Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Global analysis of the influence of environmental variables to explain ecological niches and realized thermal niche boundaries of sea snakes

  • Carlos Patrón-Rivero ,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Software, Validation, Writing – original draft, Writing – review & editing

    c.patron.rivero@comunidad.unam.mx (CPR); carlos_yanez@ciencias.unam.mx (CYA)

    Affiliation Laboratorio de Ecología Geográfica, Unidad de Conservación de la Biodiversidad, UMDI-Sisal, Facultad de Ciencias, Universidad Nacional Autónoma de Mexico, Sierra Papacal, Yucatán, Mexico

    ⨯
  • Luis Osorio-Olvera,

    Roles Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Laboratorio de Ecoinformática de la Biodiversidad, Departamento de Ecología de la Biodiversidad, Instituto de Ecología, Universidad Nacional Autónoma de México, Circuito Exterior s/n Anexo al Jardín Botánico, Ciudad Universitaria, Coyoacán, Ciudad de México, México

    ⨯
  • Octavio Rojas-Soto,

    Roles Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Laboratorio de Bioclimatología, Red de Biología Evolutiva, Instituto de Ecología A.C., Xalapa, Veracruz, México

    ⨯
  • Xavier Chiappa-Carrara,

    Roles Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Departamento de Sistemas y Procesos Naturales, Escuela Nacional de Estudios Superiores Unidad Mérida, Universidad Nacional Autónoma de Mexico, Ucú, Yucatán, Mexico

    ⨯
  • Fabricio Villalobos,

    Roles Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Laboratorio de Macroecología Evolutiva Red de Biología Evolutiva, Instituto de Ecología, A.C, Xalapa, Veracruz, México

    ⨯
  • Brooke Bessesen,

    Roles Methodology, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of Reading, Reading, United Kingdom

    ⨯
  • Kevin López-Reyes,

    Roles Data curation, Formal analysis, Validation, Writing – review & editing

    Affiliation Laboratorio de Ecología Geográfica, Unidad de Conservación de la Biodiversidad, UMDI-Sisal, Facultad de Ciencias, Universidad Nacional Autónoma de Mexico, Sierra Papacal, Yucatán, Mexico

    ⨯
  • Carlos Yañez-Arenas

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    c.patron.rivero@comunidad.unam.mx (CPR); carlos_yanez@ciencias.unam.mx (CYA)

    Affiliation Laboratorio de Ecología Geográfica, Unidad de Conservación de la Biodiversidad, UMDI-Sisal, Facultad de Ciencias, Universidad Nacional Autónoma de Mexico, Sierra Papacal, Yucatán, Mexico

    ⨯

Abstract

Understanding the factors affecting species distributions is a central topic in ecology and biogeography. However, most research on this topic has focused on species inhabiting terrestrial environments. At broad scales, abiotic variables consistently serve as primary determinants of species’ distributions. In this study, we investigated the explanatory power of different abiotic variables in determining the distribution patterns of sea snakes on a global scale. Additionally, as the boundaries of realized thermal niches have significant implications for the ecology of species and their geographic distributions, we evaluated the asymmetry of realized thermal limits (i.e., differences in variances between the upper and lower limits of the realized thermal niche). We obtained 10 marine environmental variables from global databases along with >5000 occurrence records for 51 sea snake species in 4 genera across the group’s entire known geographic range. Using these data, we employed correlative ecological niche modeling to analyze the influence of the individual variables in explaining species’ distributions. To estimate the realized thermal limits of each species, we extracted the mean, minimum, and maximum temperature values at four depths (superficial, mean benthic, minimum benthic, and maximum benthic) for each occurrence record of the species. We then evaluated the asymmetry of the realized thermal niche by measuring and comparing the variances in the upper and lower limits. Both analyses (the importance of variables and realized thermal limit asymmetry) were performed at three taxonomic levels (sea snakes as a lineage of marine-adapted elapids [true sea snakes + sea kraits], subfamily, and genus) and two spatial resolutions. Overall, we found that temperature, silicate, nitrate, salinity, and phosphate concentrations were the most influential factors in explaining the spatial distribution patterns of sea snakes, regardless of taxonomic level or spatial resolution. Similarly, we observed that the realized thermal limits were asymmetric, with a higher variance in the lower limits, and that asymmetry decreased as the taxonomic level and spatial resolution increased.

Citation: Patrón-Rivero C, Osorio-Olvera L, Rojas-Soto O, Chiappa-Carrara X, Villalobos F, Bessesen B, et al. (2024) Global analysis of the influence of environmental variables to explain ecological niches and realized thermal niche boundaries of sea snakes. PLoS ONE 19(12): e0310456. https://doi.org/10.1371/journal.pone.0310456

Editor: Michael Schubert, Laboratoire de Biologie du Développement de Villefranche-sur-Mer, FRANCE

Received: May 3, 2024; Accepted: September 2, 2024; Published: December 5, 2024

Copyright: © 2024 Patrón-Rivero et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The presence records used for this work and the code for running the analyses can be accessed through the following GitHub repository: https://github.com/PatronRiveroC/Ssnakes. The environmental layers used in this research can be accessed via Bio-Oracle (www.bio-oracle.org) and MARSPEC (www.marspec.org).

