 times, resulting in a total of 1000 model runs. Model performance was assessed using three evaluation metrics: the area under the curve (AUC), the true skill statistic (TSS), and Cohen’s kappa. Only models that met the thresholds of TSS ≥ 0.8, AUC ≥ 0.7, and kappa ≥ 0.6 were retained for building the final ensemble projections [33,34].

3. Results

3.1. CTmin in Phrynocephalus theobaldi

The CTmin values used were 0.85 °C for Phrynocephalus theobaldi. There was no significant difference in the critical thermal minimum (CTmin) of Phrynocephalus theobaldi between the two sampled populations (F1,22 = 1.816, p = 0.192) (Table 2). The variation in CTmin among individuals was also minimal, with an overall range of 0.7 °C (Table 2).

3.2. Model Performance and Evaluation

The 8 individual species distribution models constructed using the Biomod2 platform showed variable performance across 10 replicate runs. Evaluation metrics for these individual models ranged from 0.530 to 0.763 for KAPPA, 0.765 to 0.969 for AUC and 0.530 to 0.804 for TSS (Table 3).
By combining the outputs of the best-performing algorithms, the final ensemble models (EM) demonstrated excellent predictive power. In the traditional SDMs, the ensemble model for Phrynocephalus theobaldi achieved Kappa = 0.673, AUC = 0.978, and TSS = 0.877, whereas Phrynocephalus erythrurus reached Kappa = 0.706, AUC = 0.992, and TSS = 0.944. After integrating physiological predictors, TSS increased in both models. For Phrynocephalus theobaldi, ensemble model increased to Kappa = 0.705, AUC = 0.982, and TSS = 0.904. For Phrynocephalus erythrurus, the integrated ensemble yielded Kappa = 0.674, AUC = 0.989, and TSS = 0.965, indicating a high level of accuracy in predicting the distributions of both lizard species (Table 3). The final ensemble model for Phrynocephalus theobaldi integrated 300 individual models under the traditional framework and 420 models when incorporating CTmin data. For Phrynocephalus erythrurus, the ensemble model was composed of 323 traditional models and 342 physiology-informed models (Table 4).

3.3. Main Environmental Variables Among Different SDMs

In the traditional SDMs (without CTmin), the distribution of Phrynocephalus theobaldi was most influenced by six variables, each contributing over 10% to the final ensemble model: Bio12, Bio6, Bio19, Bio2, Bio15, Bio5, while Bio7 and Bio3 contributed over 5% (Figure 2a). For Phrynocephalus erythrurus, Bio15, Bio12 and Bio4 contributed more than 10%, while Bio11 and Bio5 contributed more than 5% (Figure 2b).
After incorporating the CTmin data, the set of influential predictors shifted. For the physiology-informed model of Phrynocephalus theobaldi, variables contributing over 10% became Bio13, Bio9, Bio5, Bio7, Bio2, Bio15, while Bio19 contributed over 5% (Figure 2c). For Phrynocephalus erythrurus, the model was simplified, with Bio15 and Bio12 remaining as the two variables that contributed more than 10% (Figure 2d).
In the traditional SDMs, the predicted suitability for Phrynocephalus theobaldi showed a decreasing trend with increasing Annual Precipitation (Bio12), but an increasing trend with higher Precipitation of the Coldest Quarter (Bio19) and Precipitation Seasonality (Bio15). Regarding temperature, suitability rose sharply once the Minimum Temperature of the Coldest Month (Bio6) increased above −20 °C, after which it gently declined. The optimal habitat corresponded to a Max Temperature of the Warmest Month (Bio5) of approximately 18–20 °C. The species also showed a gentle positive response to a larger Mean Diurnal Range (Bio2) (Figure S1).
For Phrynocephalus erythrurus, Precipitation Seasonality (Bio15) was the dominant factor, with suitability exhibiting a sudden jump when Bio15 approached a value of 110. Suitability also showed a positive trend with increasing Temperature Seasonality (Bio4), while the response to Annual Precipitation (Bio12) was relatively flat (Figure S2).
After incorporating CTmin data, both the key environmental factors and their response curves shifted. For Phrynocephalus theobaldi, suitability decreased with increasing Precipitation of the Wettest Month (Bio13). A significant increase in suitability was observed when the Mean Temperature of the Driest Quarter (Bio9) rose above −10 °C, after which the trend flattened. The response to Temperature Annual Range (Bio7) was complex, showing a slight dip around 30 °C before a sharp decline above 40 °C. For Mean Diurnal Range (Bio2), suitability was stable before rising markedly above a value of 12.5. The trends for Bio5 and Bio15 were consistent with the traditional model: suitability still peaked at a Max Temperature of the Warmest Month (Bio5) around 20 °C and increased with higher Precipitation Seasonality (Bio15) (Figure S3).
For Phrynocephalus erythrurus, the strong threshold for Precipitation Seasonality (Bio15) remained the primary factor, with Annual Precipitation (Bio12) as the secondary factor, although the peak suitability (Figure S4).

