Herpetozoa 38: 191-204 (2025) DOI 10.3897/herpetozoa.38.e151017 OGH Herpetozoa Austrian Herpetological Society Ecological niches and climate-driven range shifts in Hemorrhois snakes: implications for biogeography Mehmet Kiirsat Sahin’ 1 Department of Biology, Faculty of Science, Hacettepe University, 06800 Ankara, Turkiye https://zoobank. org/996523A 6-F FOD-43C5-BEB8-5CEA449233C7 Corresponding author: Mehmet Kursat Sahin (kursat.sahin@hacettepe.edu.tr) Academic editor: Yurii Korniley @ Received 21 February 2025 @ Accepted 21 July 2025 @ Published 6 August 2025 Abstract Understanding the factors shaping species distributions is essential for predicting their responses to environmental change. The genus Hemorrhois (horseshoe whip snakes) comprises ecologically diverse colubrid snakes found across the Mediterranean Ba- sin, North Africa, the Middle East, and Central Asia. Despite this broad range, their ecological niches and distributional dynamics remain understudied. This study employs ecological niche modeling (ENM) to assess the biogeography, niche differentiation, and potential climate-driven range shifts of H. algirus, H. hippocrepis, H. nummifer, and H. ravergieri under future climate scenari- os. Using species occurrence data and bioclimatic variables, I constructed ensemble models to predict suitable habitats, evaluate niche overlap, and quantify potential range changes. Results indicate significant variation in climate-driven distributional responses among species. Hemorrhois algirus is projected to expand across North Africa, whereas H. hippocrepis, H. nummifer, and H. raver- gieri may face range contractions under high-emission scenarios. Niche analyses suggest moderate overlap between H. algirus and H. hippocrepis, implying historical and ecological connectivity, while H. nummifer and H. ravergieri display distinct environmental preferences. Climatic and geographic barriers—such as the Sahara Desert, the Dardanelles and Istanbul Straits, the Alps, and the Pyrenees Mountains—play crucial roles in shaping their evolutionary trajectories. Given the increasing threats of climate change and habitat loss, this study underscores the need for conservation strategies prioritizing habitat connectivity, species-specific man- agement, and climate refugia. By integrating ecological and evolutionary perspectives, this research contributes to understanding Mediterranean and Western Palearctic reptile biogeography and their responses to environmental change. Key Words climate change, habitat loss, niche differentiation, Squamata, Western Palearctic Introduction Species distributions and their ecological niches are shaped by a combination of historical biological and geo- logical events, climatic fluctuations, and biotic interactions (Franklin 2009). Continental drift, glaciations, and envi- ronmental shifts have profoundly influenced present-day biodiversity by driving speciation and dispersal (Wiens 2011; Pelegrin et al. 2021). Within these frameworks, inter- specific competition, environmental pressures, and selec- tive forces further shape evolutionary trajectories (Angert and Schemske 2005; Sexton et al. 2009). Understanding the processes governing species’ ecological niches Is criti- cal for predicting their responses to environmental change and informing conservation efforts (Soberon and Peterson 2005; Franklin 2009; Vaissi and Mohammadi 2024). Climate change represents a significant threat to many reptilian taxa, particularly those with narrow ecological tolerances or fragmented distributions (Vaissi et al. 2023; Sahin 2024). Recent advances in ecological modeling, cou- pled with global climate and land cover datasets, now allow researchers to forecast species distributions under different environmental scenarios. Ecological niche models (ENMs) integrate occurrence data and environmental variables to Copyright Mehmet Ktrsat Sahin. This is an open access article distributed under the terms of the Creative Commons Attribution } PENSUFT. License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 192 Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois predict potential species range shifts under climate change (Phillips and Dudik 2008; Peterson et al. 2011). Howev- er, significant conceptual and statistical challenges exist in niche modeling using occurrence records that encompass a geographically representative range of species. For example, most ENMs exhibit spatial correlation among environmen- tal variables, such as temperature, which may lead to misin- terpretation of significant niche modeling results as a conse- quence of geographical distance (McCormack et al. 2010). The genus Hemorrhois (Colubridae), commonly re- ferred to as the horseshoe whip snakes, comprises four currently recognized species: Hemorrhois algirus, H. hip- pocrepis, H. nummifer, and H. ravergieri (Abreu 2017; Far- aone et al. 2020; Kazemi et al. 2023). This genus exhibits a broad yet regionally distinct distribution, primarily span- ning the Mediterranean Basin, North Africa, the Middle East, and parts of Central and South Asia (Carranza et al. 2006; Abreu 2017; Bulbul et al. 2019; Faraone et al. 2020). Hemorrhois snakes display notable ecological adaptabili- ty, occupying diverse habitats including arid and semi-arid landscapes, rocky terrains, and anthropogenically modified environments (Montes et al. 2020). The broad distribution and ecological plasticity of Hemorrhois make this genus an ideal model system for testing biogeographic and ecologi- cal hypotheses. Such studies could elucidate how ecologi- cal niches and climate-driven range shifts have shaped their current distributions while also providing insights into po- tential future range dynamics under contemporary climate change scenarios (Winter et al. 2016; Veverkova 2021). Despite their broad distributions, Hemorrhois snakes generally remain relatively understudied with regard to their ecological interactions and responses to environ- mental changes. The integration of species distribution modeling and niche differentiation assessments will clar- ify the role of climate in their speciation and habitat re- quirements for conservation planning. Given the increas- Ce es ae herpetozoa.pensoft.net ing pressures of habitat destruction and climate change, future research should prioritize identifying key refugia, assessing population viability, and implementing conser- vation strategies that account for both regional and global threats to these snakes (Bombi et al. 2011). This study provides a comprehensive assessment of the biogeography and ecological niches of the genus Hemor- rhois by integrating occurrence records, climatic variables, and predictive modeling techniques. I hypothesize that Hem- orrhois species exhibit significant climatic niche differenti- ation corresponding to their ecological and biogeographic diversity and that climate change will drive species-specif- ic range shifts. To test this hypothesis, I developed ENMs based on climatic variables and species occurrence data to (i) evaluate the relationship between current Hemorrhois distributions and observed climate and forecast potential future species distributions under different climate change scenarios (2081-2100) and (11) measure and compare cli- matic niche divergence within the genus. These findings will contribute to a better understanding of Mediterranean and Palearctic reptile biogeography and offer insights into the resilience and adaptability of Hemorrhois species in the face of ongoing environmental transformations. Materials and methods Study area and species occurrences The study area encompasses the entire distribution- al range of Hemorrhois (20°W to 80°E longitude, 25° to 48°N latitude), spanning diverse ecosystems across Southern Europe (including the Iberian Peninsula), North Africa, the Middle East, and Central and Southwestern Asia (Fig. 1). This comprehensive coverage captures the full spectrum of ecological conditions inhabited by the ay a y Figure 1. Species occurrence records for genus Hemorrhois in Asia, the Middle East, Southern Europe, and North