Abstract
Snakebite is one of the largest risks from wildlife, however little is known about venomous snake distribution, spatial variation in snakebite risk, potential changes in snakebite risk pattern due to climate change, and vulnerable human population. As a consequence, management and prevention of snakebite is hampered by this lack of information. Here we used habitat suitability modeling for 10 medically important venomous snakes to identify high snakebite risk area under climate change in Iran. We identified areas with high snakebite risk in Iran and showed that snakebite risk will increase in some parts of the country. Our results also revealed that mountainous areas (Zagros, Alborz, Kopet–Dagh mountains) will experience highest changes in species composition. We underline that in order to improve snakebite management, areas which were identified with high snakebite risk in Iran need to be prioritized for the distribution of antivenom medication and awareness rising programs among vulnerable human population.
Similar content being viewed by others
Introduction
Snakebite envenoming is a medical emergency and an important health problem worldwide1,2,3. Up to 5 million snakebites occur each year leading to about 100,000 deaths annualy 2,4,5,6. Despite recent attention to the challenge of snakebite 3,5,7,8,9,10 little is known about the distribution of venomous snakes, and therefore the spatial variation in snakebite risk and accessibility of vulnerable human population to healthcare system3. Further, the potential shift of high risk areas for snakebite due to climate change remains unknown. As a consequence, management of snakebite is hampered by this lack of information around the globe3,11. To ensure successful snakebite management it is necessary to identify high snakebite risk areas.
Identifying areas with high risk of snakebite can help us to locate target areas for awareness raising programs and to effectively provide antivenoms to the most vulnerable groups2,12,13. While our knowledge on current snakebite risk patterns remains limited2,14, venomous snakes are responding to climate change by shifting their habitats to cope with global warming2,15,16,17. Thus, the complexity of snakebite management further increases and therefore calls for more efforts to identify high snakebite risk areas under climate change2,10.
Habitat Suitability Models (HSMs) are practical tools to predict the impacts of climate and land use changes on species distribution18,19,20, identify their future distribution21, document changes in species composition22,23 by correlating species occurrence records to climatic variables18. Moreover, HSMs can be used to predict changes in distribution and interaction of species and their host or prey and predator in changing climate24,25,26. In recent years, HSMs are used in predicting snakebite risk areas under current and future climate change, identifying vulnerable human population2,13,27,28,29,30, and have even been shown to correlate with snakebite incidences28. In these studies, venomous snakes’ habitat suitability was considered as an indicator of snakebite risk2,13,27.
Iran is home to 15 terrestrial venomous snakes including one of the most famous snake in the world named Spider tailed viper, Pseudocerastes urarachnoides31. While all 15 species are medically important, four species (Echis carinatus Macrovipera lebetina, Naja oxiana, and P. persicus) are more widespread in Iran and are therefore responsible for the most snakebite incidents in the country32. Although previous studies investigated the challenge of snakebite in Iran by reporting snakebite incidents32,33 still little is known about medically important venomous snakes’ distribution and potential changes in their distribution under climate change17,34,35,36. Previous studies have shown that climate change will change distribution pattern of biodiversity including snakes in Iran37. While suitable habitats of some species will expand under climate change some others will lose considerable proportion of their suitable range. For instance, Yousefi et al.17 showed that mountains vipers of genus Montivipera will lose their suitable habitat and will shift to higher altitude to track their suitable climate. As model projections showed that species like E. carinatus and P. persicus will experience an expansion of suitable habitat area due to warming climate, this may lead to an increasing number of envenoming events38.
In order to mitigate the risk of snakebite and to identify vulnarable human populations, developing snakebite risk maps becomes crucial2. Thus, to effectively manage snakebite problem in Iran the aims of this study are (i) modeling the habitat suitability of medically important venomous snakes; (ii) determining hotspots of snakebite risk; (iii) quantifying changes in medically important venomous snakes’ composition due to climate change; (iv) identifying vulnerable populations to snakebite envenoming by considering exposure to snakebite and accessibility to health centers. The results of this study will guide responsible health organizations for the distribution of antivenom medication and awareness raising programs in the country.
