Statistical analyses
We analysed whether and how vegetation structure and other characteristics of grassland habitats affect the fine-scale occurrence of snakes by building Generalized Linear Mixed Models (GLMM) separately for each study species. In GLMMs, presence/absence of vipers was incorporated as a binary dependent variable, while the four variables characterising vegetation structure were applied as fixed explanatory variables, with grass/rock/shrub surface cover in V. graeca GLMMs and number of burrows in V. renardi GLMMs as additional fixed variables. The sampling site was incorporated in the GLMMs as a random factor to control for the spatial non-independence of the observations. We fitted the GLMMs specifying binomial error distribution using the ‘lme4’ package (Bates et al., 2014). We then used an information-theoretic framework and a model selection approach (Burnham & Anderson, 2002) to run all possible combinations of fixed effects to identify models with substantial empirical support based on Akaike differences (Δi = AICi- AICmin < 2.0) and to perform model averaging based on the relative importance of explanatory variables using the ‘MuMIn’ package in R (Bartoń, 2018).
RESULTS
The number of snakes found (presence locations) varied from 32 to 73 across the three species. V. graeca and V. ursinii were much rarer locally than V. renardi , and the search effort-corrected density was an order of magnitude higher in V. renardi than in the other two species (Table 1). Vegetation structure was recorded by white-board photography in a total of 141 presence locations and 726 random locations. Almost half of the pre-randomised locations in the alpine habitats of V. graeca were in inaccessible cliffs and 17% of pre-randomised locations fell on roads or water bodies in V. renardi habitats (Table 1).
Most of the vegetation structure variables followed a normal distribution, except for MHV in V. ursinii habitats, where vegetation was at some sampling points taller than 1 m, i.e., the height of the whiteboard (Fig. 1, 3). The correlations between vegetation structure variables were usually not significant, except between LA and HCV in V. graeca and in V. ursinii habitats and when data were pooled across species, and also between LA and FHD in V. graeca habitats (Fig. 3).
In V. graeca , the full GLMM returned no significant main effect, whereas HCV was included in all and shrub cover was included in five of the six best models (ΔAICc < 2). The averaged parameter estimate was significant and positive only for HCV (Table 2), indicating a higher chance of occurrence of V. graeca in taller and closed vegetation.
In V. renardi , LA and the number of burrows had significant explanatory power in the full model and LA, FHD and number of burrows were included in both best models (ΔAICc < 2). The effects of LA and number of burrows were positive, whereas that of FHD was negative (Table 2), indicating higher chances of V. renardi occurrence in microhabitats with higher, more homogeneous cover and more burrows.
In V. ursinii , the full model had the lowest AICc value, and in the two best models (ΔAICc < 2), LA and MHV had significant positive effects, whereas HCV had a significant negative effect (Table 2), indicating higher chances of V. ursinii occurrence in tall, high-cover but more open vegetation.
DISCUSSION
Our study provided key results in the development of field data collection and data processing methodology for quantifying the role of vegetation structure in studies of animal microhabitat selection and in understanding how vegetation affects the occurrence of snakes in grasslands. Our results demonstrate that variables relevant in describing vegetation structure can be derived from the automated processing of images taken by standardised whiteboard photographs. In addition, at least one of the variables so derived influenced the occurrence of snakes in three species in three widely differing grassland ecosystems.
Our method decreases subjectivity in quantifying vegetation structure as it returns exact cover values along with the vertical range in pixel rows and does not rely on estimates by eye, and minimises observer bias and measurement error. Moreover, it does not require arbitrarily delimited measurement classes to characterise vertical variation in structure. Our method considers image pixels as the unit of analysis, however, specification of larger units, e.g. 4, 9 or 16 image pixels combined is also possible, which allows decreasing the resolution (upscaling) and computing time. Repeating the analyses at different unit sizes can provide further insight as it offers the possibility of studying scale-dependence in habitat selection, i.e., the identification of the environmental grain size at which animal-vegetation relationships are the strongest (Gunton et al., 2014; Lengyel et al., 2014). The objectivity of the method also allows comparisons made in two or more species or ecosystems and also allows local measurements to be extrapolated to larger areas, habitat types or ecosystems.