Africa. Herpetozoa 38: 191-204 (2025) genus, enabling robust characterization of their niche preferences. Occurrence records for Hemorrhois species were sourced from multiple repositories, including her- petological literature, personal observation data, and on- line platforms (see Suppl. material 1). The data under- went a two-step process for cleaning and validation to ensure quality and accuracy. The georeferenced occur- rence data were systematically analyzed to identify and correct errors and inconsistencies. Initially, I screened the data’s geographic accuracy. Records with fewer than three decimal places in their coordinates (equivalent to a spatial uncertainty > 100 meters) were flagged for fur- ther validation. For each record, the locality description was cross-checked with the mapped coordinates using GIS software (QGIS v. 3.40.4-Bratislava) to ensure consistency (QGIS 2025). Records showing discrepan- cies between the locality description and geographic po- sition were excluded. For records obtained from online platforms (GBIF 2024; iNaturalist 2025), I prioritized those marked with high location accuracy (<100 m un- certainty) and validated with photographic evidence or detailed observation notes. Furthermore, to ensure tem- poral consistency with the bioclimatic layers (CHELSA v. 2.1; Karger et al. 2017; baseline period 1970—2000), I primarily included occurrence records collected during or close to this timeframe. Records with uncertain iden- tification, unclear locations, or poor spatial resolution were omitted from the final dataset to maximize data reliability (Chapman 2005). To reduce geographic sam- pling biases and improve the interpretation of habitat suitability analyses and niche overlap tests, the R pack- age spThin (Aiello-Lammens et al. 2015) was utilized to spatially rarefy the occurrence records for each Hem- orrhois species. One locality was retained for every 5 km linear distance to maintain a balance between data density and spatial representation. The retained records were as follows: H. algirus — 109 from 170; H. hippo- crepis — 1,071 from 7,992: H. nummifer — 239 from 547; and H. ravergieri — 161 from 281. Bioclimatic variables Bioclimatic variables were obtained from the CHEL- SA database at a spatial resolution of 30 arc-seconds (Karger et al. 2017; Brun et al. 2022), and all layers were initially clipped to a broad study area covering the Western Palearctic and adjacent regions. To improve model robustness and avoid overprediction, I defined a species-specific calibration area (M) for each Hemor- rhois species, following recommendations from ecolog- ical niche modeling theory (Soberon and Peterson 2005; Barve et al. 2011; Luna et al. 2024). These M areas were delineated based on each species’ known geographic distribution, major dispersal barriers, and regional bio- geographical boundaries. Environmental layers were masked to each species’ M area prior to modeling, en- suring that pseudo-absences and background data were 18 sampled only from regions ecologically accessible to the species. This strategy reduces model bias and 1m- proves the reliability of post-modeling analyses such as niche overlap and divergence (Aratjo and New 2007; Phillips et al. 2009; Luna et al. 2024). Pearson correlations among variables were computed using R v4.3 (R Core Team 2024), and variables with high correlations (r > |0.8]) were removed. Eight bio- climatic variables were selected for model construction across all four species: mean diurnal air temperature range (Bio 2), isothermality (Bio 3), temperature sea- sonality (Bio_4), mean daily maximum air temperature of the warmest month (Bio_5), daily mean air tem- peratures of the wettest quarter (Bio_8), daily mean air temperatures of the driest quarter (Bio_9), precipi- tation seasonality (Bio_15), and mean monthly precip- itation amount of the warmest quarter (Bio_ 18); here- after, the Bio_ codes are used throughout. All variables were applied to forecast species niches under recent (1970-2000) and future (2071-2100) climate change projections using GFDL-ESM4, IPSL-CM6A-LR, MPI- ESM1-2-HR, MRI-ESM2-0, and UKESM1-0-LL mod- els. These global circulation models were selected based on their optimal performance for the study area, and projections were evaluated under the lowest and highest shared socioeconomic pathways (SSPs) from the Cou- pled Model Intercomparison Project Phase 6 (CMIP6) (Eyring et al. 2016; Sun et al. 2022). Ecological niche modeling An ensemble model was developed to predict the po- tential suitable habitats of Hemorrhois species, utilizing six distinct algorithms: generalized linear model (GLM), generalized additive model (GAM), surface range enve- lope (SRE/BIOCLIM), domain model (DM), random forest (RF), and maximum entropy (MAXENT). This was implemented using the ENMTools and kuenm pack- ages (Cobos et al. 2019; Warren et al. 2021) in R. The ensemble model represents a proportional aggregation of the responses generated by the algorithms (Table 1). This approach leverages the strengths of these algorithms while reducing their respective weaknesses, enhances accuracy, and offers a quantifiable measure of uncertain- ty in predictions (Araujo and New 2007). Some of these algorithms require datasets that include both presence and absence data; however, obtaining real absence data is challenging. To minimize potential biases, pseudo-ab- sences were generated using a target-group sampling approach (Phillips et al. 2009). Pseudo-absences were selected based on environmental constraints, ensuring they were located in areas not occupied by the species but within suitable climatic conditions. The models were calibrated using 80% of the data (training set) and as- sessed using the remaining 20% (validation set). This process was repeated three times to enhance model training coverage and increase robustness. herpetozoa.pensoft.net 194 Mehmet Kursat Sahin: Ecological niches and climate responses in Hemorrhois Table 1. Model comparisons for species distribution of genus Hemorrhois according to AUC values. Species GLM GAM RF DM BC MaxEnt H. algirus 0.847 0.861 0.903 0.873 0.864 0.955 H. hippocrepis 0.861 0.829 0.888 0.851 0.910 0.943 H. nummifer 0.865 0.868 0.922 0.899 0.898 0.959 H. ravergieri 0.805 0.811 0.813 0.804 0.810 0.915 *GLM: General linear model, GAM: General Additive Model, RF: Random For- est, DM: Domain Model, BC: Bioclim, MaxEnt: Maximum Entropy. Model selection To ensure optimal model complexity, 341 candidate MAXENT models were tested using combinations of feature classes (hinge, threshold, product, quadratic, and linear) and regularization multipliers (ranging from 0.1 to 10). Feature class selection was based on previ- ous studies demonstrating their role in controlling mod- el complexity and preventing overfitting (Muscarella et al. 2014: Cobos et al. 2019). Regularization multipliers were adjusted to balance model generality and com- plexity, ensuring biologically meaningful predictions (Radosavljevic and Anderson 2014). The application of these combinations provided an optimal strategy for generating diverse candidate models, enabling the se- lection of those that best explained the data. Subsequently, the 341 candidate MAXENT mod- els were evaluated using a multi-criteria assessment framework. Optimal models were selected based on (1) the highest Area Under the Curve (AUC) values, (11) the lowest Akaike Information Criterion corrected for small sample sizes (AICc) (Hurvich and Tsai 1989), (111) statistical significance based on partial ROC (pROC) tests (Peterson et al. 2008), and (iv) a predictive omis- sion rate threshold set at 5% (Table 2) (Anderson et al. 2003). This combination of independent metrics offers a comprehensive and robust assessment of model per- formance and reliability, reducing the risk of overfitting and ensuring that selected models reflect both statisti- cal and ecological relevance (Araujo and New 2007). AUC values were interpreted as follows: AUC = 0.5 indicates random prediction, AUC > 0.7 suggests use- ful performance, > 0.8 is satisfactory, and > 0.9 is ex- cellent (Manel et al. 2001; Adhikari et al. 2018; Lietal. 