*3.2. Future Habitat Distribution*

We simulated the potential distribution of Siberian ibex under three greenhouse scenarios for two periods. The performance of the six eSDMs was good (Table 2 and Supplementary Materials Table S3), in which the TSS value range was 0.78–0.81 (average of 0.793), and the AUC value range was 0.94–0.95 (average of 0.947). In these scenarios, temperature seasonality, mean temperature of the wettest quarter, and precipitation of the wettest month were found to be the main climate factors affecting the distribution of Siberian ibex, which will be driving the distribution of the Siberian ibex further in this region. In addition, HII, NDVI, and elevation will also affect the distribution of ibex. Human Influence Index (HII), the indicator of human impact, is also a main factor that will influence the distribution of ibex, since a high HII is not good for the survival of the ibex.


**Table 2.** Changes in suitable habitat of Siberian ibex in Taxkorgan Nature Reserve under different climate change scenarios.

Our models indicated a tendency for a decrease in the suitable habitat of Siberian ibex in the TNR under the future climate scenarios (RCP2.6, RCP4.5, and RCP8.5 for both 2050 and 2070), while the magnitude of the reduction depends on the emission scenario, although most of the changes do not appear to be spatially substantial (Figure 3). The average decline of suitable habitat predicted in the future is 14.76%, when compared to the present. Under the RCP2.6 for 2070, the suitable habitat decreased the most, reaching 26.55% (717.4 km2) in some models. Moreover, there was no obvious change in the distribution of Siberian ibex along the altitudinal gradient under climate change (Figure 4). However, we found that the centroid of Siberian ibex distribution tended to move north under different climate scenarios (Figure 5).

**Figure 3.** The habitat suitability of Siberian ibex in Taxkorgan Nature Reserve under three greenhouse gas emissions scenarios for two periods; top panel predictions for 2050 and bottom panels for 2070. (**a**) RCP 2.6 for 2050; (**b**) RCP 4.5 for 2050; (**c**) RCP 8.5 for 2050; (**d**) RCP 2.6 for 2070; (**e**) RCP 4.5 for 2070; (**f**) RCP 8.5 for 2070.

**Figure 4.** The proportion of suitable habitat area for Siberian ibex at different elevation gradients in Taxkorgan Nature Reserve under current and future climate scenarios.

**Figure 5.** Centroid distribution of suitable habitat for Siberian ibex in Taxkorgan Nature Reserve under current and future climate scenarios. The average shift to the north is 5.81 km when the latitude difference is taken into account, and 10.20 km in the RCP2.6 for 2050.

We found that the northwest of the reserve was the most stable suitable habitat for Siberian ibex under any climate scenario (Figure 6). The area can act as a climate change refuge for Siberian ibex (Supplementary Materials Figure S2), which is the overlapping area of current suitable habitat and future stable suitable habitat. The region with major losses of current suitable habitat was mainly located in the southern part of the reserve, and the area with an increase was mainly located in the central part of the reserve. The loss rate of suitable habitat was highest under the scenarios of RCP2.6 for 2070 and RCP8.5 for 2050, reaching 35.11% and 30.86% loss, respectively. However, we found that the lost suitable habitat and the newly gained suitable habitat were not concentrated in low or high elevations under different climate scenarios. At the same time, we also found that the stable suitable habitats were mainly located in the northwest and a small part of the

northeast of the reserve. Under RCP4.5 for 2070, the area of stable suitable habitat was the largest at 2577.88 km2.

**Figure 6.** Transfer of suitable habitat for Siberian ibex between current and future climate scenarios in Taxkorgan Nature Reserve. (**a**) RCP 2.6 for 2050; (**b**) RCP 4.5 for 2050; (**c**) RCP 8.5 for 2050; (**d**) RCP 2.6 for 2070; (**e**) RCP 4.5 for 2070; (**f**) RCP 8.5 for 2070.

### **4. Discussion**

Numerous studies have confirmed the profound influence of climate warming on the change in distribution of ungulates [15,63,64]. Such change could be deleterious, leading to a reduction in species abundance or even extinction as habitats deteriorate [65,66]. In the Apennines, Italy, the Apennine chamois may undergo a drastic decline in its historical core range in the next 50 years with a 95% reduction or near-extinction at worst [67]. The Taxkorgan Nature Reserve, located in the Eastern Pamir, is home to several ungulates, including the Siberian ibex. In the present study, we, for the first time, forecast the potential habitat shifts/loss of Siberian ibex in the Eastern Pamir under climate change. We showed that climate change and human disturbance have combined negative impacts on the distribution of Siberian ibex.

