Global Potential Distribution of Sarcophaga dux and Sarcophaga haemorrhoidalis under Climate Change

: Climate change has a direct impact on biodiversity, affecting ecosystems and altering their balance. Many taxa, including insects, are likely to be affected by climate change in terms of geographic distribution. Sarcophagid ﬂies, such as Sarcophaga dux and Sarcophaga haemorrhoidalis , are important ﬂies because of their apparent ecological, forensic, and medical signiﬁcance. Global habitat suitability varies as a result of climate change. In wildlife management, models that predict species’ spatial distribution are being used more and more, which emphasizes the need for reliable methods to evaluate their accuracy. Consequently, the statistical robustness of maximum entropy was implemented in Maxent to model the current and future global distribution of both ﬂies, involving occurrence data of 155 and 87 points for S. dux and S. haemorrhoidalis , respectively. Based on the Pearson correlation and Jackknife test, ﬁve bioclimatic variables were used for current and future predictive models. For future models, two representative concentration pathways (RCPs), 2.6 and 8.5, for 2050 and 2070 were applied. Both statistical parameters, AUC and TSS, were used to assess the resulting models with values equal to 0.80 ( ± 0.01) and 0.9, respectively, for S. dux and equal to 0.86 ( ± 0.01) and 0.92 for S. haemorrhoidalis . The resulting models for S. dux showed high and very high suitability in Europe, Tropical Africa, India, Canada, the United States from Alaska to Florida, Brazil, and Australia. In the case of S. haemorrhoidalis Europe and North and South America displayed low to medium suitability, but North Africa, including Egypt; Tropical Africa; Asia, including Saudi Arabia, India, and China; and Australia showed increased suitability. Decision-makers are put in conﬂict with their duties to avert destruction in the economic, medical, and ecological sectors by such anticipated models, and use these predictive models as a cornerstone for building a control strategy for such forensically important ﬂies at local spatial scales.


Introduction
Sarcophagidae is a family of flies commonly known as flesh flies or Sarcophagid flies [1]. They are found worldwide and are typically medium to large in size, with a gray or black body and three longitudinal stripes on the thorax [1,2]. Flesh flies are a varied group of flies, with over 3000 species documented in the Sarcophagid family [3]. They range in size from medium to huge, with body lengths ranging from a few millimeters to more than a centimeter [4]. Flesh flies are important in recycling because they feed on dead and decomposing animal flesh, including human flesh [5]. They can also infest living animals, such as livestock wounds, and in some cases humans, in the very well-known phenomenon myiasis [4,5]. Furthermore, the presence and development of larvae of particular species of flesh flies are utilized in forensic entomology to establish the time of death of a cadaver [6]. The adults of some species have a very important role in pollination; some species are known to feed on nectar and pollen and can be important pollinators of certain plant species [7]. species records, in addition to a wide range of generated climatological future scenarios, help scientists predict how our biological future will be and determine what the risk is that our ecosystems will face in the near and far future due to global warming [31,32].
So, the aim of this work is to evaluate the climate change impact on two very widespread Sarcophagid species on the global scale and also to create maps for their current distribution range for the first time.

Input Data
Almost all available records of both S. dux and S. haemorrhoidalis were gathered from the literature, either as adults or larval cases [33][34][35][36][37][38]. Moreover, other distribution data from the digital database: The Global Biological Information Facility (GBIF.org https://doi.org/10.15468/dl.xp4www and https://doi.org/10.15468/dl.nbmjwu (accessed on 15 May 2023)) were used for both species. On these data, three filtration stages were carried out. The duplicate records were first deleted [39]. Additionally, data with high spatial uncertainty were removed [40]. Thirdly, to avoid any spatial redundancy, data were spatially rarefied using an ArcGIS algorithm [39,40]. Finally, the remaining georeferenced occurrence points were 155 and 87 points for S. dux and S. haemorrhoidalis, respectively. These points were converted into comma delimited (CSV) format and used to predict the current and future global distribution of S. dux and S. haemorrhoidalis ( Figure 1). several modeling tools and statistical methods and the availability of large am accumulative data from species records, in addition to a wide range of generated logical future scenarios, help scientists predict how our biological future will be termine what the risk is that our ecosystems will face in the near and far futur global warming [31,32].
So, the aim of this work is to evaluate the climate change impact on two ver spread Sarcophagid species on the global scale and also to create maps for their distribution range for the first time.