2024). By integrating these complementary evaluation methods, the final models were both statistically robust and ecologically meaningful. Projections for the recent period were generated using the optimal configuration of the statistical model, utilizing 100% of the records and conducting 5,000 iterations with 10 replicates. All model outputs were ultimately converted into bi- nary predictions using the minimum training presence threshold to distinguish between unsuitable and suit- able areas (Pearson 2007; Rodriguez-Ruiz et al. 2020). Future projections were performed using climate layers from five general circulation models (GFDL-ESM4, IPSL-CM6A-LR, MPI-ESM1-2-HR, MRI-ESM2-0, and UKESM1-0-LL) under two shared socioeconomic pathways (SSP126 and SSP585). During model trans- fer to future conditions, extrapolation was permitted but carefully constrained based on model response curves to avoid biologically implausible projections (Owens et al. 2013; Gomes et al. 2018). To assess and quantify areas of strict extrapolation—where environ- mental conditions differ substantially from those of the calibration region—a Mobility-Oriented Parity (MOP) analysis was conducted using the new MOP package (Cobos et al. 2024). MOP results were used to identify regions of higher uncertainty in future projections, and interpretations of range shifts were made cautiously, taking these uncertainty zones into account (Suppl. ma- terials 2-5). The contributions of the bioclimatic vari- ables are presented in Table 3. Median values from the replicates were used to summarize model predictions for each climate scenario (Figs 2-5). Species range change To assess and visualize potential species range change (SRC) in Hemorrhois species under climate change scenarios, a spatial analysis approach was employed to generate maps depicting regions where species may experience gains or losses in suitable conditions. The metrics include “Loss” (the number of pixels antici- pated to become unsuitable), “Absent” (the number of pixels expected to remain unsuitable), “Stable” (the number of pixels projected to remain suitable), and “Gain” (the number of pixels predicted to become suit- able). These estimates are based on model predictions and do not directly represent the species’ actual area of occupancy, as they are derived from binarized mod- el outputs using a specified threshold (Guisan et al. 2018). In addition to absolute measures, three relative Table 2. Summary statistics for the best models selected for species distribution maps of Hemorrhois species. Species Feature Candidate Statistically Mean AUC ratio Partial Omission rate AICe AAICe WAICe AUC models significant models ROC at 5% H. algirus product + threshold 341 340 1.741 0 0.044 2891.214 0 0.914 0.955 H. hippocrepis quadratic+ product 341 36 1.873 0 0.049 2893.732 0 1 0.943 HI. nummifer linear + threshold 341 175 1.729 0 0.039 6643.085 0 0.718 0.959 H. ravergieri product 341 341 1.420 0 0.090 4873.719 0 1 0.915 AICc: a corrected AIC score, used for a small sample size by increasing the cost for each parameter; wAICc: the model weight is the relative likelihood for each model divided by the total relative likelihood for all models that were considered. AAICc: the difference between the model with the lowest score (the “best” model) and the AICc score for each model; AUC: area under the curve is a measure of the accuracy of the model; mean AUC ratio > 1.00, p< 0.05 means predictions are significantly better than a random model. herpetozoa.pensoft.net Herpetozoa 38: 191-204 (2025) Table 3. Contributions of bioclimatic variables. Species / Variable % Bio2 Bio3 Bio4 Bio5 Bio8 Bio9 Bio 15 Bio 18 H. algirus 5.1 31.9 155 09 44 3.7 71 314 H. hippocrepis 1.5 154 298 43 19 27 16.7 17.7 HL. nummifer 15.6 08 1.9 7.7 195 68 25.1 12.6 H. ravergieri 13.9 41 341 183 55 411 48 15.2 metrics were developed to clarify the potential effects of climate change on the distributions of Hemorrhois species (Table 4; Suppl. material 6): 1. Percentage Loss: the proportion of currently occu- pied sites anticipated to be lost. Calculated as: Loss / (Loss + Stable). 2. Percentage Gain: the ratio of new sites anticipated for the species relative to its existing distribution size. Calculated as: Gain / (Loss + Stable). 3. Range Change: the net alteration in the species’ range size, accounting for both gains and losses. Calculated as: Percentage Gain — Percentage Loss. These metrics provide important insights into the potential impacts of climate change on the distribution of Hemorrhois species. The analysis of these metrics across different climate scenarios enables a thorough understanding of potential range shifts for each species (Guisan et al. 2018). 195 Niche analysis To evaluate ecological niche differentiation among Hem- orrhois species, I implemented a comprehensive analyt- ical framework combining traditional overlap metrics with multivariate statistical techniques. First, I calculat- ed Schoener’s D (difference-focused) and Hellinger’s I (similarity-focused) indices to quantify niche similarity, with values ranging from O (no overlap) to 1 (identical niches) (Warren et al. 2021). Subsequently, a multivariate analysis of variance (MANOVA) was conducted to assess whether the environmental conditions occupied by each species differed significantly. To further validate the re- sults of the niche overlap analyses, background similarity tests were performed using a Monte Carlo randomization approach with 500 replicates, comparing observed niche overlap metrics (Schoener’s D and Hellinger’s I) against a null distribution of randomized backgrounds (Warren et al. 2021) (Suppl. material 7). This higher number of replicates ensured a more stable and reliable estimation of null distributions, allowing for robust significance testing (Table 5; Suppl. material 8). Although climate variables are primary drivers of niche differentiation (e.g., competi- tive exclusion), habitat partitioning may further constrain species distributions. It is therefore essential to under- stand whether ecological niches remain conserved or di- verge over evolutionary time. Niche conservatism refers Hemorrhots algirus Figure 2. Recent (1970-2000) and future (2081-2100) climatic suitability for Hemorrhois algirus based on different models under the optimistic (ssp126) and pessimistic (ssp585) scenarios. (a. Recent; b. GFDL 126; e. GFDL 585; d. IPSL 126; e. IPSL 585; f. MPI 126; g. MPI 585; h. MRI 126; i. MRI 585; j. UKESM 126; k. UKESM 585). herpetozoa.pensoft.net 196 Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois Hemorrhots hippocrepis L . Low High Figure 3. Recent (1970-2000) and future (2081-2100) climatic suitability for Hemorrhois hippocrepis based on different models under the optimistic (ssp126) and pessimistic (ssp585) scenarios. (a. Recent; b. GFDL 126; e. GFDL 585; d. IPSL 126; e. IPSL 585; f. MPI 126; g. MPI 585; h. MRI 126; i. MRI 585; j. URESM 126; k. UKESM 585). Hemorrhois nummifer High Figure 4. Recent (1970-2000) and future (2081—2100) climatic suitability for Hemorrhois nummifer based on different models under the optimistic (Ssp126) and pessimistic (ssp585) scenarios. (a. Recent; b. GFDL 126; e. GFDL 585; d. IPSL 126; e. IPSL 585; f. MPI 126; g. MPI 585; h. MRI 126; i. MRI 585; j. UKESM 126; k. UKESM 585). herpetozoa.pensoft.net Herpetozoa 38: 191-204 (2025) 197 Figure 5. Recent (1970-2000) and future (2081—2100) climatic suitability for Hemorrhois ravergieri based on different models under the optimistic (Sssp126) and pessimistic (ssp585) scenarios. (a. Recent; b. GFDL 126; ec. GFDL 585; d. IPSL 126; e. IPSL 585; f. MPI 126; g. MPI 585; h. MRI 126; i. MRI 585; j. URESM 126; k. UKESM 585). Table 4. Species range change (SRC) of Hemorrhois species in recently suitable habitats (gain/loss) by 2081—2100 under optimistic (ssp126) and pessimistic (ssp585) scenarios. Species H. algirus H. hippocrepis H. nummifer H. ravergieri Models GFDL IPSL MPI MRI UKESM GFDL IPSL MPI MRI UKESM GFDL IPSL MPI MRI UKESM GFDL IPSL MPI MRI UKESM Loss% 24.835 25.452 