Our results support the role of vegetation structure in microhabitat selection of snakes. While compositional habitat diversity (plant species composition) and estimated cover of vegetation have been reported to influence the occurrence of reptiles (Nemes et al., 2006; Stumpel & van der Werf, 2012), our study confirmed that vertical aspects of vegetation structure can also be important in the habitat selection of reptiles (Mizsei et al., 2020b). For all three viper species studied, HCV and LA were the most important variables, indicating that vipers chose microhabitats where the vertical cover of the grass was higher than average, as measured in random locations. In the case of V. ursinii , a previous study (Máté & Vidéki, 2007) did not find a relationship between snake occurrence and plant species composition of the same study grassland. Our study thus also exemplifies that considering the structural aspects of vegetation can provide additional explanatory power in predicting the occurrence of snakes in microhabitats.
The role of vegetation structure in the microhabitat selection of snakes is probably determined by a trade-off between the need to hide from predators, for which the chances are better in higher or more dense vegetation (Wilgers & Home, 2007; Hansen et al., 2018), and the need to thermoregulate, for which the chances are better in lower or sparser vegetation (Muri et al., 2015). This trade-off is probably the reason why HCV had opposite effects on the occurrence of V. graeca(positive) and V. ursinii (negative) because different species may find different optimum values along the continuum of vegetation height. Considering the latter negative relationship, it has to be noted that tall wetland plants were common in V. ursinii habitats, representing more dense cover and shading, which probably reduced the possibility of sunbathing for the vipers (Muri et al., 2015), which can explain why vipers appeared to avoid microhabitats with tall and dense vegetation. Further studies of other species with different needs for hiding vs. thermoregulation will certainly shed more light on this trade-off in habitat selection in snakes. Such knowledge will be fundamental for habitat restoration and conservation management actions for snakes.
Two limitations of this study need to be mentioned for the correct interpretation of our results. First, our study was limited by the small number of presence locations for V. ursinii and the high spatial scatter of V. graeca presence locations in the study area. Unfortunately, these rare endangered species have low detectability due to their low abundance, hidden lifestyle and camouflaged body pattern. Data collection requires huge sampling effort that represents significant challenges in logistics and person-power. A potential source of error common in such studies is the assumption of absence in random locations where the species is not found because absence cannot be deduced without uncertainty as the individuals of the study species may actually live at particular random locations (Olivier & Wotherspoon, 2006; Phillips et al., 2009). In our study, the possibility of this error was high in V. renardi , which species showed extreme abundance in the study habitat and several new presence locations were found during sampling “random” locations. Another practical limitation of our method is that placing the whiteboard on uneven ground or very dense vegetation may result in changes in the vegetation next to the whiteboard, which may distort the value of vegetation structure variables. To avoid this problem, we recommend a careful trimming of the vegetation in the plane and on the backside of the board to make sure that the board is standing firmly on the ground.
In conclusion, the supplementation of standardised whiteboard photography with automated image processing allows the calculation of simple measures of vegetation structure that can provide additional insight into animal-vegetation relationships beyond the role of plant species composition. The combined use of field recording and image processing and analysis offers several options that broaden the range of cost-effective ecological survey methods and can make a substantial contribution to the design and implementation of evidence-based conservation, including the conservation of endangered, grassland specialist vipers.