3.4. Effects of Incorporating CTmin on the Prediction of Suitable Habitat Areas

The prediction results of the two SDMs showed that the high suitability areas were mainly concentrated in South China, Central China, and East China, with a few in North and Southwest China, and almost no invasion risk in most of the northeast and northwest provinces of China (Figure 3). Although both SDMs performed well, the traditional model for Phrynocephalus erythrurus predicted a larger high suitability area (146,603 km2, Figure 3a) compared to the model incorporating CTmin data (106,906 km2, Figure 3b). Furthermore, the SDMs incorporating physiological data predicted a significantly smaller area (4865.97 km2, Figure 3a) of overlapping habitat between the two Phrynocephalus species, amounting to only 29.70% of the overlap projected by the traditional models (163,82.6 km2, Figure 3b). The two models show a difference of approximately 37.1% in high suitability areas for Phrynocephalus erythrurus, primarily concentrated at the edges of the high suitability areas, and the physiology-informed model projected a distribution that expanded further southward, rather than northward, and revealed several small additional patches that met the winter constraints (Figure 3 and Figure 4). These findings indicate that the incorporation of embryo temperature tolerance data influenced the model’s predictions by reducing the extent of edges within these high suitability areas.

4. Discussion

4.1. Comparison of CTmin Between Phrynocephalus theobaldi and Other Lizards in Qinghai–Tibetan Plateau

The critical thermal minimum (CTmin) of an ectotherm is influenced by numerous factors, including developmental stage, geographical origin, and nutritional status [35]. In lizards, species inhabiting higher altitudes typically exhibit lower CTmin values. For example, Phrynocephalus erythrurus, found at the highest elevations, has a CTmin of approximately −1.58 °C [29], whereas Phrynocephalus vlangalii, which lives at relatively lower elevations, has a higher CTmin of 0.9 °C [36]. And the CTmin of the Eremias argus is approximately 1 °C [37], whereas the Takydromus sexlineatus and Takydromus septentrionalis, which live in lower latitude regions, is approximately 6 °C [38,39]. These interspecific differences likely reflect adaptations to the distinct thermal niches these species occupy. Our results showed that Phrynocephalus theobaldi possesses a CTmin intermediate to its high- and low-altitude congeners is consistent with the unique thermal environment it inhabits. This species is primarily distributed across open habitats in the southern Qinghai–Tibet Plateau, a region characterized by low annual mean temperatures (−5.1–8.3 °C) and a large diurnal temperature range (12–16 °C) [40]. The strong cold tolerance exhibited by Phrynocephalus theobaldi appears to be a key adaptation to these demanding thermal conditions.
Furthermore, the intermediate CTmin of Phrynocephalus theobaldi, which falls between that of higher- and lower-altitude congeners, indicates that tolerance is broader than in low-elevation specialists but narrower than in high-elevation specialists. This pattern implies a potential for localized niche and spatial overlap where suitable climatic and microhabitat conditions coincide, as has been noted in other lizard systems along elevational gradients [41]. Although our study did not directly test these hypotheses, the observed physiological variation underscores the role of cold tolerance in structuring species distributions on the Plateau.

4.2. Model Limitations and Behavioral Buffering

It is important to acknowledge the limitations of our modeling approach. Our predictions are based on soil temperature data at a 1 km resolution, which serves as a proxy for the thermal conditions within the lizards’ overwintering sites (hibernacula). However, this does not capture the microclimatic variations that individual animals experience. Ectotherms are known to engage in behavioral thermoregulation, and these lizards may actively select hibernacula with more favorable thermal profiles than the surrounding environment, a concept known as “behavioral buffering” [30]. For instance, lizards may choose deeper burrows or specific aspects of slopes that are better insulated from extreme cold. Studies on other lizard species have shown that females select specific nest sites to optimize thermal conditions for embryonic development (e.g., Phrynocephalus przewalskii [42], Physignathus lesueurii [43]). Previous studies indicate that repeated cold extremes or unusually long winters can overwhelm buffering capacities, leading to local extinctions despite apparently favorable microhabitat use [44].

4.3. Effects of Traditional Environmental Factors

In the traditional SDMs, three environmental factors, Bio2 (mean diurnal range), Bio3 (isothermality), and Bio15 (precipitation seasonality) were consistently retained across all four model combinations for both species. Notably, Bio15 contributed >10% in every ensemble model, indicating a strong signal of precipitation variability shaping habitat suitability on the Qinghai–Tibetan Plateau. Inspection of the response curves shows a marked increase in suitability with higher precipitation seasonality, implying that both Phrynocephalus theobaldi and Phrynocephalus erythrurus tend to occur in areas where rainfall variability is large. This pattern is ecologically plausible on the Qinghai–Tibetan Plateau, where the juxtaposition of Indian Summer Monsoon and mid-latitude westerlies produces pronounced seasonal and spatial gradients in precipitation and its variability [45]. In such environments, coarse sandy-gravel substrates and open steppe-desert vegetation co-vary with seasonal water availability, likely mediating prey and burrow microhabitats required by toad-headed lizards. Together, these results suggest that precipitation seasonality acts as a first-order filter for the two species, with thermal predictors (Bio2 and Bio3) refining the niche at finer scales.
Although elevation did not emerge as the main environmental variable across all sub-models after collinearity screening, recent SDM applications in mountainous fauna show that topography can substantially shape suitability patterns, fragmentation, and centroid shifts [46,47]. In parallel, taxonomic and phylogeographic work on Phrynocephalus lizards in Qinghai–Tibet Plateau indicates that the Plateau’s extreme topographic heterogeneity has promoted genetic differentiation and lineage/species formation [23,48,49]. Taken together, our study suggested that “elevation effects” on the both Phrynocephalus theobaldi and Phrynocephalus erythrurus are primarily mediated through temperature-related environments, as terrain can greatly influence factors such as temperature, hydrology, air pressure and solar radiation [50,51].