16.785 30.486 25.435 29.754 13.709 4.987 17.442 9.604 48.962 42.159 20.834 42.659 57.535 3.394 8.979 8.814 13.606 32.516 ssp 126 Gain% 51.704 57.618 39.717 39.004 47.649 7.031 9.614 16.539 12.587 35.054 12.032 21.548 25.842 18.103 21.942 7.071 3.204 1.554 4.423 0.716 Table 5. Identity and Background Tests for genus Hemorrhois. Hemorrhois comparisons SRC 26.869 32.166 221932 8.518 22.214 -22.723 -4.095 11.552 -4.855 25.450 -36.930 -20.611 5.008 -24.556 -35.593 3.677 -5.775 -7.260 -9.183 -31.800 Loss% 37.479 55.364 66.008 56.218 58.336 33.283 36.498 79.754 42.728 23.948 93.914 95.113 97.441 89.894 96.142 9.207 20.203 43.978 10.183 40.609 ssp 585 Gain% 87.835 74.289 52.168 66.134 69.940 14.912 21.176 9.542 14.378 45.717 26.065 23.581 9.971 24.405 25.915 6.996 6.109 1.839 8.210 1.086 SRC 50.356 18.925 -13.840 9.916 11.604 -18.371 -15.322 -70.212 -28.350 21.769 -67.849 -71,532 -87.470 -65.489 -70.227 -2.211 -14.094 -42.139 -1.973 -39.523 algirus vs. hippocrepis algirus vs. nummifer algirus vs. ravergieri hippocrepis vs. nummifer hippocrepis vs. ravergieri nummifer vs. ravergieri Identity test Background test (asymmetric) Background test (symmetric) D, D, I, re D, fig ri fi D, D, I, L, 0.429 0.830 0.716 0.976 0.454 0.674 0.744 0.901 0.446 0.689 0.728 0.898 0.275 0.727 0.551 0.918 - = 2 . = 7 A a 0.290 0.843 0.565 0.978 - ~ : ¥ : 7 - : 0.273 0.854 0.567 0.969 - = 7 7 - Z : . 0.387 0.833 0.672 0.978 - E - = : _ 2 : 0.560 0.870 0.833 0.985 0.578 0.582 0.853 0.846 0.563 0.584 0.819 0.807 herpetozoa.pensoft.net 198 Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois to the tendency of species to retain ancestral ecological traits, while niche divergence implies adaptation to dif- ferent environmental conditions. By integrating statisti- cal validation with ecological interpretation, this study provides a comprehensive framework for understanding niche dynamics in Hemorrhois. Results Ecological niche models predicted distinct environmental suitability patterns among Hemorrhois species, with consis- tent outputs across algorithms. The performance evaluation of the models—based on AUC, AICc, partial ROC, and omission rates—indicated that all selected models demon- strated high predictive accuracy and statistical significance, supporting their reliability for subsequent analyses (Tables 1, 2). The regions projected to be suitable for Hemorrhois Species demonstrated significant model performance: H. al- girus (AUC = 0.955, omission rate at 5% = 0.044), H. hippo- crepis (AUC = 0.943, omission rate at 5% = 0.049), H. num- mifer (AUC = 0.959, omission rate at 5% = 0.039), and H. ravergieri (AUC = 0.915, omission rate at 5% = 0.090). Although several statistically significant models were iden- tified for each Hemorrhois species, only one per species met the AICc criterion of < 2. The relative contributions of bio- climatic variables to the ecological niche models of Hemor- rhois species are summarized in Table 3. Model projections under future climate scenarios suggest heterogeneous responses among Hemorrhois species, with some showing potential range expansions, others contractions, and a few maintaining relatively sta- ble distributions (Figs 2—5). Projected patterns of range stability, expansion, and contraction for each Hemorrhois Species are summarized in Suppl. material 6 and Table 4. The following results describe species-specific responses of Hemorrhois taxa to future climate scenarios, highlight- ing patterns of range expansion, contraction, and stability. For H. algirus, Bio_3 (31.9%) and Bio_18 (31.4%) were the two most important variables affecting its po- tential distribution (Table 3). The habitat suitability map under recent climatic conditions indicates that North Af- rica—particularly Morocco and northern Algeria—has high suitability (Fig. 2a). While the species occurs in arid environments such as Tunisia and Western Sahara, it is projected to occupy these regions with relatively limited suitable habitat (Fig. 2b—k). In the future, the distribution range of H. algirus may expand further in North Africa. For example, under the SSP585 scenario, the GFDL (Fig. 2c) and IPSL (Fig. 2e) models predict range expansions of approximately 50.3% and 18.9%, respectively, com- pared to the current distribution (Table 4). Based on the SRC analysis, range expansion is expected under all op- timistic and pessimistic future climate scenarios, except for MPI SSP585 (—13.8%) (Suppl. material 6: fig. SSA). For H. hippocrepis, Bio 4 (29.8%) and Bio_18 (17.7%) were identified as the most influential variables shaping its potential distribution (Table 3). The habitat herpetozoa.pensoft.net suitability map under current climatic conditions indi- cates a broad and continuous distribution across southern Europe and the western Mediterranean coast. In particu- lar, the Iberian Peninsula (Spain and Portugal) and parts of Italy exhibit high suitability (Fig. 3a). The species shows a notable preference for coastal regions over inland areas (Fig. 3b—k). In the future, a significant range con- traction 1s projected under high-emission scenarios. For instance, under the SSP585 scenario in the MPI model, a loss of approximately 70.2% of suitable habitat is pre- dicted within the species’ current distribution range (Fig. 3g, Table 4). SRC analysis further supports this trend, indicating range contraction under all future climate sce- narios—except expansions in MPI SSP126 (11.5%) and both UKESM scenarios (25.4% under SSP126 and 21.7% under SSP585) (Suppl. material 6: fig. S5B). For H. nummifer, Bio_15 (25.1%) and Bio_8 (19.5%) were the most influential variables contributing to its po- tential distribution (Table 3). The habitat suitability map under current climatic conditions shows a wide distribu- tion primarily across the Levantine corridor, extending into parts of western Anatolia and the Middle East (Fig. 4a). This pattern suggests adaptation to semi-arid and Mediterranean environments. Under future climate sce- narios, a notable range contraction is projected, particu- larly under high-emission conditions. For example, under the SSP585 scenario, the MPI model predicts a loss of ap- proximately 87.4% of suitable habitat, especially in sur- rounding countries of the Levantine corridor and western Anatolia (Fig. 4f, Table 4). According to the SRC analy- sis, habitat loss is expected under nearly all future scenari- os—specifically 4 out of 5 under SSP126 and all 5 models under SSP585—except for MPI SSP126, which predicts a limited expansion (5%) (Suppl. material 6: fig. SSC). For H. ravergieri, Bio_4 (34.1%) and Bio_5 (18.3%) were the most influential variables shaping its potential dis- tribution (Table 3). The habitat suitability map under cur- rent climatic conditions highlights high suitability across South-Central Asia—particularly northwestern Kazakh- stan and Mongolia—as well as western Asia (Turktye) and the Caucasus region (Fig. 5a). In future climate scenarios, areas at higher elevations are projected to become increas- ingly suitable (Fig. 5b-k). However, under the SSP585 scenario, both the MPI and UKESM models predict nota- ble range contractions, with losses of 42.1% and 39.5% of suitable habitat, respectively, particularly in the Caucasus and western Asia (Table 4, Suppl. material 6: fig. SSD). The comparative analysis of range shift patterns among Hemorrhois species revealed contrasting responses to projected climate change. Hemorrhois algirus is gener- ally expected to expand its suitable range across North Africa, while H. hippocrepis, H. nummifer, and H. raver- gieri are projected to experience significant range con- tractions under pessimistic climate scenarios. The degree of projected range loss was highest for H. nummifer and H. ravergieri, particularly in regions of complex topog- raphy and aridification, suggesting species-specific sen- sitivity to climatic variables and geographic constraints. Herpetozoa 38: 191-204 (2025) The measured niche overlaps among all species are presented in Table 5. The null hypothesis regarding niche overlap among Hemorrhois species (excluding nummifer