REFERENCES
Barthelme, R. (2019). imager: Image Processing Library Based on ’Cimg’. R package version 0.41.2. https://CRAN.R-project.org/package=imager
Bartoń, K. (2018). MuMIn: Multi-Model Inference. R package version 1.42.1. https://CRAN.R-project.org/package=MuMIn
Bates, D., Mächler, M., Bolker, B. M., & Walker, S. (2014). lme4: Linear mixed-effects models using Eigen and S4. R package version 1.0-5. http://CRAN.R-project.org/package=lme4
Benkobi, R., Uresk, D. W., Schenbeck, G., & King, R. M. (2000). Protocol for monitoring standing crop in grasslands using visual obstruction. Journal of Range Management, 53, 627–633. https://doi.org/10.2307/4003158
Bergstedt, J., Westerberg, L., & Milberg, P. (2009). In the eye of the beholder: bias and stochastic variation in cover estimates. Plant Ecology, 204(2), 271–283. https://doi.org/10.1007/s11258-009-9590-7
Block, W. M., Franklin, A. B., Ward, J. P., Ganey, J. L., & White, G.C. (2001). Design and implementation of monitoring studies to evaluate the success of ecological restoration on wildlife. Restoration Ecology, 9, 293–303. https://doi.org/10.1046/j.1526-100x.2001.009003293.x
Böhm, M., Collen, B., Baillie, J. E. M., Bowles, P., Chanson, J., Cox, N., …, & Zug, G. (2013). The conservation status of the world’s reptiles. Biological Conservation, 157, 372–385. https://doi.org/10.1016/j.biocon.2012.07.015
Burnham, K. P., & Anderson, D. R. (2002). Model Selection and Multimodel Inference, 2nd edn. Springer-Verlag, New York, NY.
Faria, N., & Silva, J. P. (2010). Habitat selection of the little bustard during the beginning of an agricultural year. Ardeola, 57, 363–373.
Ferchaud, A.-L., Ursenbacher, S., Cheylan, M., Luiselli, L., Jelić, D., Halpern, B., Major, Á., Kotenko, T., Keyan, N., Crnobrnja-Isailović, J., Tomović, L., Ghira, I., Ioannidis, Y., Arnal, V., & Montgerald, C. (2012). Phylogeography of the Vipera ursinii complex (Viperidae): mitochondrial markers reveal an east–west disjunction in the Palaearctic region. Journal of Biogeography, 39, 1836–1847. https://doi.org/10.1111/j.1365-2699.2012.02753.x
Freitas, I., Ursenbacher, S., Mebert, K., Zinenko, O., Schweiger, S., Wüster, W., Brito, J.C., Crnobrnja-Isailovic, J., Halpern, B., Fahd, S., Santos, X., Plegeuzuelos, J.M., Joger, U., Orlov, N., Mizsei, E., Lourdais, O., Zuffi, M., Strugariu, A., Zamfirescu, S., Martínez-Solano, I., Velo-Antón, G., Kaliontzopoulou, A., & Martínez-Freifría, F. (2020). Evaluating taxonomic inflation: towards evidence-based species delimitation in Eurasian vipers (Serpentes: Viperidae).Amphibia-Reptilia, 41, 285–311. https://doi.org/10.1163/15685381-bja10007
Fretwell, S. D., & Lucas, H. J. (1970). On territorial behaviour and other factors influencing habitat distribution in birds. Acta Biotheoretica, 19, 16–36. https://doi.org/10.1007/BF01601955
Garden, J. G., McAlpine, C. A., Possingham, H. P., & Jones, D. N. (2007), Habitat structure is more important than vegetation composition for local-level management of native terrestrial reptile and small mammal species living in urban remnants: A case study from Brisbane, Australia. Austral Ecology, 32, 669–685. https://doi.org/10.1111/j.1442-9993.2007.01750.x
Gibbons, J.W., Scott, D. E., Ryan, T. J., Buhlmann, K. A., Tuberville, T. D., Metts, B. S., Greene, J. L., Mills T., Leiden, Y., Poppy, S., & Winne, C. T. (2000). The Global Decline of Reptiles, Déjà Vu Amphibians: Reptile species are declining on a global scale. Six significant threats to reptile populations are habitat loss and degradation, introduced invasive species, environmental pollution, disease, unsustainable use, and global climate change. BioScience, 50, 653–666. https://doi.org/10.1641/0006-3568(2000)050[0653:TGDORD]2.0.CO;2
Gunton, R. M., Klenke, R. A., Paloniemi, R., Gavish, Y., Marsh, C. J., Kunin, W. E., & Henle, K. (2014). The meaning of “scale”. In Pp. 19–21 in Scaling in Ecology and Biodiversity Conservation (pp. 19–22). Pensoft Publishers, Sofia. ISBN 978-954-642-739-7
Hansen, N., Sato, C. F., Michael, D. R., Lindenmayer, D. B., & Driscoll, D. B. (2018). Predation risk for reptiles is highest at remnant edges in agricultural landscapes. Journal of Applied Ecology, 56, 31–43. https://doi.org/ 10.1111/1365-2664.13269
Jellinek, S., Parris, K. M., McCarthy, M. A., Wintle, B. A., & Driscoll, D. A. (2014). Reptiles in restored agricultural landscapes: the value of linear strips, patches and habitat condition. Animal Conservation, 17, 544–554. https://doi.org/ 10.1111/acv.12121
Johnson, C. J., Seip, D. R., & Boyce, M. S. (2004). A quantitative approach to conservation planning: using resource selection functions to map the distribution of mountain caribou at multiple spatial scales.Journal of Applied Ecology, 41, 238–251. https://doi.org/10.1111/j.0021-8901.2004.00899.x
Karr, J. R., & Roth, R. R. (1971). Vegetation structure and avian diversity in several New World areas. American Naturalist, 105,423–435. https://doi.org/ 10.1086/282735
Krecsák, L., Zamfirescu, S., & Korsós, Z. (2003). An updated overview of the distribution of the Moldavian steppe viper (Vipera ursinii moldavica Nilson, Andrén et Joger, 1993). Russian Journal of Herpetology, 10, 199–206.