Results
Modeling the impacts of climate change (years 2041–2070 SSP126 (Fig. S1), SSP585 (Fig. S2) and 2071–2100 SSP126 (Fig. S3), SSP585 (Fig. 1)) on habitat suitability of the 10 medically important venomous snakes showed that suitable habitats of five species (E. carinatus, Gloydius halys, Montivipera latifii, P. persicus and Walterinnesia aegyptia) will increase while five species will lose (M. lebetina, M. raddei, N. oxiana, P. urarachnoides and Vipera eriwanensis) some of their suitable habitat under climate (Fig. 1). Results of the evaluation metrics of the models were presented in Fig. 2. The evaluation metrics values were highest for M. latifii while the values were lowest for the E. carinatus.
Variable importance
Estimating variable importance for the 10 medically important venomous snakes showed that growing degree days heat sum above 10 °C followed by annual precipitation were the most important predictor of habitat suitability of the 10 snakes when averaged across all species (Fig. 3). Growing degree days heat sum above 10 °C was the most important determinants of suitable habitats of M. raddei, N. oxiana, P. persicus and V. eriwanensis, annual precipitation was identified the most influential predictor of habitat suitability for E. carinatus, M. lebetina and W. aegyptia and precipitation of driest quarter turn out to be the most important predictor of habitat suitability of G. halys, M. latifii and P. urarachnoides.
Hotspot of snakebite in Iran
By multiplying snakebite risk models of the 10 medically important venomous snakes (Fig. 4) we found that hotspots of snakebite in Iran are located in north west of Iran, Alborz mountains, and west of Zagros mountains. Zagros mountains have low richness of medically important venomous snakes under current climatic conditions but under climate change snakebite risk will increase in the mountains in particular under SSP585 scenario for years 2071–2100.
Species composition change
We compared similarity of medically important venomous snakes’ composition by comparing current and future distribution of the 10 medically important venomous snakes (Fig. 5). We found that under SSP585 for years 2041–2070 and 2071–2100 mountainous areas (Zagros, Alborz, Kopet–Dagh mountains) will experience highest changes in species composition.
Vulnerability to snakebite
Vulnerability to snakebite was modeled by considering exposure to snakebite plus accessibility to healthcare centers. Results showed that central parts of Iran and north east of the country have largest areas which are vulnerable to snakebite. Under climate change geographic pattern of snakebite risk will change especially in Zagros Mountains and areas vulnerable to snakebite will increase (Fig. 6).
Discussion
Snakebite is a neglected tropical disease and important health problem39, leading to thousands of fatal incidents and many more disabilities annually. Hereby, the rural population is at a higher risk of snakebite due to limited access to healthcare facilities, resulting in delayed administration of antivenom, which is a critical factor in the treatment of snakebite14,40. To ensure successful snakebite management and reduce snakebite incidents, identifying high snakebite risk areas under current and future climate is highly relevant. Here we mapped areas with high risk snakebite in Iran under current and future climate based on 10 medically important venomous snakes.
We found that the North west of Iran, Alborz mountains, and west of Zagros mountains have the highest richness of medically important venomous snakes and thus are regions with high snakebite risk. Our study confirms the results of previous studies on the importance of climate on distribution of venomous snakes of Iran13,15,34,38. Further, we showed that growing degree days heat sum above 10 °C followed by annual precipitation were the most important predictors of habitat suitability of the 10 snakes in Iran. Growing degree days heat sum above 10 °C is the most important variable because it influences the growth and development of plants and insects during the growing season which directly and indirectly influence prey availability for the snakes41.
Developing snakebite management strategies based solely on the current distribution of medically significant venomous snake species may be risky, as recent studies have demonstrated that climate change is capable of altering the potential distribution of these species10,29. Furthermore, model predictions suggest that climate change may lead to changes in the composition and species richness of medically significant venomous snake species, which could subsequently modify the pattern of snakebite risk10,29. Predicted changes in medically important venomous snakes’ hotspot and composition in mountains regions of Iran may result in previously less affected populations to venomous snakes becoming more exposed. The future snakebite risk models developed in this study can be useful to raise awareness among the people who might be at risk of envenoming under climate change and by considering potential changes in snakebite risk pattern for antivenom distribution in the future. Together, this may support reducing the numbers of deaths and cases of disability due to snakebite envenoming2,10,42.