vs. ravergieri and hippocrepis vs. algirus) was rejected, as the empirical values for Schoener’s D and Hellinger’s I test statistics differed significantly from the null distribution of overlap tests for each species comparison (Suppl. material 7: fig. S6A-F) (t test, df = 99, P < 0.05). The ecological niche models of the majority of these species were not equivalent. The asymmetric and symmetric background tests for the parapatric H. algirus and H. hippocrepis, as well as H. nummifer and H. ravergieri, confirmed partial niche overlap between these species with respect to global bioclimatic variables (Suppl. material 8: fig. S7A—D). Discussion This study provides a comprehensive examination of the ecological niche attributes and future range dynamics of four Hemorrhois species. The biogeographical pat- terns of these species are influenced by both historical and contemporary ecological processes. The presence of H. hippocrepis in the western Mediterranean, for in- stance, suggests a complex history of dispersal and vi- cariance events (Bombi et al. 2011). Specifically, physical barriers such as the Sahara Desert, the Alps, the Pyrenees, and the Dardanelles and Istanbul straits have historically limited gene flow and contributed to speciation through allopatric divergence (Machado et al. 2021). The initial divergence between H. algirus and H. hippocrepis has been estimated to have occurred during the late Mio- cene to early Pliocene, approximately 4—7 million years ago (Carranza et al. 2006). Subsequent climatic events during the Pleistocene, including sea-level fluctuations at the Strait of Gibraltar, likely influenced the secondary contact, distributional shifts, or population structure of Hemorrhois species. Based on current distributions and phylogenetic inference, H. algirus and H. hippocrepis may have originated in North Africa, with H. hippocre- pis subsequently dispersing into the Iberian Peninsula (Carranza et al. 2006). Instead of direct transmarine mi- gration following the opening of the Strait of Gibraltar at the end of the Messinian, climatic events during the Pleistocene glaciations resulted in a sea-level drop of ap- proximately 130 m (Anderson and Borns Jr. 1997) at Ca- marinal Sill, where water depths range from 40 to 150 m (Brandt et al. 1996). Consequently, some elevated areas in this region temporarily became small islands, poten- tially allowing certain terrestrial vertebrates—such as the snakes examined here—to traverse the Strait of Gibraltar relatively recently (Carranza et al. 2006). In addition, the Alps (Central Europe) and Pyrenees (between France and Spain) may act as major dispersal barriers due to their cold temperatures, steep terrain, and limited prey avail- ability for H. hippocrepis, which prefers lower elevations with stable temperatures. High-altitude mountain zones often present unsuitable conditions (e.g., frost, low prey 192 density) (Pleguezuelos and Feriche 2002). These findings suggest that climatic conditions and geographic features may have significantly shaped the historical distributions of Hemorrhois species in the region. Notably, physical barriers such as the Dardanelles and Istanbul straits to the north—and the Sahara Desert to the south—could have influenced dispersal and population structure, especially for H. nummifer and H. ravergieri. The Sahara, with its extreme heat, scarce water, and limited prey, likely served as an inhospitable barrier. Likewise, although the Darda- nelles and Istanbul straits are narrow, they probably acted as persistent water barriers between Europe and Asia, re- stricting gene flow in low-dispersal species—particularly given the absence of land bridges during glaciations, unlike the Strait of Gibraltar. Importantly, similar patterns have been documented in other colubrid snakes. For example, the dice snake (Natrix tessellata) exhibits ancestral area reconstruction findings indicating that deserts and moun- tain corridors in Central Asia and Anatolia significantly shaped lineage distributions (beginning ~3.7 Ma), with Pleistocene glacial refugia emerging as key range mod- ifiers (Salvi et al. 2018; Liz et al. 2021; Romero-Iraola et al. 2023; Jablonski et al. 2024). This supports the idea that both climatic events and geographic barriers have historically constrained colubrid dispersal. Incorporating a similar ancestral-area framework for Hemorrhois could further substantiate these biogeographic inferences. Con- sequently, the current distribution patterns of Hemorrhois may also result from allopatric speciation influenced by vicariance, wherein physical barriers caused the separa- tion and divergence of populations. These contrasting responses to climate change may re- flect differences in environmental plasticity, which influ- ence each species’ ability to tolerate or adapt to changing conditions. Species occupying broader climatic niches (H. algirus) are more resilient, while those reliant on me- sic or montane habitats (H. nummifer, H. ravergieri) show heightened vulnerability. Insular and coastal specialists (H. hippocrepis) are additionally constrained by limited dispersal options. These findings highlight that ecological generalists may fare better under climate change, whereas specialists are at greater risk. Hemorrhois algirus occupies broad arid and semi-ar- id habitats in North Africa and is projected to expand its range under future climate scenarios. Its adaptation to harsh environmental conditions, along with ecological flexibility, likely confers resilience to increasing tempera- tures and habitat changes. In contrast, H. hippocrepis, which inhabits coastal and insular Mediterranean regions, is expected to experience moderate range contraction. Geographic isolation on islands and shorelines, coupled with limited dispersal ability, may make this species more vulnerable to habitat loss and climate-driven shifts. Hemorrhois nummifer, distributed across mesic habi- tats in the Levant, shows the highest projected range con- traction. Its dependence on relatively humid conditions renders it particularly sensitive to the aridification trends forecasted under future scenarios. Similarly, H. ravergieri, herpetozoa.pensoft.net 200 which occupies montane and steppe habitats in Western and Central Asia, is projected to lose a substantial portion of its range—especially in high-altitude regions, which are disproportionately affected by temperature increases. These observed patterns are consistent with findings in other reptilian taxa. Genera Timon and Lacerta exhibit niche conservatism and gradual phenotypic shifts linked to historical climatic stability (Enriquez-Urzelai et al. 2022). Similarly, Zarentola mauritanica and genus Anato- lolacerta demonstrate niche divergence and conservatism depending on environmental pressures (Rato et al. 2015; Sahin et al. 2022). Studies on rat snakes (Zamenis spp.) revealed both niche divergence and conservatism among lineages (Vaissi et al. 2024). Such comparisons reinforce the generality of the mechanisms driving niche evolution and distributional dynamics observed in Hemorrhois. Given the projections of range contraction for several Hemorrhois species, conservation strategies must prior- itize habitat connectivity, preservation of climatic refu- gia, and management of cross-border habitats. Adaptive conservation planning tailored to each species’ ecological needs will be crucial. Species like H. algirus may benefit from proactive habitat expansion opportunities, whereas H. nummifer and H. ravergieri will require strategies to mitigate habitat fragmentation and loss. This study also conducted niche analyses to assess the ecological niche overlap among four Hemorrhois spe- cies. The results demonstrated a range of niche differen- tiation—from low to moderate to substantial—reflecting varying degrees of ecological specialization within the genus. The identity and background tests provide insights into niche dynamics, revealing statistical support for niche overlap in only 2 of 8 pairwise comparisons (Table 5; Suppl. materials 7, 8). The comparisons among allo- patric