Funding: This study was partially supported by the Consejo Nacional de Ciencia y Tecnología (CONACYT) through the grant CVU: 1010703. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Anthropogenic activities have led to rapid global changes that create unprecedented climatic conditions, which could compromise the viability of many species [1, 2]. Understanding how environmental conditions affect the survival of species’ populations and shape their distributions is fundamental for establishing conservation strategies [3]. However, estimating the relationships between species and their environments is a complex task, as it involves multiple factors [4–6]. These factors can be summarized into three categories: 1) limiting or regulatory factors, which control the ecophysiological responses of the species, 2) disturbance factors, which describe the historical modifications in the occupation of habitats by both natural and anthropogenic processes, and 3) resource factors, which represent the supplies necessary for the survival of organisms [7]. Some causal relationships between these factors and species’ attributes, such as abundance and distribution, have been demonstrated through physiological experiments in laboratories and wildlife demographic studies [8–10]. Conducting laboratory experiments or fieldwork to obtain demographic data has certain drawbacks, however. First, both are generally costly and time-consuming. Second, implementing physiological experiments is not ethically feasible for many organisms [11]. Finally, some environmental variables are found to be more relevant at broad spatial scales (>200 km); for instance, limiting or regulating factors associated with climate are considered most relevant in explaining the distribution of species across their entire range [12–14].

An alternative approach to studying species-environment relationships in a spatial context is the use of correlative methods, known generically as ecological niche modeling (ENM) and species distribution modeling (SDM) [14–16]. The inputs required by correlative methods are easier to acquire, because they are generally available in online repositories. These inputs are sets of georeferenced localities where the presence of the species in question has been observed (some methods also require points that represent absence), and spatial surfaces that describe the environmental conditions of the area of interest [17]. Thus, correlative methods are extensively used to model species-environment relationships and estimate their actual and potential distributions, both in terrestrial [18, 19] and marine ecosystems [3, 20]. Although correlative methods have certain limitations [e.g., it is difficult to discern whether their results are merely statistical associations or causal relationships; see 21], they allow us to recognize diverse ecological patterns [22]. As a result, these methods have become popular for analyzing the importance of environmental variables as limiting factors for distribution ranges and various ecological aspects across multiple taxa [3, 20, 23].

One of the most studied limiting factors is temperature, which is important for many physiological and ecological processes [10, 24, 25]. Thus, studying the thermal niche (TN) and realized thermal niche (RTN) of species is a major focus of ecological research [23, 26–28]. According to Gvoždík [26], TN refers to the range in body temperature that individuals of a species require for the population to experience positive growth, whereas RTN refers to the environmental range of temperatures to which individuals of a species are exposed [28, 29]. Based on the concepts of TN and RTN, two hypotheses have been proposed to explain the patterns of temperature tolerance and organismal response to global warming. The first hypothesis suggests that tropical and polar terrestrial species have lower TN than those from temperate climates [30, 31]. The second hypothesis is realized thermal limit asymmetry, which refers to significant differences in the variation of upper limits in a taxon compared with the variation of lower limits within the same taxon, with the variation being greater in the lower limits. Lower variances in upper thermal limits have been reported to severely impact marine species, which may face increasing sea temperatures under climate change scenarios with the potential to impact both survivability and habitat suitability [28, 29, 32–34]. Ideally, evaluating these hypotheses requires direct estimation of the species’ TN through physiological population growth-mortality experiments [35–38], and so despite the high cost, time and ethical constraints, there continue to be global initiatives to analyze those patterns [39]. Unsurprisingly, the characterization of the TNs for most taxa is still lacking.

Because correlative methods do not require experimentation and management of organisms, they are a more feasible option for studying species’ TNs. Consequently, they have been commonly applied to many taxonomic groups [37, 40–42]. Most TN analyses have focused on species from terrestrial environments [39], though using correlative methods, it is possible to characterize the RTN as the range of temperatures associated with the presence of a species [43–45], allowing investigations of many important marine groups, such as sea snakes.

Sea snakes are an ecologically diverse and marine-adapted lineage of elapids that includes two subfamilies: Hydrophiinae (true sea snakes) and Laticaudinae (sea kraits) [46–48]. This group comprises >70 species [49] that are widely distributed in the tropical and subtropical regions of the Indian and Pacific Oceans [50, 51]. Sea snake lineages have evolved unique physiological adaptations to survive in marine environments [52, 53]. Research has shown temperature to play a critical role in their distribution [54–56]. Moreover, individuals exhibit different diving behaviors and metabolic rates in response to changes in water temperature [57, 58]. These findings highlight the complex ways in which temperature influences the ecology and physiology of sea snakes; yet, to date, there is limited information on the asymmetry of their realized thermal limits or the broader environmental factors that influence their distribution. In this study, we used correlative ecological niche modeling (ENM) to evaluate the relative importance of environmental variables in explaining the distribution of a majority of sea snake species globally. We also analyzed whether there was asymmetry in the dispersion of the lower and upper realized thermal limits. We underscore the importance of understanding the thermal biology of these reptiles for conservation and management efforts. The results of our research provide a better understanding of the biogeographical phenomena that determine the distribution patterns in this group and will allow us to infer their vulnerability to future ocean warming scenarios.