Lawlor, L. R., & Smith, J. M. (1976). The coevolution and stability of competing species. American Naturalist, 110, 79–99. https://doi.org/10.1086/283049
Lengyel, S., Déri, E., & Magura, T. (2016). Species richness responses to structural or compositional habitat diversity between and within grassland patches: A multi-taxon approach. PLOS ONE, 11,e0149662. https://doi.org/10.1371/journal.pone.0149662
Lengyel, S., Kosztyi, B., Ölvedi, T. B., Gunton, R. M., Kunin, W. E., Schmeller, D. S., & Henle, K. (2014). Conservation strategies across spatial scales. In Scaling in Ecology and Biodiversity Conservation (pp. 133–136). Pensoft Publishers, Sofia. ISBN 978-954-642-739-7
Loke, L. H. L., Ladle, R. J., Bouma, T. J., & Todd, P. A. (2015). Creating complex habitats for restoration and reconciliation.Ecological Engineering, 77, 307–313. https://doi.org/10.1016/j.ecoleng.2015.01.037
MacArthur, R. H., & MacArthur, J. W. (1961). On bird species diversity.Ecology, 42, 594–598. https://doi.org/10.2307/1932254
MacArthur, R. H., MacArthur, J. W., & Preer, J. (1962). On bird species diversity II. Prediction of bird census from habitat measurements.American Naturalist, 96, 167–174. https://doi.org/10.1086/282219
Máté, A., & Vidéki, R. (2007). A rákosi vipera által preferált gyepek szerkezetének monitorozó jellegű vizsgálata. Rosalia, 3,125–141.
Michael, D. R., Crane, M., Florance, D., & Lindenmayer, D. B. (2018). Revegetation, restoration and reptiles in rural landscapes: Insights from long-term monitoring programmes in the temperate eucalypt woodlands of south-eastern Australia. Ecological Management and Restoration, 19, 32–38. https://doi.org/10.1111/emr.12294
Milberg, P., Bergstedt, J., Fridman, J., Odell, G., & Westerberg, L. (2008). Observer bias and random variation in vegetation monitoring data. Journal of Vegetation Science, 19(5), 633–644. https://doi.org/10.3170/2008-8-18423
Mizsei, E., Fejes, Z., Malatinszky, Á., Lengyel, S., & Vadász, C. (2020a). Reptile responses to vegetation structure in a grassland restored for an endangered snake. Community Ecology 21, 203–212. https://doi.org/10.1007/s42974-020-00019-2
Mizsei, E., Jablonski, D., Roussos, S. A., Dimaki, M., Ioannidis, Y., Nilson, G., & Nagy, Z. T. (2017). Nuclear markers support the mitochondrial phylogeny of Vipera ursinii–renardi complex (Squamata: Viperidae) and species status for the Greek meadow viper.Zootaxa, 4227, 75–88. https://doi.org/10.11646/zootaxa.4227.1.4
Mizsei, E., Szabolcs, M., Dimaki, M., Roussos, S. A., & Ioannidis, Y. (2018a). Vipera graeca . The IUCN Red List of Threatened Species 2018: e.T53789650A53790137.