Hemorrhois species indicated that their niches are not more similar than expected by chance, yet they are not equivalent (Suppl. material 3). Studies on allopatric Neurergus species in Anatolia (Gul 2019) and the specia- tion dynamics of endemic lizards in Madagascar (Nunes et al. 2022) indicate that variations in their climatic nich- es align with the abiotic environmental conditions of the geographical regions occupied by these allopatric spe- cies. Species that are adapted to specific climatic or lo- cal conditions experience niche differentiation due to the unique adaptations required for survival and reproduction (Nakazato et al. 2010). The case of niche overlap in parapatric speciation, as illustrated by the comparisons between H. hippocrepis and H. algirus and between H. nummifer and H. ravergi- eri, requires further discussion due to the restricted distri- butions of southern Europe and North Africa for H. hip- pocrepis and H. algirus and the Eastern Mediterranean and Western Asia for H. nummifer and H. hippocrepis. The utilization of the niche is significantly influenced by various ecological interactions. Therefore, incorporat- ing data on diverse selective regimes may aid in analyz- ing the speciation dynamics of these parapatric species (Gavrilets et al. 2000; Mammola et al. 2018). herpetozoa.pensoft.net Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois These findings suggest that, although certain species have established unique ecological niches, a general pat- tern of niche conservatism is evident within the Hemor- rhois genus. This tendency toward niche conservatism supports the theory that speciation in Hemorrhois may be influenced by the preservation of ancestral ecological fea- tures, in line with results reported for other reptile groups (e.g., Morales-Castilla et al. 2011; Pomara et al. 2014; Vaissi et al. 2024). It is important to interpret these findings with caution, given the inherent limitations of correlative ENMs, which do not incorporate factors such as dispersal constraints, physiological tolerances, or species interactions (Kearney and Porter 2009; Peterson et al. 2015). These limitations underscore the potential value of future research integrat- ing mechanistic or hybrid models. Conclusion This study presents a species-level evaluation of ecolog- ical niches and projected future distributions for Hemor- rhois snakes using correlative ecological niche models based on bioclimatic variables. The projections indicate that H. algirus may experience range expansion under future climate conditions, whereas H. nummifer and H. ravergieri are likely to face substantial habitat reductions. These outcomes should be interpreted within the scope of the modeling framework, as correlative ENMs do not in- corporate physiological tolerances, dispersal limitations, or biotic interactions. The results suggest that niche dif- ferentiation has occurred among species within the ge- nus, with varied ecological preferences likely shaped by historical isolation and climatic gradients. The evolution- ary history of Hemorrhois species appears to have been influenced by past climatic fluctuations and geographic obstacles such as deserts, mountain systems, and sea bar- riers. Areas identified as climatically stable under both current and future scenarios may represent important cli- mate refugia that could support long-term population per- sistence. These findings support the prioritization of such regions in conservation planning and emphasize the need to integrate ecological modeling with physiological, ge- netic, and dispersal-based approaches in future research. Acknowledgments I thank Mr. Hanley Garner for proofreading and Assoc. Prof. Muammer Kurnaz, the respected anonymous review- ers, and the section editor for their valuable suggestions. References Abreu JMB (2017) Phylogenetic and diversity patterns of the Algerian whip snake Hemorrhois algirus. Master Thesis, University of Porto, Porto, Portugal. Herpetozoa 38: 191-204 (2025) Adhikari P, Shin M-S, Jeon J-Y, Kim HW, Hong S, Seo C (2018) Po- tential impact of climate change on the species richness of subal- pine plant species in the mountain national parks of South Korea. Journal of Ecology and Environment 42: 36. https://doi.org/10.1186/ s41610-018-0095-y Aiello-Lammens ME, Boria RA, Radosavljevic A, Vilela B, Ander- son RP (2015) spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 38(5): 541-545. https://doi.org/10.1111/ecog.01132 Anderson BG, Borns Jr HW (1997) The ice age world. An introduction to quaternary history and research with emphasis on North America and Europe during the last 2.5 million years. Scandinavian Univer- sity Press, Oslo, Norway, 208 pp. Anderson RP, Lew D, Peterson AT (2003) Evaluating predictive models of species’ distributions: Criteria for selecting optimal models. Eco- logical Modelling 162(3): 211-232. https://doi.org/10.1016/S0304- 3800(02)00349-6 Angert AL, Schemske DW (2005) The evolution of species’ distribu- tions: Reciprocal transplants across the elevation ranges of Mimulus cardinalis and M. lewisii. Evolution 59(8): 1671-1684. https://doi. org/10.1111/J.0014-3820.2005.TB0O1817.X Araujo MB, New M (2007) Ensemble forecasting of species distri- butions. Trends in Ecology & Evolution 22(1): 42-47. https://doi. org/10.1016/j.tree.2006.09.010 Boback SM, Nafus MG, Yackel Adams AA, Reed RN (2020) Use of vi- sual surveys and radiotelemetry reveals sources of detection bias for a cryptic snake at low densities. Ecosphere 11(1): e03000. https:// doi.org/10.1002/ecs2.3000 Bombi P, Capula M, D’Amen M, Luiselli L (2011) Climate change threatens the survival of highly endangered Sardinian populations of the snake Hemorrhois hippocrepis. Animal Biology 61(3): 239-248. https://doi.org/10.1163/157075511X584191 Brandt P, Alpers W, Backhaus JO (1996) Study of the generation and propagation of internal waves in the Strait of Gibraltar using a nu- merical model and synthetic aperture radar images of the European ERS 1 satellite. Journal of Geophysical Research: Oceans 101(C6): 14237-14252. https://doi.org/10.1029/96JC00540 Brun P, Zimmermann NE, Hari C, Pellissier L, Karger DN (2022) Glob- al climate-related predictors at kilometer resolution for the past and future. Earth System Science Data 14(12): 5573-5603. https://doi. org/10.5194/essd-14-5573-2022 Bulbul U, Ko¢ H, Bayrak MO, Kutrup B (2019) New locality record and morphological data of Hemorrhois ravergieri (Mén¢étries, 1832) (Serpentes: Colubridae) in Turkey. Turkish Journal of Bioscience and Collections 3(2): 59-62. https://doi.org/10.26650/tjbc.20190005 Carranza S, Arnold EN, Pleguezuelos JM (2006) Phylogeny, biogeog- raphy, and evolution of two Mediterranean snakes, Malpolon mon- spessulanus and Hemorrhois hippocrepis (Squamata, Colubridae), using mtDNA sequences. Molecular Phylogenetics and Evolution AQ(2): 532-546. https://doi.org/10.1016/j.ympev.2006.03.028 Chapman AD (2005) Principles and Methods of Data Cleaning. Primary Species and Species-Occurrence Data, v 1.0. Report for the Global Biodiversity Information Facility, Copenhagen, 75 pp. Cobos ME, Peterson AT, Barve N, Osorio-Olvera L (2019) kuenm: an R package for detailed development of ecological niche models using Maxent. PeerJ 7: e6281. https://doi.org/10.7717/peerj.6281 Cobos ME, Owens HL, Soberon J, Peterson AT (2024) Detailed multi- variate comparisons of environments with mobility oriented parity. 