4.4. Effects of Incorporating Temperature Tolerance Data on the SDMs

Our study demonstrates that incorporating physiological data into SDMs significantly refines predictions of habitat suitability. A key finding is that the physiology-informed SDMs, which included CTmin data during overwintering, projected a smaller area of suitable habitat for Phrynocephalus erythrurus and distributional overlap between two lizards compared to the traditional climate-only model, which more accurately reflects the allopatric distribution of these two species observed in the field [27]. This better-predicting result aligns with a growing body of research suggesting that traditional SDMs often overestimate potential distributions because they do not account for critical physiological thresholds that can limit a species’ survival, even when climatic conditions appear favorable [52,53]. By failing to consider the lethal and sub-lethal effects of extreme cold events on overwintering lizards, traditional models may incorrectly identify large areas as suitable habitat. These eastern patches represent potential suitability rather than confirmed occurrences. They may reflect local relaxation of winter constraints, but they could also arise from sampling gaps or dispersal/historical barriers; moreover, year-round persistence may be limited by summer heat or aridity. We therefore recommend targeted surveys to evaluate these sites.
Similar findings have been reported in other taxa. For example, Laeseke et al. found that the distribution of Capreolia implexa was restricted by geographical barriers, causing climate-only SDMs to underestimate suitable areas [54]. Conversely, invasive species such as the red-eared slider (Trachemys scripta elegans) can disperse widely due to human activities, and climate-only SDMs often overestimate their potential range if they ignore physiological constraints [55]. When embryonic thermal limits were incorporated, unsuitable edge regions were excluded and high-risk areas were more realistically restricted to warmer provinces in southern and central China [56,57]. Likewise, Gamliel et al. reported that physiology-informed SDMs of marine organisms performed better than purely climatic models, especially at the edges of species’ ranges [4]. For intertidal crabs, adding experimentally determined thermal tolerance data reduced overly optimistic range predictions and provided a more conservative assessment of suitable habitats [58]. Collectively, these cases demonstrate that ignoring physiological bottlenecks can lead to both underestimation and overestimation of distribution ranges depending on species and context.
By integrating overwintering cold tolerance, our model effectively distinguishes between areas remain suitable through the overwintering period. For instance, a traditional model might identify a region as suitable based on warm summer temperatures, but our physiology-informed approach correctly excludes it if winter soil temperatures fall below the species’ lethal CTmin. This refinement is particularly evident at the edges of the predicted distribution, providing a more ecologically realistic depiction of the species’ fundamental niche. More broadly, for high-mountain species the reliability of projections benefits from incorporating key, biology-linked and terrain-linked determinants (such as cold tolerance, preferred body temperature, thermal performance, stand metabolic rate and reproductive trait) [59,60,61], thereby reducing overprediction in climatically complex landscapes. This underscores the necessity of integrating key biological information to produce more reliable predictions [3,30].

5. Conclusions

Our study indicated that the embryo temperature tolerance data are important for the prediction of SDMs. We found that traditional species distribution models (SDMs) based on climatic and topographic variables overestimated the suitable habitat for both Phrynocephalus theobaldi and Phrynocephalus erythrurus, critically predicting a larger area of potential habitat overlap. By integrating CTmin data, the physiology-informed models corrected these inaccuracies, projecting smaller and more ecologically plausible habitats that align with the species’ known allopatric distributions. This research highlights that while broad climatic conditions may appear suitable, extreme physiological challenges, such as lethal winter temperatures, are a key limiting factor defining the true boundaries of species’ niches in extreme environments. In conclusion, our research provides a scientific basis for the conservation of these endemic species of the Qinghai–Tibetan Plateau. By providing more accurate habitat maps, our work can help guide targeted conservation strategies.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ani15223297/s1, Figure S1. Response curves of environmental variables in traditional SDMs for Phrynocephalus theobaldi. Variables are grouped by contribution to the model: (a) > 10 %; (b) ≥ 5 %; (c) < 5 %; Figure S2. Response curves of environmental variables in traditional SDMs for Phrynocephalus erythrurus. Variables are grouped by contribution to the model: (a) > 10 %; (b) ≥ 5 %; (c) < 5 %; Figure S3. Response curves of environmental variables in SDMs incorporating CTmin data for Phrynocephalus theobaldi. Variables are grouped by contribution to the model: (a) > 10 %; (b) ≥ 5 %; (c) < 5 %. Figure S4. Response curves of environmental variables in SDMs incorporating CTmin data for Phrynocephalus erythrurus. Variables are grouped by contribution to the model: (a) > 10 %; (b) < 5 %.

Author Contributions

Conceptualization, X.T.; methodology, X.T. and Y.L.; software, X.T. and Y.L.; validation, X.T., Y.L. and J.L.; formal analysis, X.T., and Y.L.; investigation, X.T., Y.L. and J.L.; resources, X.T.; data curation, X.T., Y.L. and J.L.; writing—original draft preparation, X.T.; writing—review and editing, X.T. and Y.L.; visualization, X.T. and Y.L.; supervision, X.T.; project administration, X.T.; All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The work was carried out in compliance with laws on animal welfare and research in China, and approved by the Animal Research Ethical Committees of Nanjing Normal University (Approval number: IACUC-20200511, approval date: 26 June 2020).