Access to antivenom can help to reduce mortality and morbidity of snakebite envenoming42. Mapping medically important venomous snakes’ current and future distribution is critical for knowing where antivenin for each species is needed2. Beside determining areas with less accessibility to medical facilities is necessary to identify the most vulnerable populations2,42. Our models, showing the most vulnerable areas to snakebite in Iran can guide antivenom distribution by the Ministry of Health and Medical Education across the country.
WHO introduced community empowerment and engagement as an important strategy to prevent and control snakebite envenoming39. Areas identified at risk of snakebite have high priority to guide snakebite mitigation and training measures for health teams and local people in Iran2. Farmers, shepherd and nomads who encounter snakes in their everyday life while working on farms or grazing their livestock are the most vulnerable groups. Thus, awareness raising programs should focus on these population groups. In most snakebite accidents the snakes were not identified and therefore no specific treatment can be applied. Educating local communities in areas with high snakebite risk to be able to identify venomous snakes should therefore be prioritized40. Moreover, despite the importance of the snakebite in Iran still health workers are not familiar with all medically important venomous snakes of Iran. Thus, beside local people, it is necessary to regularly educate health workers in the country. Some occupations are associated with higher risk of encountering venomous snakes such as protected areas rangers, foresters and tour guides. Thus, organizations responsible for these occupations like Department of Environment and Department of Natural Resources and Watershed Management can run awareness raising and training workshops in high snakebite risk areas to educate them how to avoid snakebite and to seek health care instead seeking care from traditional healers39.
Venomous snakes have secretive behavior and life history thus studying snakebite risk is a challenging task especially in developing countries like Iran43,44. Using citizen science is a promising avenue to overcome lack of knowledge on distribution of venomous snake43. We encourage using citizen science in Iran as increasing number of young people are getting interested in herping in the country. Their observations on venomous snakes’ distribution in areas with limited data can be used to develop high resolution distribution maps and more accurate models. To be able to use herpers observations in snakes and snakebite studies their observations should be documented with a photograph and deposited in relevant databases like iNaturalist or HerpMapper43.
Conclusions
Identifying high snakebite risk areas under climate change, distribution of effective antivenoms, and education are four key factors in snakebite management. In this study, we predicted the impacts of climate change on 10 medically important venomous snakes, their richness, community composition and identified vulnerable snakebite area. This research study has important implications for snakebite management in Iran. As shown in other studies13,27,28,29,30,45,46 and confirmed in this research HSMs developed based on distribution of medically important venomous snakes can guide public policies for the distribution of antivenom medication and identification of regions with higher snakebite risk to guide mitigation and training measures for vulnerable populations as well as health teams. It should be noted that population density, educational level, income, etc. could are also important variables that affect people’s living standards and their exposure to snake habitat proximity. But since these predictors are not available at high spatial resolution, we were unable to consider them. Apart from Iran, snakebite is an important challenge in other Middle Eastern countries32,47. Our approach in modeling snakebite risk and mapping vulnerability to snakebite under current and future climate can be applied in other countries to reduce snakebite incidents in the region.
Methods
Species data
In this study, distribution records of medically important venomous snakes of Iran31 gathered from different sources as follows: fieldworks, online databases (GBIF, VertNet), published books and papers ranging from 2000 to 2022. After gathering distribution records from different sources, we removed duplicates and thinned distribution records to match with climatic layers’ resolution (1 km2). Considering some species like M. kuhrangica, Bungarus sindnus, Cerastes gasperettii, and Eristicophis macmahoni have restricted range31 and limited distribution records it was not possible to develop robust habitat suitability models for them thus they were removed from further analysis31. We considered M. lebetina and M. razii together as one taxonomic unit31. We found enough distribution records to build habitat suitability model for 10 medically important venomous snake species (E. carinatus (148 points), G. halys (34 points), M. lebetina (150 points), M. latifii (20 points), M. raddei (23 points), N. oxiana (51 points), P. persicus (151 points), P. urarachnoides (22 points), V. eriwanensis (54 points), W. aegyptia (32 points)) in Iran31 (Figs. 7 and 8).