201 Frontiers of Biogeography 17: e132916. https://doi.org/10.21425/ fob.17.132916 Elith J, Phillips SJ, Hastie T, Dudik M, Chee YE, Yates CJ (2011) A statistical explanation of MaxEnt for ecologists: Statistical expla- nation of MaxEnt. Diversity and Distributions 17(1): 43-57. https:// doi.org/10.1111/).1472-4642.2010.00725.x Enriquez-Urzelai U, Martinez-Freiria F, Freitas I, Perera A, Martinez- Solano I, Salvi D, Velo-Anton G, Kaliontzopoulou A (2022) Allopat- ric speciation, niche conservatism and gradual phenotypic change in the evolution of European green lizards. Journal of Biogeography 49(12): 2193-2205. https://doi.org/10.1111/jbi.14497 Esparza-Estrada CE, Alencar LRV, Terribile LC, Rojas-Soto O, Yafiez-Arenas C, Villalobos F (2023) Vipers on the Scene: As- sessing the Relationship Between Speciation and Climatic Niche Evolution in Venomous Snakes (Reptilia: Viperidae). Evolutionary Biology 50: 264—273. https://do1.org/10.1007/s11692-023-09604-5 Eyring V, Bony S, Meehl GA, Senior CA, Stevens B, Stouffer RJ, Taylor KE (2016) Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geoscien- tific Model Development 9(5): 1937-1958. https://doi.org/10.5194/ gmd-9-1937-2016 Faraone FP, Melfi R, Di Nicola MR, Giacalone G, Valvo ML (2020) Phylogenetic relationships of the Italian populations of Horseshoe Whip snake Hemorrhois hippocrepis (Serpentes, Colubridae). Acta Herpetologica 15(2): 129-135. https://doi. org/10.13128/a_h-9058 Franklin J (2009) Mapping species distributions. Ecology, biodiversity and conservation. Cambridge University Press, Cambridge, 320 pp. https://doi.org/10.1017/CBO97805118 10602 Gavrilets S, LiH, Vose MD (2000) Patterns of parapatric speciation. Evo- lution 54(4): 1126-1134. https://doi.org/10.1111/j.0014-3820.2000. tb00548.x GBIF (2024) GBIF Occurrence Download. https://doi.org/10.15468/ dl.6jkmk9 Gomes VHF, IJff SD, Raes N, Amaral IL, Salom&o RP, de Souza Coelho L, et al. (2018) Species Distribution Modelling: Contrasting pres- ence-only models with plot abundance data. Scientific Reports 8(1): 1003. https://do1.org/10.1038/s41598-017-18927-1 Guisan A, Thuiller W, Zimmermann NE (2018) Habitat suitability and distribution models: with applications in R. Cambridge University Press, 477 pp. https://doi.org/10.1017/9781139028271 Gul S (2019) Is there an ecological barrier between allopatric spotted newts (Neurergus strauchii and Neurergus crocatus) in Turkey? A view to the glacier mountains of Hakkari. In: Salamanders: Habitat, Behavior and Evolution. Nova Science Publishers, New York, 63-83. Hurvich CM, Tsai C-L (1989) Regression and time series model se- lection in small samples. Biometrika 76(2): 297-307. https://doi. org/10.1093/biomet/76.2.297 iNaturalist (2025) Observations of Hemorrhois species. https://www. inaturalist.org [on 4 January 2025] Jablonski D, Mebert K, Masroor R, Simonov E, Kukushkin O, Abdu- raupov T, Hofmann S (2024) The Silk roads: Phylogeography of Central Asian dice snakes (Serpentes: Natricidae) shaped by rivers in deserts and mountain valleys. Current Zoology 70(2): 150-162. https://do1.org/10.1093/cz/zoad008 Karger DN, Conrad O, Bohner J, Kawohl T, Kreft H, Soria-Auza RW, Zimmermann NE, Linder HP, Kessler M (2017) Climatologies at herpetozoa.pensoft.net 202 Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois high resolution for the earth’s land surface areas. Scientific Data 4: 170122. https://doi.org/10.1038/sdata.2017.122 Kazemi SM, Jahan-Mahin MH, Mohammadian-Kalat T, Hosseinzadeh MS, Weinstein SA (2023) Local envenoming by the coinsnake or Asian racer, Hemorrhois nummifer and mountain racer or leopard snake, Hemorrhois ravergieri (Serpentes: Colubridae, Colubrinae) in Iran: a reminder of the importance of species identification in the medical management of snakebites. Toxicon 226: 107070. https:// doi.org/10.1016/j.toxicon.2023.107070 Kearney M, Porter W (2009) Mechanistic niche modelling: combining physiological and spatial data to predict species’ ranges. Ecology Let- ters 12: 334—350. https://doi.org/10.1111/j.1461-0248.2008.01277.x Li Y, Wang Y, Zhao C, Du X, He P, Meng F (2024) Predicting the spatial distribution of three Ephedra species under climate change using the MaxEnt model. Heliyon 10: e32696. https://doi.org/10.1016/j. heliyon.2024.e32696 Liz AV, Rédder D, Goncalves DV, Velo-Antén G, Fonseca MM, Ge- niez P, ..., Brito JC (2021) The role of Sahara highlands in the diversification and desert colonization of the Bosc’s fringe-toed lizard. Journal of Biogeography 48(11): 2891-2906. https://doi. org/10.1111/jbi.14250 Luiselli L, Capula M, Rugiero L, Salvi D, Akani GC (2012) Does in- terspecific competition with a stronger competitor explain the rarity of an endangered snake on a Mediterranean island? Ecological Re- search 27(3): 649-655. https://doi.org/10.1007/s11284-012-0936-6 Luna S, Pefia-Peniche A, Mendoza-Alfaro R (2024) Species distribution model accuracy is strongly influenced by the choice of calibration area. Biodiversity Informatics 18: 43-55. https://doi.org/10.17161/ bi.v181.22655 Machado L, Harris DJ, Salvi D (2021) Biogeographic and demographic history of the Mediterranean snakes Malpolon monspessulanus and Hemorrhois hippocrepis across the Strait of Gibraltar. BMC Ecology and Evolution 21: 210. https://doi.org/10.1186/s12862-021-01941-3 Mammola S, Arnedo MA, Pantini P, Piano E, Chiappetta N, Isaia M (2018) Ecological speciation in darkness? Spatial niche partitioning in sibling subterranean spiders (Araneae: Linyphiidae: Troglohy- phantes). Invertebrate Systematics 32(5): 1069-1082. Manel S, Williams HC, Ormerod SJ (2001) Evaluating presence—ab- sence models in ecology: the need to account for prevalence. Journal of Applied Ecology 38(5): 921-931. https://doi.org/10.1046/j.1365- 2664 .2001.00647.x McCormack JE, Zellmer AJ, Knowles LL (2010) Does niche divergence accompany allopatric divergence in Aphelocoma jays as predicted under ecological speciation?: Insights from tests with niche mod- els. Evolution 64(5): 1231-1244. https://doi.org/10.1111/j.1558- 5646.2009.00900.x Montes E, Feriche M, Alaminos E, Pleguezuelos JM (2020a) The Horseshoe whip snake (Hemorrhois hippocrepis) on Ibiza: predator release in an invasive population. Amphibia-Reptilia 42(2): 249- 254. https://dot.org/10.1163/15685381-bjal0039 Montes E, Feriche M, Ruiz-Sueiro L, Alaminos E, Pleguezuelos JM (2020b) Reproduction ecology of the recently invasive snake Hem- orrhois hippocrepis on the island of Ibiza. Current Zoology 66(4): 363-371. https://doi.org/10.1093/cz/z0z059 Montes E, Gallo-Barneto R, Cabrera-Pérez MA (2021) Presence of the horseshoe whip snake (Hemorrhois hippocrepis) on Gran Canaria, Spain. Boletin de la Asociacion Herpetologica Espafiola 32(1): 97-100. herpetozoa.pensoft.net Morales-Castilla I, Olalla-Tarraga MA, Bini LM, De Marco Jr P, Hawkins BA, Rodriguez MA (2011) Niche conservatism and species richness patterns of squamate reptiles in eastern and southern Africa: Squamate richness patterns in Africa. Aus- tral Ecology 36(5): 550-558. https://doi.org/10.1111/).1442- 9993.2010.02186.x Muscarella R, Galante PJ, Soley-Guardia M, Boria RA, Kass JM, Uri- arte M, Anderson RP (2014) ENM eval: An R package for conduct- ing spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods in Ecol- ogy and Evolution 5(11): 1198-1205. https://doi.org/10.1111/2041- 210X.12261 Nakazato T, Warren DL, Moyle LC (2010) Ecological and geographic modes of species divergence in wild tomatoes. American Journal of Botany 97(4): 680-693. https://do1.org/10.3732/ajb.0900216 Nunes LA, Raxworthy CJ, Pearson RG (2022) Evidence for ecological processes driving speciation among endemic lizards of Madagascar. Evolution 76(1): 58-69. https://doi.org/10.1111/evo.14409 Pearson RG (2007) Species’ distribution modeling for conservation educators and practitioners. Synthesis. Lessons in Conservation 3: 54-89. https://doi.org/10.5531/cbe.linc.3.1.3 Pelegrin N, Winemiller KO, Vitt LJ, Fitzgerald DB, Pianka ER (2021) How do lizard niches conserve, diverge or converge? Further explo- ration of saurian evolutionary ecology. BMC Ecology and Evolution 21: 1-13. https://doi.org/10.1186/s12862-021-01877-8 Peterson AT, Papes M, Soberon J (2008) Rethinking receiver operat- ing characteristic analysis applications in ecological niche model- ing. Ecological Modelling 213(1): 63-72. https://doi.org/10.1016/). ecolmodel.2007.11.008 Peterson AT, Soberon J, Pearson RG, Anderson RP, Martinez-Meyer E, Nakamura M, Araujo MB (2011) Ecological niches and geographic distributions (MPB-49). Princeton University Press, 328 pp. https:// doi.org/10.23943/princeton/978069 1 136868.001.0001 Peterson AT, Papes M, Soberon J (2015) Mechanistic and correlative models of ecological niches. European Journal of Ecology 1(2): 28-38. https://doi.org/10.1515/eje-2015-0014 