Input Data
Almost all available records of both S. dux and S. haemorrhoidalis were gather the literature, either as adults or larval cases [33][34][35][36][37][38]. Moreover, other distribut from the digital database: The Global Biological Information Facility (G https://doi.org/10.15468/dl.xp4www and https://doi.org/10.15468/dl.nbmjwu (acce 15 May 2023)) were used for both species. On these data, three filtration stages w ried out. The duplicate records were first deleted [39]. Additionally, data with hig uncertainty were removed [40]. Thirdly, to avoid any spatial redundancy, data w tially rarefied using an ArcGIS algorithm [39,40]. Finally, the remaining georefere currence points were 155 and 87 points for S. dux and S. haemorrhoidalis, resp These points were converted into comma delimited (CSV) format and used to pre current and future global distribution of S. dux and S. haemorrhoidalis ( Figure 1). About nineteen bioclimatic data with a spatial resolution of approximatel were downloaded from the WorldClim data (www.worldclim.org, (accessed on 2023)). Using ArcGIS version 10.7, fifteen bioclimatic variables for current bioclim were transformed into ASCII format to assess the most important biological fac contribute to the present model of habitat suitability for both S. dux and S. haemor [40]. Bioclimatic variables 8-9 and 18-19 were excluded from the analysis bec known geographic anomalies [41]. Person correlation function with a value equa ≥ 0.8) was used to reduce multicollinearity among the fifteen bioclimatic data [41,4 sequently, only five bioclimatic layers were used for further analysis.
For future predictions, we used parallel datasets from the WorldCli (www.worldclim.org, (accessed on 1 April 2023)) of two typical concentratio About nineteen bioclimatic data with a spatial resolution of approximately 5 km 2 were downloaded from the WorldClim data (www.worldclim.org, (accessed on 1 April 2023)). Using ArcGIS version 10.7, fifteen bioclimatic variables for current bioclimatic data were transformed into ASCII format to assess the most important biological factors that contribute to the present model of habitat suitability for both S. dux and S. haemorrhoidalis [40]. Bioclimatic variables 8-9 and 18-19 were excluded from the analysis because of known geographic anomalies [41]. Person correlation function with a value equal to (|r| ≥ 0.8) was used to reduce multicollinearity among the fifteen bioclimatic data [41,42]. Consequently, only five bioclimatic layers were used for further analysis.
For future predictions, we used parallel datasets from the WorldClim data (www.worldclim.org, (accessed on 1 April 2023)) of two typical concentration paths (RCPs),  [40]. These bioclimatic layers were also changed into ASCII format via ArcGIS v.10.7 [43]. Using the MRI-CGCM3 global climate model, the climatic data for the selected variables were then extended to the years 2050 and 2070 in order to assess how climate change may affect the appropriateness of S. dux and S. haemorrhoidalis habitat in the future. The most recent GCM climate estimates used in the IPCC Fifth Assessment Report contain this information. Moreover, gain and loss of habitat suitability ranges were calculated by subtracting the predicted future maps from the current one using map algebra function implemented in ArcGIS v.10.7 to produce gain and loss maps. Finally, we calculated the predicted areas in kilometers between current and future predictive models via ArcGIS v. 10.7 (Table S1).

Environmental Niche Modeling
The ecological niche and habitat suitability of S. dux and S. haemorrhoidalis were assessed using the maximum entropy method in Maxent v3.4.3e [44]. The statistical power of this method resulted in robust predictive models [45]. It is widely used due to the capability to create models using only presence records [46]. In modeling studies with small sample numbers, it can also get rid of duplicate data in the same cell [47]. For each of our models, 25% of the records were used for testing, and 75% were used for training [45][46][47]. The maximum background points and iterations permitted were ten thousand and one thousand, respectively [48]. Additionally, a 10-fold cross-validation was carried out, which improved the performance of the model [47,48].

Model Robustness
The robustness of the resulting models is typically evaluated using the Area Under Curve (AUC) [34]. AUC values above 0.9 are considered to represent excellent model performance, and they range from 0.5 to 1.0 [45]. In order to assess the distribution of the target species, important bioclimatic parameters were found using the Jackknife test [46]. The expected model accuracy was also measured using the True Skill Statistic (TSS) [44]. The TSS value can be anywhere between −1 and 1, with close values to 1 indicating a strong link and values to −1 indicating a weak relationship between the distribution and the predictive model [47,49].