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

We thank anonymous reviewers for their helpful comments.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Travis, J. Climate change and habitat destruction: A deadly anthropogenic cocktail. Proc. R. Soc. B 2003, 270, 467–473. [Google Scholar] [CrossRef]
  2. Elith, J.; Graham, C.H. Do they? How do they? WHY do they differ? On finding reasons for differing performances of species distribution models. Ecography 2009, 32, 66–77. [Google Scholar] [CrossRef]
  3. Rodríguez, L.; García, J.J.; Carreño, F.; Martínez, B. Integration of physiological knowledge into hybrid species distribution modelling to improve forecast of distributional shifts of tropical corals. Divers. Distrib. 2019, 25, 715–728. [Google Scholar] [CrossRef]
  4. Gamliel, I.; Buba, Y.; Guy-Haim, T.; Garval, T.; Willette, D.; Rilov, G.; Belmaker, J. Incorporating physiology into species distribution models moderates the projected impact of warming on selected Mediterranean marine species. Ecography 2020, 43, 1090–1106. [Google Scholar] [CrossRef]
  5. Sunday, J.M.; Bates, A.E.; Kearney, M.R.; Colwell, R.K.; Dulvy, N.K.; Longino, J.T.; Huey, R.B. Thermal-safety margins and the necessity of thermoregulatory behavior across latitude and elevation. Proc. Natl. Acad. Sci. USA 2014, 111, 5610–5615. [Google Scholar] [CrossRef]
  6. Gangloff, E.J.; Telemeco, R.S. High temperature, oxygen, and performance: Insights from reptiles and amphibians. Integr. Comp. Biol. 2018, 58, 9–24. [Google Scholar] [CrossRef] [PubMed]
  7. Niu, Z.; Li, M.; Pu, P.; Wang, H.; Zhang, T.; Tang, X.; Chen, Q. Effects of temperature on the locomotor performance and contraction properties of skeletal muscle from two Phrynocephalus lizards at high and low altitude. J. Comp. Physiol. B 2021, 191, 907–916. [Google Scholar] [CrossRef]
  8. Irwin, J.T.; Lee, R.E., Jr. Cold winter microenvironments conserve energy and improve overwintering survival and potential fecundity of the goldenrod gall fly, Eurosta solidaginis. Oikos 2003, 100, 71–78. [Google Scholar] [CrossRef]
  9. Hahn, D.A.; Denlinger, D.L. Energetics of insect diapause. Annu. Rev. Entomol. 2011, 56, 103–121. [Google Scholar] [CrossRef]
  10. Du, W.G.; Shine, R. The behavioural and physiological strategies of bird and reptile embryos in response to unpredictable variation in nest temperature. Biol. Rev. 2015, 90, 19–30. [Google Scholar] [CrossRef] [PubMed]
  11. Noble, D.W.; Stenhouse, V.; Schwanz, L.E. Developmental temperatures and phenotypic plasticity in reptiles: A systematic review and meta-analysis. Biol. Rev. 2018, 93, 72–97. [Google Scholar] [CrossRef] [PubMed]
  12. Logan, M.L.; Huynh, R.K.; Precious, R.A.; Calsbeek, R.G. The impact of climate change measured at relevant spatial scales: New hope for tropical lizards. Glob. Change Biol. 2013, 19, 3093–3102. [Google Scholar] [CrossRef] [PubMed]
  13. Kubisch, E.L.; Fernández, J.B.; Ibargüengoytía, N.R. Vulnerability to climate warming of Liolaemus pictus (Squamata, Liolaemidae), a lizard from the cold temperate climate in Patagonia, Argentina. J. Comp. Physiol. B 2016, 186, 243–253. [Google Scholar] [CrossRef] [PubMed]
  14. Valdecantos, S.; Martínez, V.; Lobo, F.; Cruz, F.B. Thermal biology of Liolaemus lizards from the high Andes: Being efficient despite adversity. J. Therm. Biol. 2013, 38, 126–134. [Google Scholar] [CrossRef]
  15. Freeman, B.G.; Tobias, J.A.; Schluter, D. Behavior influences range limits and patterns of coexistence across an elevational gradient in tropical birds. Ecography 2019, 42, 1832–1840. [Google Scholar] [CrossRef]
  16. Sinervo, B.; Mendez-De-La-Cruz, F.; Miles, D.B.; Heulin, B.; Bastiaans, E.; Villagrán-Santa Cruz, M.; Lara-Resendiz, R.; Martínez-Méndez, N.; Calderón-Espinosa, M.L.; Meza-Lázaro, R.N.; et al. Erosion of lizard diversity by climate change and altered thermal niches. Science 2010, 328, 894–899. [Google Scholar] [CrossRef]
  17. Bombi, P.; Calò, F.; Salvi, D. Living Beyond the Edge: Impacts of Climate Change on Rock Lizards at the Niche Margin. Anim. Conserv. 2025, 28, 699–710. [Google Scholar] [CrossRef]