Phillips SJ, Dudik M (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evalua- tion. Ecography 31(2): 161-175. https://doi.org/10.1111/j.0906- 7590.2008.5203.x Phillips SJ, Dudik M, Elith J; Graham CH, Lehmann A, Leathwick J, Ferrier S (2009) Sample selection bias and presence-only dis- tribution models: implications for background and pseudo-ab- sence data. Ecological Applications 19(1): 181-197. https://doi. org/10.1890/07-2153.1 Pleguezuelos JM, Feriche M (2002) Coluber hippocrepis Linnaeus, 1758. Culebra de herradura. In: Pleguezuelos JM, Marquez R, Li- zana M (Eds) Atlas y libro rojo de los anfibios y reptiles de Espafia, AHE-MMaA, Madrid, 266-268. Pomara LY, LeDee OE, Martin KJ, Zuckerberg B (2014) Demograph- ic consequences of climate change and land cover help explain a history of extirpations and range contraction in a declining snake species. Global Change Biology 20(7): 2087-2099. https://doi. org/10.1111/gceb.12510 QGIS. org (2025). QGIS Geographic Information System. Open Source Geospatial Foundation Project; QGIS Association, Switzerland. http://www.qgis.org Herpetozoa 38: 191-204 (2025) R Core Team (2024) R: A language and environment for statistical computing, R foundation for statistical computing, Vienna. https:// www.t-project.org/ Radosavljevic A, Anderson RP (2014) Making better Maxent mod- els of species distributions: complexity, overfitting and eval- uation. Journal of Biogeography 41(4): 629-643. https://doi. org/10.1111/jb1.12227 Rato C, Harris DJ, Perera A, Carvalho SB, Carretero MA, Rodder D (2015) A combination of divergence and conservatism in the niche evolution of the Moorish gecko, Zarentola mauritanica (Gekko- ta: Phyllodactylidae). PLoS ONE 10(5): e0127980. https://doi. org/10.1371/journal.pone.0127980 Rodriguez-Ruiz R, Juarez-Agis A, Garcia-Sanchez S, Olivier-Salome B, Zeferino-Torres J, Rivas-Gonzalez M (2020) Modelling the cur- rent and future potential distribution of Maconellicoccus hirsutus (Green, 1908) a pest of importance for Mexico. Agro Productividad 13(8): 47-52. https://doi.org/10.32854/agrop.vi.1653 Romero-Iraola I, Freitas I, Jiménez-Ruiz Y, Geniez P, Garcia-Paris M, Martinez-Freiria F (2023) Phylogeographic and paleoclimatic mod- elling tools improve our understanding of the biogeographic his- tory of Hierophis viridiflavus (Colubridae). Animals 13(13): 2143. https://do1.org/10.3390/ani13132143 Salvi D, Mendes J, Carranza S, Harris DJ (2018) Evolution, bio- geography and systematics of the western Palaearctic Zame- nis ratsnakes. Zoologica Scripta 47(4): 441-461. https://doi. org/10.1111/zsc.12295 Sexton JP, McIntyre PJ, Angert AL, Rice KJ (2009) Evolution and ecol- ogy of species range limits. Annual Review of Ecology, Evolution and Systematics 40: 415-436. https://doi.org/10.1146/annurev.ecol- sys.110308.120317 Soberon J, Peterson AT (2005) Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodiversity In- formatics 2: 1-10. https://doi.org/10.17161/bi.v210.4 Sun C, Zhu L, Liu Y, Wei T, Guo Z (2022) CMIP6 model simulation of concurrent continental warming holes in Eurasia and North Amer- ica since 1990 and their relation to the Indo-Pacific SST warming. Global and Planetary Change 213: 103824. https://doi.org/10.1016/j. gloplacha.2022.103824 Sahin MK, Candan K, Karakasi D, Lymberakis P, Poulakakis N, Kumlutas Y, Yildirim E, Ilgaz © (2022) Ecological niche dif- ferentiation in the Anatolian rock lizards (Genus: Anatololacer- ta) (Reptilia: Lacertidae) of the Anatolian Peninsula and Ae- gean Islands. Acta Herpetologica 17(2): 165-175. https://doi. org/10.36253/a_h-13089 Sahin MK (2024) The potential range and future distribution of the en- dangered lizard Darevskia clarkorum in the Caucasus Biodiversity Hotspot under climate change scenarios. Zoology in the Middle East 70(1): 1-11. https://doi.org/10.1080/09397 140.2024.2314338 Vaissi S, Kurnaz M, Sahin MK, Hernandez A (2023) Climatic niche di- vergence and conservatism promote speciation in snake-eyed skinks (Sauria: Scincidae): New insight into the evolution and diversifica- tion of Ablepharus species. Evolutionary Biology 50(2): 249-263. https://doi.org/10.1007/s11692-023-09603-6 Vaissi S, Mohammadi A (2024) Climate-driven distribution shifts of Iranian amphibians and identification of refugia and hotspots for effective conservation. Scientific Reports 14(1): 31610. https://doi. org/10.1038/s41598-024-79293-3 203 Vaissi S, Sahin MK, Kurnaz M (2024) Niche dynamics and climate change sensitivity in western Palearctic Zamenis ratsnakes (Rep- tilia: Colubridae). Amphibia-Reptilia 46(1): 33-50. https://doi. org/10.1163/15685381-bjal0209 Veverkova B (2021) Ecological effects of climate change on snakes. Bachelor Thesis, Charles University, Prague, Czechia. Warren DL, Matzke NJ, Cardillo M, Baumgartner JB, Beaumont LJ, Turelli M, Glor RE, Huron NA, Simées M, Iglesias TL, Piquet JC, Dinnage R (2021) ENMTools 1.0: an R package for comparative ecological biogeography. Ecography 44(4): 504-511. https://doi. org/10.1111/ecog.05485 Wiens JJ (2011) The niche, biogeography and species interactions. Phil- osophical Transactions of the Royal Society B: Biological Sciences 366: 2336-2350. https://doi.org/10.1098/rstb.2011.0059 Wiens JJ, Graham CH (2005) Niche conservatism: integrating evolu- tion, ecology, and conservation biology. Annual Review of Ecology, Evolution, and Systematics 36(1): 519-539. https://doi.org/10.1146/ annurev.ecolsys.36.102803.095431 Winter M, Fiedler W, Hochachka WM, Koehncke A, Meiri S, De La Riva I (2016) Patterns and biases in climate change research on amphibians and reptiles: a systematic review. Royal Society Open Science 3(9): 160158. https://doi.org/10.1098/rsos. 160158 Supplementary material 1 Raw species occurrence records of Hemorrhois species from literature, online source databases, and personal trips Author: Mehmet Kursat Sahin Data type: docx Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/ odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for oth- ers, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl1 Supplementary material 2 Mobility-Oriented Parity (MOP) analysis for projected distribution of Hemorrhois algirus under future climate conditions Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/li- censes/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl2 herpetozoa.pensoft.net 204 Mehmet Kutrsat Sahin: Ecological niches and climate responses in Hemorrhois Supplementary material 3 Mobility-Oriented Parity (MOP) analysis for projected distribution of Hemorrhois hippocrepis under future climate conditions Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/li- censes/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl3 Supplementary material 4 Mobility-Oriented Parity (MOP) analysis for projected distribution of Hemorrhois nummifer under future climate conditions Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/ odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for oth- ers, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl4 Supplementary material 5 Mobility-Oriented Parity (MOP) analysis for projected distribution of Hemorrhois ravergieri under future climate conditions Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/ odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for oth- ers, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl5 herpetozoa.pensoft.net Supplementary material 6 Species range change (SRC) of Hemorrhois species in recently suitable habitats Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/li- censes/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl6 Supplementary material 7 Results of identity tests for each Hemorrhois species Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/li- censes/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl7 Supplementary material 8 Results of asymmetric and symmetric background similarity tests for each parapatric Hemorrhois species Author: Mehmet Kursat Sahin Data type: tiff Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/li- censes/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://do1.org/10.3897/herpetozoa.38.e151017.suppl8