Mizsei, E., Szabolcs, M., Szabó, L., Boros, Z., Mersini, K., Roussos, S. A., Dimaki, M., Ioannidis, Y., Végvári, Z., & Lengyel, S. (2020b). Determining priority areas for an Endangered cold-adapted snake on warming mountaintops. Oryx, 1–10. https://doi.org/10.1017/S0030605319000322
Mizsei, E., Zinenko, O., Sillero, N., Ferri, V., Roussos, S. A., & Szabolcs, M. (2018b). The distribution of meadow and steppe vipers (Vipera graeca , V. renardi and V. ursinii ): a revision of the New Atlas of Amphibians and Reptiles of Europe.Basic and Applied Herpetology, 32, 77–83. https://doi.org/10.11160/bah.94
Morris, D. W. (2003). Toward an ecological synthesis: a case for habitat selection. Oecologia, 136, 1–13. https://doi.org/10.1007/s00442-003-1241-4
Muri, D., Schuerch, J., Trim, N., Golay, J., Baillifard, A., El Taher, A., & Dubey, S. (2015). Thermoregulation and microhabitat choice in the polymorphic asp viper (Vipera aspis ). Journal of Thermal Biology, 53, 107–112. https://doi.org/10.1016/j.jtherbio.2015.06.009
Mushinsky, H. R., & McCoy, E. D. (2016). Measuring microhabitats used by non-avian reptiles. Reptile Ecology and Conservation: A Handbook of Techniques. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198726135.003.0019
Nemes, S., Vogrin, M., Hartel, T., & Öllerer, K. (2006). Habitat selection at the sand lizard (Lacerta agilis ): ontogenetic shifts. North-Western Journal of Zoology, 2, 17–26.
Nilson, G., & Andrén, C. (2001). The meadow and steppe vipers of Europe and Asia – the Vipera (Acridophaga ) ursiniicomplex. Acta Zoologica Academiae Scientarium Hungaricae, 47,87–267.
Oksanen, J., Blanchet, G. F., Friendly, M., Kindt, R., Legendre, R., McGlinn, D., Minchin, P. R., O’Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M. H., Szoecs, E., & Wagner, H. (2019). vegan: Community Ecology Package. R package version 2.5-5. https://CRAN.R-project.org/package=vegan
Olivier, F., & Wotherspoon, S. J. (2006). Modelling habitat selection using presence-only data: Case study of a colonial hollow nesting bird, the snow petrel. Ecological Modelling, 195, 187–204. https://doi.org/10.1016/j.ecolmodel.2005.10.036
Péchy, T., Halpern, B., Sós, E., & Walzer, C. (2015). Conservation of the Hungarian meadow viper Vipera ursinii rakosiensis .International Zoo Yearbook, 49, 89–103. https://doi.org/10.1111/izy.12088
Philips, S. J., Dudík, M., Elith, J., Graham, C. H., Lehmann, A., Leathwick, J., & Ferrier, S. (2009). Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. Ecological Applications, 191, 81–197. https://doi.org/10.1890/07-2153.1
Pianka, E. R. (1973). The structure of lizard communities. Annual Review of Ecology and Systematics, 4, 53-74. https://doi.org/10.1146/annurev.es.04.110173.000413
Pringle, R. M., Webb, J. K., & Shine, R. (2003). Canopy structure, microclimate, and habitat selection by a nocturnal snake,Hoplocephalus bungaroides . Ecology, 84, 2668–2679. https://doi.org/10.1890/02-0482
R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
Rosenzweig, M. L. (1981). A theory of habitat selection. Ecology, 62, 2327–2335. https://doi.org/ 10.2307/1936707
Ruiz-Jaen, M. C., & Aide, M. T. (2005). Restoration success: how is it being measured? Restoration Ecology, 13, 569–577. https://doi.org/10.1111/j.1526-100X.2005.00072.x
Stein, A., Gerstner, K., & Kreft, H. (2014). Environmental heterogeneity as a universal driver of species richness across taxa, biomes and spatial scales. Ecology Letters, 17, 866–880. https://doi.org/10.1111/ele.12277
Stumpel, A. H. P., & van der Werf, B. D. C. (2012). Reptile habitat preference in heathland: implications for heathland management.Herpetological Journal, 22, 179–182.