  18. Ma, L.; Buckley, L.B.; Huey, R.B.; Du, W. A global test of the cold-climate hypothesis for the evolution of viviparity of squamate reptiles. Glob. Ecol. Biogeogr. 2018, 27, 679–689. [Google Scholar] [CrossRef]
  19. Moore, D.; Stow, A.; Kearney, M.R. Under the weather?—The direct effects of climate warming on a threatened desert lizard are mediated by their activity phase and burrow system. J. Anim. Ecol. 2018, 87, 660–671. [Google Scholar] [CrossRef]
  20. Yan, C.; Song, M.; Jiang, D.; Ren, J.; Lv, Y.; Chang, J.; Huang, S.; Zaher, H.; Li, J. Genomic evidence reveals intraspecific divergence of the hot-spring snake (Thermophis baileyi), an endangered reptile endemic to the Qinghai-Tibet plateau. Mol. Ecol. 2023, 32, 1335–1350. [Google Scholar] [CrossRef]
  21. IUCN. Phrynocephalus theobaldi. Available online: https://www.iucnredlist.org/species/170415/123299193 (accessed on 10 October 2025).
  22. IUCN. Phrynocephalus erythrurus. Available online: https://www.iucnredlist.org/species/107660333/107660347 (accessed on 10 October 2025).
  23. Jin, Y.; Liu, N. Phylogeography of Phrynocephalus erythrurus from the Qiangtang Plateau of the Tibetan Plateau. Mol. Phylogenet. Evol. 2010, 54, 933–940. [Google Scholar] [CrossRef]
  24. Warren, D.; Dinnage, R. ENMTools: Analysis of Niche Evolution Using Niche and Distribution Models. R Package Version 1.1.5. Available online: https://CRAN.R-project.org/package=ENMTools (accessed on 10 October 2025).
  25. R Core Team, R. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2023; Available online: https://www.R-project.org/ (accessed on 10 October 2025).
  26. Dayananda, B.; Murray, B.R.; Webb, J.K. Hotter nests produce hatchling lizards with lower thermal tolerance. J. Exp. Biol. 2017, 220, 2159–2165. [Google Scholar] [CrossRef]
  27. Zhao, E.; Zhao, K.; Zhou, K. Fauna Sinica, Reptilia, Volume 2, Squamata, Lacertilia; Science Press: Beijing, China, 1999. [Google Scholar]
  28. Wang, Z.; Ma, L.; Shao, M.; Ji, X. Are viviparous lizards more vulnerable to climate warming because they have evolved reduced body temperature and heat tolerance? Oecologia 2017, 185, 573–582. [Google Scholar] [CrossRef]
  29. Li, X.; Wang, Y.; Li, M.; Men, S.; Pu, P.; Tang, X.; Chen, Q. Comparison of cold hardiness of two toad-headed lizards from different altitudes. Sichuan J. Zool. 2017, 36, 300–305. [Google Scholar]
  30. Sun, B.; Ma, L.; Wang, Y.; Mi, C.; Buckley, L.B.; Levy, O.; Lu, H.; Du, W. Latitudinal embryonic thermal tolerance and plasticity shape the vulnerability of oviparous species to climate change. Ecol. Monogr. 2021, 91, e01468. [Google Scholar] [CrossRef]
  31. Thuiller, W.; Georges, D.; Gueguen, M.; Engler, R.; Breiner, F.; Lafourcade, B.; Patin, R.; Blancheteau, H. biomod2: Ensemble Platform for Species Distribution Modeling. R Package Version 4.2-6-2. Available online: https://CRAN.R-project.org/package=biomod2 (accessed on 10 October 2025).
  32. Barbet-Massin, M.; Jiguet, F.; Albert, C.H.; Thuiller, W. Selecting pseudo-absences for species distribution models: How, where and how many? Methods Ecol. Evol. 2012, 3, 327–338. [Google Scholar] [CrossRef]
  33. Fielding, A.H.; Bell, J.F. A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ. Conserv. 1997, 24, 38–49. [Google Scholar] [CrossRef]
  34. Allouche, O.; Tsoar, A.; Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 2006, 43, 1223–1232. [Google Scholar] [CrossRef]
  35. Hutchison, V.H. Factors influencing thermal tolerances of individual organisms. In Thermal Ecology, Vol. II; Esch, G.W., McFarlane, R.M., Eds.; National Technical Information Service; Springfield: Augusta, GA, USA, 1976; pp. 10–26. [Google Scholar]
  36. Shu, L.; Zhang, Q.; Qu, Y.; Ji, X. Thermal tolerance, selected body temperature and thermal dependence of food assimilation and locomotor performance in the Qinghai toad headed lizard, Phrynocephalus vlangalii. Acta Ecol. Sin. 2010, 30, 2036–2042. [Google Scholar]
  37. Luo, L.; Qu, Y.; Ji, X. Thermal dependence of food assimilation and sprint speed in a lacertid lizard (Eremias argus) from northern China. Acta Zool. Sin. 2006, 52, 256–262. [Google Scholar]
  38. Zhang, Y.; Ji, X. The thermal dependence of food assimilation and locomotor performance in southern grass lizards, Takydromus sexlineatus (Lacertidae). J. Therm. Biol. 2004, 29, 45–53. [Google Scholar] [CrossRef]