Model Evaluation and Contribution of Environmental Covariates
For both species, the Maxent predictive model achieved a high AUC values of 0.80 (±0.01) for S. dux and 0.86 (±0.01) for S. haemorrhoidalis. The AUC values were often higher in continuous species distribution modeling (SDM) than discontinuous SDM. Additionally, the TSS value was quite high, with values equal to 0.9 and 0.92 for S. dux and S. haemorrhoidalis, respectively, indicating outstanding model performance. TSS readings over 0.5 are often considered acceptable. The percentage contribution of the bioclimatic variables was demonstrated by the Jackknife test of the predictive model (Figure 2a,b, Table 1) . In the case of S. dux, the annual mean temperature (bio 1) was the most effective bioclimatic covariate followed by the min temperature of the coldest month (bio 6), the annual precipitation (bio 12), the mean diurnal range (bio 2), and the precipitation of the driest month (bio 14), respectively. But, for S. haemorrhoidalis, the (bio 1) was the most effective bioclimatic covariate followed by (bio 12), (bio 6), (bio 2), and (bio 14), respectively. Furthermore, the range of the response curves for (bio 1), the most significant environmental covariate, was from 5 • C to 20 • C and from 10 • C to 30 • C for S. dux and S. haemorrhoidalis, respectively (Figure 2c,d).

Two-Dimensional Niche Analysis
The enveloped test was used to generate the two-dimension niche for both S. dux and S. haemorrhoidalis. The most effective bioclimatic variables used in studying these insects are as follow: the annual mean temperature (bio 1) and the annual precipitation (bio 12) (Figure 2e,f). The results indicate the impressive adaptability of both species for different environmental conditions. In S. dux, the annual mean temperature can range from 10 °C to 27 °C and the amount of rain can range from 300 mm to 2600 mm per year. In S. haemorrhoidalis, the annual mean temperature can range from 14 °C to 28 °C and the amount of rain can range from 0 mm to 2700 mm per year.

Current Habitat Suitability Models
The predictive current global distribution of S. dux and S. haemorrhoidalis agrees with the present-day occurrence of both flies. Our predictive maps showed that S. dux and S. haemorrhoidalis are found in different parts of the world (Figure 3a,b). In Europe, S. dux showed very high suitability from Portugal and Spain in the east, along with France, the UK, Germany, and Italy, to Ukraine and the eastern borders of Russia in the west. While S. haemorrhoidalis showed medium suitability, especially in France, Germany, Italy, and Spain (Figure 3a,b). Asia, India, Thailand, Vietnam, and southern parts of Saudi Arabia showed high and very high suitability for S. dux, but China and Saudi Arabia showed very high suitability for S. haemorrhoidalis. Also, Malaysia, the Philippines, and Indonesia

Two-Dimensional Niche Analysis
The enveloped test was used to generate the two-dimension niche for both S. dux and S. haemorrhoidalis. The most effective bioclimatic variables used in studying these insects are as follow: the annual mean temperature (bio 1) and the annual precipitation (bio 12) (Figure 2e,f). The results indicate the impressive adaptability of both species for different environmental conditions. In S. dux, the annual mean temperature can range from 10 • C to 27 • C and the amount of rain can range from 300 mm to 2600 mm per year. In S. haemorrhoidalis, the annual mean temperature can range from 14 • C to 28 • C and the amount of rain can range from 0 mm to 2700 mm per year.