Predictor variables
For climate data, we used CHELSA high resolution climatologies version 2.148.This platform provides downscaled climate data for current and future climate conditions. We downloaded the bioclimatic variables at a 30 arcsec (~ 1 km) resolution for current climate and future climate (years 2041–2070 and 2071–2100) scenarios from five different CMIP6 Global Circulation Models (GCMs) (GFDL-ESM4, IPSL-CM6A-LR, MPI-ESM1-2-HR, MR-ESM2-0, UKESM1-0-LL). CMIP6 GCMs are state-of-the-art and using several GCMs is reducing the uncertainty coming from the single GCMs. The selected scenarios are corresponding to the available data. Further, we used two (SSP126, SSP585) different shared socioeconomic pathways (SSPs) for each GCM to represent the uncertainty of future CO2 emissions and therefore the expected change in climate. As predictor variables for habitat suitability modelling, we selected four bioclimatic variables including growing degree days heat sum above 10 °C (GDD), temperature seasonality (Bio4), annual precipitation (Bio12) and precipitation of the driest quarter (Bio17). Bioclimatic variables are gridded layers containing information about the climate in a particular region. They are derived variables from the monthly min, max, mean temperature, and mean precipitation values and were developed for species distribution modeling and related ecological applications. Those variables have shown to be reliable predictors for SDMs in a multitude of studies. Growing degree days are the heat sum of all days above the 10 °C temperature accumulated over 1 year, temperature seasonality is defined as standard deviation of the monthly mean temperatures, annual precipitation is the accumulated precipitation amount over 1 year and precipitation of the driest quarter is defined as the accumulated precipitation within the driest quarter. The selection was a combination of choosing species specific biologically relevant predictors and removing autocorrelated variables13,34,38. Additionally, we added slope as a fifth predictor variable. Slope was calculated from the Shuttle Radar Topography Mission (SRTM) digital elevation model49 at a 30 arcsec (~ 1 km) resolution with the terrain function of the raster package (version 3.4–1350). We included slope because it is an important predictor for habitat suitability of snakes. While climatic variables change over time, slope remains constant. However, future habitat suitability will still depend highly on slope. For instance, species occurring in flat areas will not adapt to terrain with steep slopes, even though the climate might become more suitable in the steep slopes. Including this variable thus is important to not overestimate the magnitude of change. For the SDM predictor variables, we chose biologically relevant variables and removed some correlated variables during this process as described in the methods section. We started with all bioclimatic variables and reduced the number of predictors in an empirical process of adding and removing variables to the SDMs in order to improve their performance. As all bioclimatic variables are derived from temperature and precipitation, some of those variables are highly correlated (e.g. maximum temperature and temperature of the warmest month). Excluding auto-correlated variables can help to avoid violating the assumption of independence of data points. Therefore, it is important to reduce the number of variables and select those which are less correlated and therefore improve the accuracy and reliability of the SDM.
Modeling snakebite risk under current and future climate
For modelling the suitable habitat of the snake species under current and future climate, we used an equally weighted ensemble approach of five different habitat suitability modeling algorithms (HSM algorithms). We used GLMs (base R-package; R environment51), GAMs (gam R-package version 1.20.152), GBMs (gbm R-packge version 2.1.853), RandomForests (randomForest R-package version 2.1.854) and Maxent (dismo R-package version 1.3-555). For each species, we generated 5000 pseudo-absences by randomly sampling coordinates from the ecoregions where the species occurred and down- weighted the absences in the model algorithms to balance the presence-absence prevalence to 0.556. To assess model performance, we used a split-sample approach (70% training data and 30% evaluation data) with 20 repetitions. The total number of points was 685 in which 480 points (70%) were used for as training data. Performance was measured using the area under the receiver operating characteristic curve (AUC)57,58 and True Skill Statistics (TSS)59. TSS values range from − 1 to + 1, where + 1 indicates perfect performance and value of zero meaning random predictions. AUC values range from 0 to 1, a value of 0.5 indicates that the performance of the model is not better than random, while values closer to 1.0 indicate better model performance57,58,59. We additionally checked the projections of the models visually (expert-based evaluation). Binary classifications were done using the sensitivity–specificity sum maximization approach (R package presenceAbsence 1.1.960). Model variable importance was assessed using the variables_importance function of the biomod2 package (version 3.5.161) for GLMs, GAMs and GBMs, the importance function of the randomForest package (version 2.1.854) for RandomForest models and the ecospat.maxentvarimport function of the ecospat package (version 3.262) for Maxent models. We evaluated the importance of each variable across the different algorithms to show the variation in variable importance.