Suitable habitat for Siberian ibex currently accounts for only 16.6% of the reserve. Suitable habitat would further contract in the future because any newly gained suitable habitat would be smaller than the habitat being lost. The extent of habitat loss could be up to 29.15% and 27.99% at most by 2050 and 2070, respectively, depending on the climate change scenarios we used in our models. This is partially because the Pamirs are a complex ecosystem that is highly vulnerable to climate change. For example, the glaciers in the Eastern Pamirs that play a crucial role in the water cycle in high elevation areas [68] continue to accelerate their retreat [69] due to a significant upward trend in temperature and precipitation [70]. Meanwhile, reliable water sources remain crucial for the survival of animals [71,72] and the survival and distribution of plant communities. In fact, our models confirmed that the distribution of Siberian ibex in this region is strongly affected by precipitation (precipitation seasonality and the precipitation of the wettest month) and temperature (temperature seasonality). This is consistent with the results for Siberian ibex in Tajikistan [29] and in Pakistan [73]. The rise in temperature will likely cause Siberian ibex and other alpine ungulates to show behavioral and physiological responses, such as

heat avoidance and heat stress, leading them to move to higher elevations or seek shade in the short term [16]. Furthermore, the precipitation of the wettest months may affect the distribution of ibex via the change in the phenology and richness of plants [17,74]. In addition, we found that Siberian ibex prefer to inhabit areas with high NDVI, i.e., where there is high vegetation cover allowing species to acquire food without moving too far [29,75]. Ruggedness is also a vital factor that influences the distribution of Siberian ibex in the TNR. The rugged terrain helps ungulates avoid predators [29,76]. In our study, the probability of the presence of ibex decreases accordingly when HII is greater than 15. A higher HII value means more serious human disturbance, which means ibex prefer to inhabit areas with low human disturbance.

Much wildlife has shifted its geographic distribution toward higher elevations or latitudes [5,67,77–79], and Chen et al.'s (2011) meta-analysis showed that the distribution of many species has recently shifted to higher elevations at a median rate of 11.0 m per decade [9]. Interestingly, we have not found such a shift to higher elevations, i.e., the distribution of Siberian ibex at different elevations remained unchanged under future climate scenarios. Similarly, no change in elevation was found for most ungulates in the Tibetan plateau [63]. Non-elevational shifts in those ungulates may be the result of physiological temperature thresholds and precipitation tolerance on the one hand. On the other hand, due to the high elevation in the plateau, there are very few areas available at high elevation for most ungulate species to colonize near their range boundaries and upward migration is thus limited [63,80,81]. As part of the Tibetan Plateau, the Pamir plateau may put the same stressors on wildlife, meaning that Siberian ibex may not have any additional room to climb to higher elevations. Furthermore, another study on ungulates, including Tibetan antelope *Pantholops hodgsoni*, Kiang *Equus kiang*, and wild yak *Bos mutus* living on the Tibetan Plateau, revealed that the distribution of the main forage plants will be reduced by more than 50% in response to climate change [64]. Indeed, research by Walther, et al. [82] showed that there has been an upward shift in alpine plants. Meanwhile, the reduction in mountain surface area with rising elevation [83] may not provide Siberian ibex with contiguous patches of habitat. Therefore, the Siberian ibex may simply run out of space to move to.

In our study, human disturbance may further contribute to the non-elevational shift of Siberian ibex. In the TNR, the lower elevation region experiences strong human disturbances, including settlements and infrastructure (HII in our model) [36], where Tajik herding people graze their livestock [38]. Meanwhile, there is strong food competition between Siberian ibex and domestic sheep, and Siberian ibex show strong avoidance behavior towards domestic sheep, which reduces their range of activity [84]. Thus, a series of human disturbances forces Siberian ibex to abandon potential suitable habitats at lower elevations leading to a reduction in the vertical distribution range of ungulates [85], which is similar to the effects found in primates and Walia ibex (*Capra walie*) [85–87].

In addition to responding to climate change by moving to higher elevations, species may also move to higher latitudes to acquire new habitats [14,88]. Global species distribution shifts under climate change are estimated to move toward higher latitudes at a rate of 16.9 km/decade [5,9,14]. By analyzing the centroid of suitable habitat for Siberian ibex at present and the centroid under different climate scenarios, we found that Siberian ibex in our study area showed a trend of moving northward under future climate scenarios. This is consistent with the study of Marco Polo sheep [22] and ibex [29] on the Pamir plateau in Tajikistan, which have the same shift trend in response to climate change. Similarly, research on Britain's mammals has shown that they have shifted 22 km north in the past 25 years [14]. In addition, previous studies have found that birds in different temperature zones have different distribution mechanisms in response to climate change [89]. Birds from the northern temperate regions migrate to higher latitudes [90–92], while those from the tropics migrate to higher elevations [89,93]. In China, Tibetan antelope and goitered gazelle (*Gazella subgutturosa*), both located in the northern temperate zone, also showed a trend of moving northward under climate change [17]. Although we have no evidence that

all animals living in the north temperate zone exhibit poleward migration in response to climate change, our results suggest that this could be one of the strategies of Siberian ibex in response to climate change.