Current Habitat Suitability Models
The predictive current global distribution of S. dux and S. haemorrhoidalis agrees with the present-day occurrence of both flies. Our predictive maps showed that S. dux and S. haemorrhoidalis are found in different parts of the world (Figure 3a,b). In Europe, S. dux showed very high suitability from Portugal and Spain in the east, along with France, the UK, Germany, and Italy, to Ukraine and the eastern borders of Russia in the west. While S. haemorrhoidalis showed medium suitability, especially in France, Germany, Italy, and Spain (Figure 3a,b). Asia, India, Thailand, Vietnam, and southern parts of Saudi Arabia showed high and very high suitability for S. dux, but China and Saudi Arabia showed very high suitability for S. haemorrhoidalis. Also, Malaysia, the Philippines, and Indonesia showed medium habitat suitability in the case of S. dux and very high suitability in the case of S. haemorrhoidalis (Figure 3a,b). Tropical Africa showed medium to very high suitability, and North Africa, including Northern parts of Egypt, Algeria, Tunisia, and Morocco, showed high suitability for both species (Figure 3a,b). In North America, S. dux showed high and very high suitability across Canada and the United States, from Alaska to Florida, while S. haemorrhoidalis showed low suitability. Southern parts of Brazil and northern parts of Argentina showed high to very high suitability for S. dux, whereas Colombia, Venezuela, and Brazil showed high habitat suitability for S. haemorrhoidalis (Figure 3a,b). Generally, our models showed high and very high suitability for both fly species in the overall Australian territories, except the southern parts showed low suitability for S. haemorrhoidalis (Figure 3a,b). Finally, a niche overlapping map illustrated the common range shared by both flies (Figure S1), and changes in the areas in kilometers illustrated the differences between current and future global predictive models (Table S1).
FOR PEER REVIEW 6 of 12 showed medium habitat suitability in the case of S. dux and very high suitability in the case of S. haemorrhoidalis (Figure 3a,b). Tropical Africa showed medium to very high suitability, and North Africa, including Northern parts of Egypt, Algeria, Tunisia, and Morocco, showed high suitability for both species (Figure 3a,b). In North America, S. dux showed high and very high suitability across Canada and the United States, from Alaska to Florida, while S. haemorrhoidalis showed low suitability. Southern parts of Brazil and northern parts of Argentina showed high to very high suitability for S. dux, whereas Colombia, Venezuela, and Brazil showed high habitat suitability for S. haemorrhoidalis ( Figure  3a,b). Generally, our models showed high and very high suitability for both fly species in the overall Australian territories, except the southern parts showed low suitability for S. haemorrhoidalis (Figure 3a,b). Finally, a niche overlapping map illustrated the common range shared by both flies (Figure S1), and changes in the areas in kilometers illustrated the differences between current and future global predictive models (Table S1).

Future Habitat Suitability Models
The produced Maxent models for the global potential distributions of S. dux and S. haemorrhoidalis under future climate change scenarios RCP 2.6 and RCP 8.5 for the years 2050 and 2070 are illustrated in Figure 4. In Europe, our predictive models showed an ascending increase in habitat suitability in the future for S. dux from RCP 2.6 2050 to RCP 8.5 2070, especially in Russia (Figure 4a-d), while the habitat suitability for S. haemorrhoidalis showed no remarkable change in the future RCPs except RCP 8.5 2070, which assures an increase in suitability in Italy, Spain, and Turkey (Figure 4e-h). In Asia, future models of S. dux showed a decrease in suitability in Western Saudi Arabia, Iraq, Kuwait, and Iran (Figure 4a-d), while S. haemorrhoidalis showed an increase in habitat suitability in the west-

Future Habitat Suitability Models
The produced Maxent models for the global potential distributions of S. dux and S. haemorrhoidalis under future climate change scenarios RCP 2.6 and RCP 8.5 for the years 2050 and 2070 are illustrated in Figure 4. In Europe, our predictive models showed an ascending increase in habitat suitability in the future for S. dux from RCP 2.6 2050 to RCP 8.5 2070, especially in Russia (Figure 4a-d), while the habitat suitability for S. haemorrhoidalis showed no remarkable change in the future RCPs except RCP 8.5 2070, which assures an increase in suitability in Italy, Spain, and Turkey (Figure 4e-h). In Asia, future models of S. dux showed a decrease in suitability in Western Saudi Arabia, Iraq, Kuwait, and Iran (Figure 4a-d), while S. haemorrhoidalis showed an increase in habitat suitability in the western coast of Saudi Arabia (Figure 4e-h, Table S1). In Africa, the calibration maps of S. dux showed a loss in suitability in Zambia, Botswana, and Namibia and a gain in Niger and Mali (Figure 5a-d), but S. haemorrhoidalis showed a high loss of suitability in Northern Africa, including Egypt, Tunisia, and Morocco, and West Africa, including Mauritania and Senegal (Figure 5e-h). In North America, S. dux showed an increase in suitability in the eastern parts of the United States ( Figure  5a-d), whereas S. haemorrhoidalis showed a gain in suitability in the south of the United States (Figure 5e-h). South America showed an increase in suitability for S. haemorrhoidalis in Southern Brazil in RCP 8.5 2070. Finally, S. dux showed a loss of suitability in Australia; however, S. haemorrhoidalis showed a gain in middle Australia (Figure 5a-h) In Africa, the calibration maps of S. dux showed a loss in suitability in Zambia, Botswana, and Namibia and a gain in Niger and Mali (Figure 5a-d), but S. haemorrhoidalis showed a high loss of suitability in Northern Africa, including Egypt, Tunisia, and Morocco, and West Africa, including Mauritania and Senegal (Figure 5e-h). In North America, S. dux showed an increase in suitability in the eastern parts of the United States (Figure 5a-d), whereas S. haemorrhoidalis showed a gain in suitability in the south of the United States (Figure 5e-h). South America showed an increase in suitability for S. haemorrhoidalis in Southern Brazil in RCP 8.5 2070. Finally, S. dux showed a loss of suitability in Australia; however, S. haemorrhoidalis showed a gain in middle Australia (Figure 5a-h).