Mapping snakebite hotspot, community composition changes and vulnerability to snakebite
To map the snakebite hotspot, we calculated the species richness of the 10 medically important venomous snakes by stacking binary habitat suitability maps of all species and calculating the sum for each grid cell in R51. For the community composition analysis, species richness maps were converted to matrices and then compared with the ecospat.CommunityEval function of the ecospat package (version 3.262). Finally, to map vulnerability to snakebite across the country we considered two factors a) distance to cities which can provide primary health care services (accessibility to health care) and b) exposure to medically important venomous snakes (binary habitat suitability maps). We calculated a distance-to-city layer as the distance of each grid cell to the nearest city using the distance function of the raster package (version 3.4-1350). For each species we masked the binary habitat suitability map with the distance-to-city layer to get a vulnerability to snakebite map for each medically important venomous snake. For the final risk maps, we multiplied the species richness maps with the distance-to-city layer. From this we extracted the values from the vulnerability to snakebite maps for each cell to get the vulnerability index.
Data availability
The datasets generated and analysed during the current study are available from the sources described in the manuscript and also the corresponding author on reasonable request.
References
Kasturiratne, A. et al. The global burden of snakebite: A literature analysis and modeling based on regional estimates of envenoming and deaths. PLoS Med. 5, 1591–1604 (2008).
Gutiérrez, J. M. et al. Snakebite envenoming. Nature 3, 17063 (2017).
Pintor, A. F. et al. Addressing the global snakebite crisis with geo-spatial analyses—Recent advances and future direction. Toxicon: X 11, 100076 (2021).
Chippaux, J. P. Snake-bites: Appraisal of the global situation. Bull. World Health Organ. 76, 515 (1998).
Kasturiratne, A., Lalloo, D. G. & de Silva, H. J. Chronic health effects and cost of snakebite. Toxicon: X 9–10, 100074 (2021).
Williams, D. J. et al. Strategy for a globally coordinated response to a priority neglected tropical disease: Snakebite envenoming. PLoS Negl. Trop. Dis. 13, e0007059 (2019).
Barnes, K. et al. Delays, fears and training needs: Perspectives of health workers on clinical management of snakebite revealed by a qualitative study in Kitui County, Kenya. Toxicon: X 11, 100078 (2021).
van Oirschot, J., Ooms, G. I., Waldmann, B. & Kadam, P. Snakebite incidents, prevention and care during COVID-19: Global key-informant experiences. Toxicon: X 9–10, 100075 (2021).
Duda, R., Monteiro, W. M. & Giles-Vernick, T. Integrating lay knowledge and practice into snakebite prevention and care in central Africa, a hotspot for envenomation. Toxicon: X 11, 100077 (2021).
Martín, G. et al. Implications of global environmental change for the burden of snakebite. Toxicon: X 9–10, 100069 (2021).
Malhotra, A. et al. Promoting co-existence between humans and venomous snakes through increasing the herpetological knowledge base. Toxicon: X 12, 100081 (2021).
Longbottom, J. et al. Vulnerability to snakebite envenoming: A global mapping of hotspots. The Lancet 392, 673–684 (2018).
Yousefi, M., Kafash, A., Khani, A. & Nabati, N. Applying species distribution models in public health research by predicting snakebite risk using venomous snakes’ habitat suitability as an indicating factor. Sci. Rep. 10, 1–11 (2020).
de Castañeda, R. R., Bolon, I. & Gutiérrez, J. M. A transdisciplinary approach to snakebite envenoming. Toxicon: X 13, 100088 (2022).
Penman, T. D., Pike, D. A., Webb, J. K. & Shine, R. Predicting the impact of climate change on Australia’s most endangered snake, Hoplocephalus bungaroides. Divers. Distrib. 16, 109–118 (2010).
Terribile, L. C. et al. Reducing Wallacean shortfalls for the coralsnakes of the Micrurus lemniscatus species complex: Present and future distributions under a changing climate. PLoS ONE 13, e0205164 (2018).
Yousefi, M. et al. Upward altitudinal shifts in habitat suitability of mountain vipers since the last glacial maximum. PLoS ONE 10, e0138087 (2015).
Guisan, A., Thuiller, W. & Zimmermann, N. E. Habitat Suitability and Distribution Models: With Applications in R (Cambridge University Press, 2017).