  39. Ji, X.; Du, W.; Sun, P. Body temperature, thermal tolerance and influence of temperature on sprint speed and food assimilation in adult grass lizards, Takydromus septentrionalis. J. Therm. Biol. 1996, 21, 155–161. [Google Scholar] [CrossRef]
  40. Jia, X.; Lee, H.F.; Cui, M.; Liu, C.; Zeng, L.; Yue, R.P.; Zhao, Y.; Lu, H. Habitat variability and ethnic diversity in Northern Tibetan Plateau. Sci. Rep. 2017, 7, 918. [Google Scholar] [CrossRef]
  41. Díaz de la Vega-Pérez, A.H.; Barrios-Montiel, R.; Jiménez-Arcos, V.H.; Bautista, A.; Bastiaans, E. High-mountain altitudinal gradient influences thermal ecology of the Mesquite Lizard (Sceloporus grammicus). Can. J. Zool. 2019, 97, 659–668. [Google Scholar] [CrossRef]
  42. Li, S.; Hao, X.; Wang, Y.; Sun, B.; Bi, J.; Zhang, Y.; Janzen, F.J.; Du, W. Female lizards choose warm, moist nests that improve embryonic survivorship and offspring fitness. Funct. Ecol. 2018, 32, 416–423. [Google Scholar] [CrossRef]
  43. Doody, J.S.; Guarino, E.; Georges, A.; Corey, B.; Murray, G.; Ewert, M. Nest site choice compensates for climate effects on sex ratios in a lizard with environmental sex determination. Evol. Ecol. 2006, 20, 307–330. [Google Scholar] [CrossRef]
  44. Üveges, B.; Mahr, K.; Szederkényi, M.; Bókony, V.; Hoi, H.; Hettyey, A. Experimental evidence for beneficial effects of projected climate change on hibernating amphibians. Sci. Rep. 2016, 6, 26754. [Google Scholar] [CrossRef] [PubMed]
  45. Maussion, F.; Scherer, D.; Mölg, T.; Collier, E.; Curio, J.; Finkelnburg, R. Precipitation seasonality and variability over the Tibetan Plateau as resolved by the High Asia Reanalysis. J. Clim. 2014, 27, 1910–1927. [Google Scholar] [CrossRef]
  46. Chen, K.; Ma, L.; Jiang, W.; Wang, L.; Wei, L.; Zhang, H.; Yang, R. Anthropogenic Disturbance and Climate Change Impacts on the Suitable Habitat of Sphenomorphus incognitus in China. Ecol. Evol. 2025, 15, e70848. [Google Scholar] [CrossRef]
  47. Liu, H.; Liu, Q.; Cui, X.; Peng, J.; Zhou, S.; Wang, F.; Zhong, L.; Wang, X.; Zheng, H.; Yang, C.; et al. Prediction of Potential Suitable Habitats for Elaphodus cephalophus in China Under Climate Change Scenarios. Ecol. Evol. 2025, 15, e72194. [Google Scholar] [CrossRef]
  48. Chen, Y.; Zhu, L.; Wu, Q.; Hu, C.; Qu, Y.; Ji, X. Geological and climatic influences on population differentiation of the Phrynocephalus vlangalii species complex (Sauria: Agamidae) in the northern Qinghai-Tibet Plateau. Mol. Phylogenet. Evol. 2022, 169, 107394. [Google Scholar] [CrossRef]
  49. Xu, R.; Song, Q.; Chen, D.; Guo, X. Lineage Diversification and Population Dynamics of the Qinghai Toad-Headed Agama (Phrynocephalus vlangalii) on the Qinghai-Tibet Plateau, with Particular Attention to the Northern Slope of the Kunlun-Arjin Mountains. Animals 2025, 15, 400. [Google Scholar] [CrossRef]
  50. Moore, I.D.; Grayson, R.B.; Ladson, A.R. Digital terrain modelling: A review of hydrological, geomorphological, and biological applications. Hydrol. Process. 1991, 5, 3–30. [Google Scholar] [CrossRef]
  51. Stanton, J.C.; Pearson, R.G.; Horning, N.; Ersts, P.; Reşit Akçakaya, H. Combining static and dynamic variables in species distribution models under climate change. Methods Ecol. Evol. 2012, 3, 349–357. [Google Scholar] [CrossRef]
  52. Byrne, M.; Gall, M.; Wolfe, K.; Agüera, A. From pole to pole: The potential for the Arctic seastar Asterias amurensis to invade a warming Southern Ocean. Glob. Change Biol. 2016, 22, 3874–3887. [Google Scholar] [CrossRef] [PubMed]
  53. Coulin, C.; de la Vega, G.J.; Chifflet, L.; Calcaterra, L.A.; Schilman, P.E. Linking thermo-tolerances of the highly invasive ant, Wasmannia auropunctata, to its current and potential distribution. Biol. Invasions 2019, 21, 3491–3504. [Google Scholar] [CrossRef]
  54. Laeseke, P.; Martínez, B.; Mansilla, A.; Bischof, K. Future range dynamics of the red alga Capreolia implexa in native and invaded regions: Contrasting predictions from species distribution models versus physiological knowledge. Biol. Invasions 2020, 22, 1339–1352. [Google Scholar] [CrossRef]