Discussion
Climate change endangers ecosystems and species distribution in a variety o including by causing severe droughts in some regions, more frequent and intense and floods in other regions, and habitat loss and fragmentation [50]. Moreover, change may provide new opportunities for invasive species to establish and thrive locations, potentially resulting in native species' displacement and changes in eco functions [51]. Furthermore, many species are predicted to face increased extinctio as a result of climate change, particularly those that are already threatened or have ranges [52]. For these reasons, the prediction of how species range will be in the f considered an important need by scientists and decision-makers to mitigate change or decrease its effect on our planet [53].
The present work forms a step in this direction, as it shows how the future wi two very important and widespread flesh flies on the global scale. The current distr of the two species is very compatible with their real situation on the ground (Figur S. dux is distributed through southern Asia, southern Europe, parts of the Midd sub-Saharan Africa, and southern North America. Also, many parts of South A show high habitat suitability for this species, although this has not been recorde before, and this information is very critical, as if S. dux accidentally enters this regi will create a very serious invasive species with a highly suitable habitat (Figure the other hand, S. haemorrhoidalis appears to have a large range through tropical a tropical habitats around the world (Figure 3b). Unlike its cousin, S. dux, this specie common in Europe or large parts of North America. The generated current map two species will be very helpful, especially for forensic sectors throughout the w they will help investigators identify these species' ranges.
The accuracy of the generated maps is considered to be very high, either for th test or the TSS test, for the two species, even when compared to several previous that have dealt with other dipterous species [39,43]. AUC and TSS are both helpful