Vargas-Jaimes, J. et al. Impact of climate and land cover changes on the potential distribution of four endemic salamanders in Mexico. J. Nat. Conserv. 64, 126066 (2021).
Gonçalves, F., Sales, L. P., Galetti, M. & Pires, M. M. Combined impacts of climate and land use change and the future restructuring of Neotropical bat biodiversity. Perspect. Ecol. Conserv. 19, 454–463 (2021).
González-Fernández, A., Manjarrez, J., García-Vázquez, U., D’Addario, M. & Sunny, A. Present and future ecological niche modeling of garter snake species from the Trans-Mexican Volcanic Belt. PeerJ 6, e4618 (2018).
Maiorano, L. et al. The future of terrestrial mammals in the Mediterranean basin under climate change. Philos. Trans. R. Soc. Lond. B Biol. Sci. 366, 2681–2692 (2011).
Buras, A. & Menzel, A. Projecting tree species composition changes of European forests for 2061–2090 under RCP 4.5 and RCP 8.5 scenarios. Front. Plant Sci. 9, 1986 (2019).
Gómez-Ruiz, E. P. & Lacher, T. E. Jr. Climate change, range shifts, and the disruption of a pollinator-plant complex. Sci. Rep. 9, 14048 (2019).
Grünig, M., Mazzi, D., Calanca, P., Karger, D. N. & Pellissier, L. Crop and forest pest metawebs shift towards increased linkage and suitability overlap under climate change. Commun. Biol. 3, 233 (2020).
Brambilla, M. et al. Species interactions and climate change: How the disruption of species co-occurrence will impact on an avian forest guild. Glob. Change Biol. 26, 1212–1224 (2020).
Yañez-Arenas, C., Peterson, A. T., Mokondoko, P., Rojas-Soto, O. & Martínez-Meyer, E. The use of ecological niche modeling to infer potential risk areas of snakebite in the Mexican state of Veracruz. PLoS ONE 9, e100957 (2014).
Yañez-Arenas, C., Peterson, A. T., Rodríguez-Medina, K. & Barve, N. Mapping current and future potential snakebite risk in the new world. Clim. Change 134, 697–711 (2016).
Nori, J., Carrasco, P. A. & Leynaud, G. C. Venomous snakes and climate change: Ophidism as a dynamic problem. Clim. Change 122, 67–80 (2014).
Zacarias, D. & Loyola, R. Climate change impacts on the distribution of venomous snakes and snakebite risk in Mozambique. Clim. Change 152, 195–207 (2019).
Rajabizadeh, M. Snakes of Iran (IranShenasi Publishing, 2018).
Dehghani, R., Fathi, B., Panjeh Shahi, M. & Jazayeri, M. Ten years of snakebites in Iran. Toxicon 90, 291–298 (2014).
Monzavi, S. M., Dadpour, B. & Afshari, R. Snakebite management in Iran: Devising a protocol. J. Res. Med. Sci. 19, 153–163 (2014).
Kazemi, E., Kaboli, M., Khosravi, R. & Khorasani, N. Evaluating the importance of environmental variables on spatial distribution of Caspian cobra Naja oxiana (Eichwald, 1831) in Iran. AHR 10, 129–138 (2019).
Rastegar-Pouyani, E. et al. A re-evaluation of taxonomic status of Montivipera (Squamata: Viperidae) from Iran using a DNA barcoding approach. Biochem. Syst. Ecol. 57, 350–356 (2014).
Behrooz, R. et al. Habitat modeling and conservation of the endemic Latifi’s viper (Montivipera latifii) in Lar National Park Northern Iran. Herpetol. Conserv. Biol. 10, 572–582 (2015).
Yousefi, M. et al. Climate change is a major problem for biodiversity conservation: A systematic review of recent studies in Iran. Contemp. Probl. Ecol. 12, 394–403 (2019).
Yousefi, M. & Kafash, A. Venomous snakes of Iran under climate change. In SCCS Europe-4th Hungarian Student Conference on Conservation Science, 4–8 September (2018).
WHO. Snakebite Envenoming: A Strategy for Prevention and Control (World Health Organization, 2016).
Moos, B. et al. A scoping review of current practices on community engagement in rural East Africa: Recommendations for snakebite envenoming. Toxicon: X 11, 100073 (2021).