Tews, J., Brose, U., Grimm, V., Tielborger, K., Wichmann, M. C., Schwager, M., & Jeltsch, F. (2004). Animal species diversity driven by habitat heterogeneity/diversity: the importance of keystone structures.Journal of Biogeography, 31, 79–92. https://doi.org/10.1046/j.0305-0270.2003.00994.x
Triska, M. D., Craig, M. D., Stokes, V. L., Pech, R. P., & Hobbs, R. J. (2016). The relative influence of in situ and neighbourhood factors on reptile recolonization in post-mining restoration sites.Restoration Ecology, 24, 517–527.https://doi.org/10.1111/rec.12340
Tupikov, A. I., & Zinenko, O. I. (2015). Distribution of the steppe viper Vipera renardi (Reptilia, Viperidae) in Kharkiv region. Visnyk of Dnipropetrovsk University. Biology, Ecology, 23,172–176.
Vermeire, L. T., & Gillen, R. L. (2001). Estimating herbage standing crop with visual obstruction in tallgrass prairie. Journal of Range Management, 54, 57–60. https://doi.org/10.2458/azu_jrm_v54i1_vermeire
Volesky, J. D., Schacht, W. H., & Reece, P. E. (1999). Leaf area, visual obstruction, and standing crop relationships on Sandhills rangeland. Journal of Range Management, 52, 494–499.
Wilgers, D. J., & Horne, E. A. (2007). Spatial variation in predation attempts on artificial snakes in a fire-disturbed tallgrass praire.Southwestern Naturalist, 52, 263–270. https://doi.org/10.1894/0038-4909(2007)52[263:SVIPAO]2.0.CO;2
Zinenko, O., Stümpel, N., Mazanaeva, L., Bakiev, A., Shiryaev, K., Pavlov, A., Kotenko, T., Kukushkin, O., Chikin, Y., Duisebayeva, T., Nilson, G., Orlov, N. L., Tuniyev, S., Ananjeva, N. B., Murphy, R. W., & Joger, U. (2015). Mitochondrial phylogeny shows multiple independent ecological transitions and northern dispersion despite of Pleistocene glaciations in meadow and steppe vipers (Vipera ursinii andVipera renardi ). Molecular Phylogenetics and Evolution, 84, 85–100. https://doi.org/10.1016/j.ympev.2014.12.005
ACKNOWLEDGEMENTS
We thank Galyna Mykytynets, Csaba Vadász, the volunteers of the Greek Meadow Viper Working Group, the Conservation Herpetology Lab and the LIFE HUNVIPHAB project for their assistance in the field. EM and GR were supported by the ÚNKP-19-3-II-DE-46 and ÚNKP-20-1-I-ÁTE-1 grants respectively within the frame of the New National Excellence Program of the Ministry for Innovation and Technology (Hungary). EM, MS, and SL were funded by a grant from the National Research, Development and Innovation Office of Hungary (OTKA K 134391). Collecting data onVipera ursinii were financially supported by the European Commission (LIFE HUNVIPHAB LIFE18 NAT/HU/000779), by the Hungarian Ministry of Agriculture and the Kiskunság National Park Directorate.
CONFLICT OF INTEREST
The authors declare no confilct of interest.
AUTHORS’ CONTRIBUTION
E.M., M.B., G.R., G.D. and S.L. conceived the ideas and designed methodology; E.M., M.B., G.R., B.B., D.R., M.S. and A.M. collected data; E.M. programmed image processing and analysed the data, E.M. and S.L. led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.
DATA AVAILABILITY STATEMENT
This study was conducted via the implementation of an R script consisting automated image processing to measure leaf area (LA), height of closed vegetation (HCV), maximum height of vegetation (MHC), and foliage height diversity (FHD). The R script, example images and example data are available at Zenodo (url).
TABLES
Table 1. Number of snakes found (presence locations), search effort, density of snakes and number of random locations studied in the three species.