Discussion
Climate change endangers ecosystems and species distribution in a variety of ways, including by causing severe droughts in some regions, more frequent and intense storms and floods in other regions, and habitat loss and fragmentation [50]. Moreover, climate change may provide new opportunities for invasive species to establish and thrive in new locations, potentially resulting in native species' displacement and changes in ecosystem functions [51]. Furthermore, many species are predicted to face increased extinction risks as a result of climate change, particularly those that are already threatened or have limited ranges [52]. For these reasons, the prediction of how species range will be in the future is considered an important need by scientists and decision-makers to mitigate climate change or decrease its effect on our planet [53].
The present work forms a step in this direction, as it shows how the future will be for two very important and widespread flesh flies on the global scale. The current distribution of the two species is very compatible with their real situation on the ground (Figure 3a,b). S. dux is distributed through southern Asia, southern Europe, parts of the Middle East, sub-Saharan Africa, and southern North America. Also, many parts of South America show high habitat suitability for this species, although this has not been recorded there before, and this information is very critical, as if S. dux accidentally enters this region, this will create a very serious invasive species with a highly suitable habitat (Figure 3a). On the other hand, S. haemorrhoidalis appears to have a large range through tropical and subtropical habitats around the world (Figure 3b). Unlike its cousin, S. dux, this species is not common in Europe or large parts of North America. The generated current maps of the two species will be very helpful, especially for forensic sectors throughout the world, as they will help investigators identify these species' ranges.
The accuracy of the generated maps is considered to be very high, either for the AUC test or the TSS test, for the two species, even when compared to several previous studies that have dealt with other dipterous species [39,43]. AUC and TSS are both helpful metrics for assessing the effectiveness of SDMs, but they each have unique advantages and disadvantages. Although the AUC is a widely used metric in the literature, it has drawn criticism for being impervious to changes in prevalence and failing to account for the model's specificity [54]. In contrast, the TSS considers both sensitivity and specificity and is less sensitive to changes in prevalence, although it is less frequently used in the literature and may be trickier to interpret [55].
Bio 1 (annual mean temperature) forms the most effective bioclimatic variable in the distribution of the two species, but each one has a completely different pattern toward this variable (Figure 2). S. dux has a wide range of temperatures that it could live in, while S. haemorrhoidalis has a very narrow range of temperatures that it could live in. This is very similar to several previously studied species, especially in a very similar family such as Calliphoridae. The species Old world screw worm Chrysomya bezziana has a very narrow temperature range, while the species of the same genus Chrysomya albiceps has a wide temperature range [29,34]. Precipitation-related variables such as bio 12 (annual precipitation) have a very low impact on the two species' distribution and have a small contribution to the generated model ( Figure 2). This could be due to the variety of habitats that the two species live in, from the heavy rain forests of the Indonesian tropics to the extremely dry deserts of Arabia.
The importance of temperature and humidity in the life cycles of these two species has been discussed previously in several works. S. haemorrhoidalis larvae require warm temperatures for accelerated growth and development, and temperature has a significant impact on how the species develops. The larvae of S. haemorrhoidalis developed most quickly in lab experiments at temperatures between 24 • C and 30 • C. It also experiences higher mortality due to high temperatures, with larvae and pupae being especially susceptible to heat stress [56]. In the same way, temperature has a significant impact on S. dux growth, just like it does on S. haemorrhoidalis. S. dux larvae developed most quickly in lab tests at temperatures between 25 • C and 30 • C. S. dux larvae were discovered to be relatively heattolerant, in contrast to S. haemorrhoidalis, with some studies finding effective development at temperatures as high as 38 • C, but this is not supported by the results generated here [57].
The future situation for the two species shows completely different results. S. haemorrhoidalis will not be in a significantly different position from its current situation, while the worst climate change scenario of 2070 shows a high degree of habitat loss, especially in tropical Africa (Figures 4e,f and 5e-h). S. dux, on the other hand, will increase and expand through different parts of its range, especially in Europe, in all studied scenarios. Such changes in the S. dux range will surely affect the surrounding ecosystem that it lives in (Figures 4a-d and 5a-d).
Climate change may cause changes in the interactions of Sarcophagid flies with other species [6]. Changes in temperature and precipitation patterns could impact the timing and success of other decomposer species, such as bacteria and fungus, which play essential roles in carrion decomposition [12]. Changes in precipitation patterns may alter the availability and decomposition of carrion by flesh flies, which are scavengers that feed on dead animal flesh [18]. For example, if precipitation becomes more intermittent or intense, it may change the rate at which carcasses decompose, thereby decreasing the availability of food for flesh flies. These changes may have an impact on the availability and quality of food for flesh flies, potentially leading to population decreases [58,59].

Conclusions
The present work provides valuable insights into the potential impacts of climate change on insect populations and highlights the need for continued research and monitoring in this area. By better understanding the effects of climate change on insect populations, we can develop more effective strategies for dealing with such changes and help mitigate the impacts of climate change on our planet's biodiversity. It is clear that climate change has a significant influence on the global potential distribution of Sarcophaga dux and Sarcophaga haemorrhoidalis. The maximum entropy model was used to predict the current and future distributions of these flies and illustrate that their potential habitat suitability is likely to shift in response to changing climatic conditions. These findings have important implications for wildlife management and conservation efforts, as well as for forensic and medical applications of these flies. Overall, this study highlights the need for reliable methods to evaluate the accuracy of species distribution models in the face of ongoing climate change. Furthermore, the study demonstrates the utility of maximum entropy modeling as a tool for predicting species distribution under changing climatic conditions. This approach can be used to identify areas of high conservation value as well as inform management decisions related to invasive species and habitat restoration. However, such models are subject to certain limitations and uncertainties, and further research is needed to improve their accuracy and reliability.
Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d15080903/s1, Figure S1: Niche overlapping between S. dux and S. haemorrhoidalis; Table S1: Comparison between current and future distribution area of S. dux and S. haemorrhoidalis based on natural Jenks map classification.