Pettorelli, N. The Normalized Differential Vegetation Index (Oxford University Press, 2013).
Potet, J. et al. Access to antivenoms in the developing world: A multidisciplinary analysis. Toxicon X 12, 100086 (2021).
Durso, A. M. et al. Citizen science and online data: Opportunities and challenges for snake ecology and action against snakebite. Toxicon: X 9–10, 100071 (2021).
Steen, D. A. Snakes in the grass: Secretive natural histories defy both conventional and progressive statistics. Herpetol. Conserv. Biol. 5, 183–188 (2010).
Melo Araújo, S. C., Ceron, K. & Guedes, T. B. Use of geospatial analyses to address snakebite hotspots in mid-northern Brazil: A direction to health planning in shortfall biodiversity knowledge areas. Toxicon 213, 43–51 (2022).
Ochoa, C. et al. Estimating and predicting snakebite risk in the Terai region of Nepal through a high-resolution geospatial and One Health approach. Sci. Rep. 11, 23868 (2021).
Amr, Z. S., Abu Baker, M. A. & Warrell, D. A. Terrestrial venomous snakes and snakebites in the Arab countries of the Middle East. Toxicon 177, 1–15 (2020).
Karger, D. N. et al. Climatologies at high resolution for the Earth land surface areas. Sci. Data. 4, 170122 (2017).
Jarvis, A., Reuter, H. I., Nelson, A. & Guevara, E. Hole-filled SRTM for the globe Version 4. Available from the CGIAR-CSI SRTM 90m Database. https://srtm.csi.cgiar.org. Accessed on 15 Apr 2015 (2008).
Hijmans, R. J. raster: Geographic Data Analysis and Modeling. R package version 3.4-13 (2021).
R Core Team. R: A Language and Environment for Statistical Computing (R Core Team, 2020).
Hastie, T. gam: Generalized Additive Models. R package version 1.20.1 (2022).
Greenwell, B., Boehmke, B., Cunningham, J. & Developers, GBM. gbm: Generalized Boosted Regression Models. R package version 2.1.8 (2020).
Liaw, A. & Wiener, M. Classification and Regression by randomForest. R News 2, 18–22 (2002).
Hijmans, R. J., Phillips, S., Leathwick, J. & Elith, J. dismo: Species Distribution Modeling. R package version 1.3-5 (2021).
Stockwell, D. The GARP modelling system: Problems and solutions to automated spatial prediction. Int. J. Geogr. Inf. Sci. 13, 143–158 (1999).
Hanley, J. A. & McNeil, B. J. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143, 29–36 (1982).
Fielding, A. H. & Bell, J. F. A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ. Conserv. 24, 38–49 (1997).
Allouche, O., Tsoar, A. & Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 43, 1223–1232 (2006).
Freeman, E. A. & Moisen, G. PresenceAbsence: An r package for presence absence model analysis. J. Stat. Softw. 23, 1–31 (2008).
Thuiller, W., Damien, G., Maya, G., Robin, E. & Frank, B. biomod2: Ensemble Platform for Species Distribution Modeling. R package version 3.5.1 (2021).
Broennimann, O., Di Cola, V. & Guisan, A. ecospat: Spatial Ecology Miscellaneous Methods. R package version 3.2 (2021).
Acknowledgements
We would like to thank the Iranian National Science Foundation (Project Number: 99021952) for supporting this research.
Author information
Authors and Affiliations
Contributions
M.Y. conceived and designed this study. M.Y., M.R., A.K., S.H.Y. and E.R.P. collected distribution data, M.G. performed statistical analyses. M.Y. and A.K prepared the maps. M.Y. wrote the manuscript with inputs from M.G. (method section). All authors read, revised, and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Yousefi, M., Yousefkhani, S.H., Grünig, M. et al. Identifying high snakebite risk area under climate change for community education and antivenom distribution. Sci Rep 13, 8191 (2023). https://doi.org/10.1038/s41598-023-35314-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-35314-1
This article is cited by
-
Projecting the impact of climate change on honey bee plant habitat distribution in Northern Ethiopia
Scientific Reports (2024)
-
Predictive spatial correlation analysis of snakebites of Krishna District, India
Microsystem Technologies (2024)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.