  55. Gong, S.; Gao, Y.; Duan, H.; Ge, Y.; Wei, Y. Incorporating physiological data into species distribution models to predict the potential distribution range of the red-eared slider in China. Ecol. Indic. 2023, 154, 110749. [Google Scholar] [CrossRef]
  56. Gong, S.; Yang, J.; Ge, Y.; Gaillard, D. Extent and Mechanisms of the Increasing Geographic Distribution of the Alien Red-Eared Slider (Trachemys scripta elegans) in China. Chin. J. Wildl. 2018, 39, 373–378. [Google Scholar]
  57. Zhao, M.; Cao, G.; Cao, S.; Liu, F.; Yuan, J.; Zhang, Z.; Diao, E.; Fu, J. Spatial-temporal variation characteristics of land surface temperature in Qinghai Province from 1980 to 2017. Arid. Zone. Res. 2021, 38, 178–187. [Google Scholar]
  58. Cong, J.; Tao, Z.; Xu, Y.; Li, X. From thermal tolerance to distribution resilience: Physiology-informed species distribution models moderated range shifts for an intertidal crab species under climate change. Glob. Ecol. Conserv. 2025, 62, e03809. [Google Scholar] [CrossRef]
  59. Jiang, Z.; Ma, L.; Mi, C.; Du, W. Effects of hypoxia on the thermal physiology of a high-elevation lizard: Implications for upslope-shifting species. Biol. Lett. 2021, 17, 20200873. [Google Scholar] [CrossRef] [PubMed]
  60. Tao, X.; Xie, F.; Zhu, L.; Wu, Q. Correlation of Morphology and Metabolism of Reproductive Traits in the Genus Phrynocephalus Around the Qinghai-Tibetan Plateau. Ecol. Evol. 2025, 15, e72029. [Google Scholar] [CrossRef] [PubMed]
  61. Zhu, Z.; Du, W.; Zhang, C.; Yu, W.; Zhao, X.; Liu, Z.; Zeng, Z. Altitudinal variation in thermal vulnerability of Qinghai-Tibetan Plateau lizards under climate warming. Curr. Zool. 2025, 71, 99–108. [Google Scholar] [CrossRef] [PubMed]
Animals 15 03297 g001
Figure 1. Presence data of two Phrynocephalus lizards used for SDMs generation, including CTmin sampling sites for Phrynocephalus theobaldi.
Figure 1. Presence data of two Phrynocephalus lizards used for SDMs generation, including CTmin sampling sites for Phrynocephalus theobaldi.
Animals 15 03297 g001
Animals 15 03297 g002
Figure 2. Mean contribution rates of sixteen environmental variables to the final ensemble traditional SDMs without incorporating CTmin data (a) Phrynocephalus theobaldi and (b) Phrynocephalus erythrurus, and with incorporating CTmin data (c) Phrynocephalus theobaldi and (d) Phrynocephalus erythrurus.
Figure 2. Mean contribution rates of sixteen environmental variables to the final ensemble traditional SDMs without incorporating CTmin data (a) Phrynocephalus theobaldi and (b) Phrynocephalus erythrurus, and with incorporating CTmin data (c) Phrynocephalus theobaldi and (d) Phrynocephalus erythrurus.
Animals 15 03297 g002
Animals 15 03297 g003
Figure 3. Prediction of suitable habitat for two lizards in Qinghai–Tibetan Plateau. (a) traditional SDMs, (b) physiology-informed SDMs.
Figure 3. Prediction of suitable habitat for two lizards in Qinghai–Tibetan Plateau. (a) traditional SDMs, (b) physiology-informed SDMs.
Animals 15 03297 g003
Animals 15 03297 g004
Figure 4. The difference between the two SDMs (with/without incorporating CTmin data) in predicting habitat of the (a) Phrynocephalus theobaldi and (b) Phrynocephalus erythrurus in Qinghai–Tibetan Plateau. Red areas indicate the reduction in suitable habitat after incorporating CTmin data, while blue areas indicate the expansion.
Figure 4. The difference between the two SDMs (with/without incorporating CTmin data) in predicting habitat of the (a) Phrynocephalus theobaldi and (b) Phrynocephalus erythrurus in Qinghai–Tibetan Plateau. Red areas indicate the reduction in suitable habitat after incorporating CTmin data, while blue areas indicate the expansion.
Animals 15 03297 g004
Table 1. Predictor variables considered in potential distribution modeling.
Table 1. Predictor variables considered in potential distribution modeling.
TypesVariablesDescriptionPhrynocephalus theobaldPhrynocephalus erythrurus
Without
CTmin Data		Incorporating
CTmin Data		Without
CTmin Data		Incorporating
CTmin Data
Climatic
variables		Bio1Annual Mean Temperature
Bio2Mean Diurnal Range (Mean of monthly (max temp–min temp))✓✓✓✓
Bio3Isothermality (Bio2/Bio7) (×100)✓✓✓✓
Bio4Temperature Seasonality (standard deviation × 100) ✓
Bio5Max Temperature of Warmest Month✓✓✓
Bio6Min Temperature of Coldest Month✓