It is a crucial step from theory to practice to put forward conservation management opinions based on study results. According to our study, the northwest part of the reserve may be a refuge which will promote Siberian ibex survival under increasingly adverse climatic conditions. This area should therefore be protected in the future. The key conservation issue in this area is finding an effective way for Siberian ibex to survive under climate change. Meanwhile, reducing human activity and grazing in and around the current suitable habitat will help the Siberian ibex obtain more habitat. In addition, the Siberian ibex is an ungulate widely distributed in Central Asia [25]. Identifying and conserving the ecological corridor linking suitable habitat in the reserve and outside the reserve (including domestic and overseas) will help support gene exchange in the population and promote genetic diversity.

Our study revealed that climate change has a strong effect on the distribution of Siberian ibex in the Eastern Pamir, China. One of the expected effects of recent warming is that it may force animals to move to higher, cooler elevations [94]. Siberian ibex generally show seasonal movement along an elevational gradient, i.e., they migrate to higher elevations in summer and to lower elevations in winter due to phenological changes and variable temperatures at high and low elevations. However, intensive human interference at lower elevations may prevent Siberian ibex from migrating to lower elevations, further limiting their distribution and making the range of their distribution along elevation even narrower [95]. Such distribution patterns may reduce the resistance of Siberian ibex to environmental changes. Although we are incapable of further analyzing whether and how interspecific interactions will alter the distribution of the species due to the lack of data on natural predators (such as snow leopards), domestic animals, and the feeding habits of the species in the region, our research is nevertheless conducive to the conservation of ungulates in the plateau ecosystem. Future studies should focus on the effects of intraand interspecific interactions on species distribution, and the construction of ecological corridors should be considered to better protect species.

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/d14090750/s1, Figure S1: Response curves of the current ensemble species distribution models for Siberian ibex in Taxkorgan Nature Reserve to each variable, Figure S2: Stable suitable habitat for Siberian ibex in Taxkorgan Nature Reserve under climate change; Table S1: The number of line transects per month for each year (2018–2021) during the survey, Table S2: Ranking of variables' importance affecting the distribution of Siberian ibex in Taxkorgan Nature Reserve, Table S3: The evaluation of ensemble species distribution models (eSDMs, including AUC and TSS) of Siberian ibex in Taxkorgan Nature Reserve under current and different climate change scenarios.

**Author Contributions:** Conceptualization, Y.Z., M.W. and W.Y.; methodology, Y.Z. and M.W.; software, Y.Z.; validation, M.W. and W.Y.; formal analysis, Y.Z. and M.W.; investigation, M.W. and W.Y.; resources, Y.Z., M.W. and W.Y.; data curation, Y.Z., M.W. and W.Y.; writing—original draft preparation, Y.Z.; writing—review and editing, M.W., B.Z., K.E.R., A.A.d.S., W.Y. and J.A.; visualization, Y.Z.; supervision, M.W. and W.Y.; project administration, M.W. and W.Y.; funding acquisition, M.W., B.Z., A.A.d.S., W.Y. and J.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the Western Young Scholar Program-B of the Chinese Academy of Sciences (grant numbers 2021-XBQNXZ-014); the Second Tibetan Plateau Scientific Expedition and Research Program (STEP, grant numbers 2019QZKK0501); Yunnan Applied Basic Research Projects (No. 202101AT070296 to B.Z.) National Natural Science Foundation of China (No. 32101408 to B.Z.), and the Shanghai cooperation organization partnership and international technology cooperation plan of science and technology projects (2021E01020). J.A. and A.A.d.S. were supported by the R&D Unit Centre for Functional Ecology—Science for People and the Planet (CFE), with reference UIDB/04004/2020, financed by FCT/MCTES through national funds (PIDDAC) and by TERRA Associate Laboratory (LA/P/0092/2020).

**Institutional Review Board Statement:** Not applicable because the study only involved long distance observation from the animals.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We thank the staff of Taxkorgan Nature Reserve for all their help during the field survey. We are grateful to the anonymous reviewers for their thoughtful comments on our manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**