Bio7Temperature Annual Range (Bio5–Bio6)✓✓ ✓
Bio8Mean Temperature of Wettest Quarter
Bio9Mean Temperature of Driest Quarter ✓ ✓
Bio10Mean Temperature of Warmest Quarter ✓
Bio11Mean Temperature of Coldest Quarter ✓
Bio12Annual Precipitation✓ ✓✓
Bio13Precipitation of Wettest Month ✓
Bio14Precipitation of Driest Month✓✓ ✓
Bio15Precipitation Seasonality (Coefficient of Variation)✓✓✓✓
Bio16Precipitation of Wettest Quarter
Bio17Precipitation of Driest Quarter ✓
Bio18Precipitation of Warmest Quarter
Bio19Precipitation of Coldest Quarter✓✓ ✓
Geographical
factors		AltitudeDigital elevation model
SlopeDerived from DEM ✓✓
AspectDerived from DEM ✓✓
Physiological
factors		Cold stress
frequency		total number of cold stress days
during the overwintering period		 ✓ ✓
✓ indicates that the variable was included in this model.
Table 2. Descriptive statistics for critical thermal minimum of Phrynocephalus theobaldi.
Table 2. Descriptive statistics for critical thermal minimum of Phrynocephalus theobaldi.
PopulationNSnout-Vent Length/mmCritical Thermal Minimum/°CRange
Gar1243.53 ± 1.580.9 ± 0.20.5–1.2
Zhongba1249.10 ± 3.620.8 ± 0.10.6–1.0
Table 3. Evaluation indices of species distribution model, expressed as mean ± SD.
Table 3. Evaluation indices of species distribution model, expressed as mean ± SD.
SpeciesPhrynocephalus theobaldiPhrynocephalus erythrurus
ModelKAPPAAUCTSSKAPPAAUCTSS
Traditional
SDMs		CTA0.530 ± 0.1550.795 ± 0.0860.557 ± 0.1590.688 ± 0.1470.904 ± 0.0680.799 ± 0.144
GAM0.692 ± 0.1250.924 ± 0.0470.715 ± 0.1140.566 ± 0.1890.895 ± 0.0730.650 ± 0.198
GBM0.669 ± 0.1170.921 ± 0.0430.681 ± 0.1230.633 ± 0.1980.954 ± 0.0480.629 ± 0.217
GLM0.683 ± 0.1300.893 ± 0.0660.695 ± 0.1300.533 ± 0.1930.805 ± 0.1090.600 ± 0.216
MARS0.704 ± 0.1290.932 ± 0.0540.708 ± 0.1220.623 ± 0.1960.938 ± 0.0540.691 ± 0.199
MAXENT0.533 ± 0.1610.765 ± 0.0840.530 ± 0.1670.587 ± 0.1770.836 ± 0.0970.672 ± 0.193
RF0.697 ± 0.1220.936 ± 0.0390.687 ± 0.1300.615 ± 0.2440.956 ± 0.0400.575 ± 0.260
XGBOOST0.605 ± 0.1650.877 ± 0.0690.600 ± 0.1690.629 ± 0.2050.923 ± 0.0850.622 ± 0.218
EM *0.6730.9780.8770.7060.9920.944
SDMs
With
CTmin data		CTA0.662 ± 0.1590.870 ± 0.0840.697 ± 0.1520.707 ± 0.1350.902 ± 0.0610.803 ± 0.121
GAM0.705 ± 0.1140.934 ± 0.0380.720 ± 0.1170.617 ± 0.1950.916 ± 0.0780.674 ± 0.217
GBM0.763 ± 0.1130.948 ± 0.0410.769 ± 0.1150.707 ± 0.1680.964 ± 0.0340.722 ± 0.182
GLM0.681 ± 0.1390.885 ± 0.0730.694 ± 0.1390.532 ± 0.2020.814 ± 0.1090.616 ± 0.225
MARS0.737 ± 0.1220.932 ± 0.0490.749 ± 0.1210.654 ± 0.1810.942 ± 0.0460.727 ± 0.183
MAXENT0.603 ± 0.1400.801 ± 0.0740.602 ± 0.1480.512 ± 0.1880.798 ± 0.1020.595 ± 0.203
RF0.757 ± 0.1020.954 ± 0.0270.757 ± 0.1070.694 ± 0.2000.969 ± 0.0320.657 ± 0.220
XGBOOST0.656 ± 0.1440.904 ± 0.0560.652 ± 0.1450.686 ± 0.1810.925 ± 0.0720.685 ± 0.199
EM *0.7050.9820.9040.6740.9890.965
* Final ensemble model.
Table 4. Evaluation indices of species distribution model.
Table 4. Evaluation indices of species distribution model.
SpeciesPhrynocephalus theobaldiPhrynocephalus erythrurus
ModelTraditional SDMsSDMs with CTmin DataTraditional SDMsSDMs with CTmin Data
CTA18547267
GAM59563734
GBM44683853
GLM49502630
MARS48664045
MAXENT15213623
RF40693443
XGBOOST27364047
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

Tao, X.; Li, Y.; Li, J. The Effects of Cold Tolerance on the Distribution of Two Extreme Altitude Lizard Species in the Qinghai–Tibetan Plateau. Animals 2025, 15, 3297. https://doi.org/10.3390/ani15223297

AMA Style

Tao X, Li Y, Li J. The Effects of Cold Tolerance on the Distribution of Two Extreme Altitude Lizard Species in the Qinghai–Tibetan Plateau. Animals. 2025; 15(22):3297. https://doi.org/10.3390/ani15223297

Chicago/Turabian Style

Tao, Xiaqiu, Yiyi Li, and Jiasheng Li. 2025. "The Effects of Cold Tolerance on the Distribution of Two Extreme Altitude Lizard Species in the Qinghai–Tibetan Plateau" Animals 15, no. 22: 3297. https://doi.org/10.3390/ani15223297

APA Style

Tao, X., Li, Y., & Li, J. (2025). The Effects of Cold Tolerance on the Distribution of Two Extreme Altitude Lizard Species in the Qinghai–Tibetan Plateau. Animals, 15(22), 3297. https://doi.org/10.3390/ani15223297

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