Materials and methods

Presence records

We searched the presence records with geographic coordinates for all sea snakes in the public online repositories following the available phylogenies and agreements [49, 59–61]. These platforms are the Global Biodiversity Information Facility (gbif.org; DOI information associated with data are available in S1 Table), VertNet (vertnet.org), the Ocean Biogeographic Information System (obis.org), the Atlas of Living Australia (ala.org.au), the Online Zoological Collections of Australian Museums (ozcam.org.au) and the UNAM open data portal (datosabiertos.unam.mx). We integrated the presence records obtained from the different platforms into a single database for each species. We then cleaned the data by eliminating duplicate records, uncertain data (e.g., taxonomic mistakes, records outside the known distribution of the group), and obvious errors (e.g., records in terrestrial environments). Next, we eliminated occurrences that might represent sink populations through an environmental outliers’ analysis. In this process, we considered the environmental variables previously identified in the literature as important to the group (see below), using the outliers function of the “biogeo” package [62] in R [63], which searches for presence records that are located at a distance above 1.5 times the interquartile range of the data. Additionally, we employed the gridSample function of the “dismo” package [64] to minimize data bias towards more intensely sampled environments. Finally, we split the data that include historical and other with only contemporary records for constructing the M area and for the modeling process, respectively (see below). These processed records established a study database of >5000 usable presence records comprising a total of 51 species in four genera: Aipysurus, Emydocephalus, Hydrophis, and Laticauda, with record counts by species ranging from 5–788 (x̄ = 100; Table 2).

Environmental data

From the global repositories Bio-Oracle [65] and MARSPEC [66], we downloaded raster formats of environmental variables commonly used in spatial modeling of marine environments: calcite (Cal), current velocity (Cvel), dissolved molecular oxygen (Doxy), iron (Iro), nitrate (Nit), pH, phosphate (Pho), salinity (Sal), silicate (Sil), and temperature (Tem) obtained at four depths (superficial, benthic maximum, benthic mean, and benthic minimum). We then classified the variables into five groups: Bsurf (superficial Bio-Oracle; n = 10), Bma (maximum benthic Bio-Oracle; n = 8), Bme (mean benthic Bio-Oracle; n = 8), Bmi (minimum benthic Bio-Oracle n = 8), and Msurf (superficial MARSPEC; n = 6). This protocol was used to select the combination of repository/depth that maximized the explanatory capacity of the predictors (Table 1). To standardize the spatial resolution of the variables, we resampled the MARSPEC variables to match the 5 arc-minutes (arcmin) cell resolution of Bio-Oracle. Subsequently, we resampled both repositories at a resolution of 10 arcmin using the same method to assess the effect of spatial resolution on our results [67].

thumbnail
Download:
Table 1. Variable groups with description of each individual variable by repository and depth (in parenthesis).

https://doi.org/10.1371/journal.pone.0310456.t001

For the study range of each species, we defined a polygon representing a hypothesis of historically accessible areas [area M; sensu 68] based on the marine biogeographical provinces of the world [69]. We first selected all provinces with at least one presence record of the species in question and then we took species dispersal into account, using ocean currents information (earth.nullschool.net/#current/ocean/surface/currents/patterson) [70]. If the M area appeared fragmented into “provinces,” we systematically evaluated the likelihood of those provinces being connected. For provinces containing numerous and also widely distributed presence records, we hypothesized that ocean currents could serve as dispersal channels creating connectivity between them [71]. In this case, we added provinces to form a unified area. Meanwhile, for species with unlinked provinces showing only isolated records, we considered currents as potential physical barriers and removed any provinces between. We also defined a calibration area and masked the variables by selecting only the areas within the intersection between the presence records and the marine eco-regions, a finer sub-regionalization than the provinces. Finally, to reduce collinearity and dimensionality of predictors, we eliminated individually for each species, variables in each of the five groups that had a pairwise Pearson correlation >0.8 with the function correlation_finder in “ntbox” package [72–74].

Ecological niche models

To characterize the ecological niches of sea snakes, we used the MaxEnt 3.4.1 algorithm [75]. We used the jackknife cross-validation procedure to improve the accuracy and minimize the variability of the models in taxa with 5–25 localities. In this method, one presence record is excluded randomly from the modeling, and the procedure is repeated as many times as the data allow; all records are excluded just once. In each iteration, n-1 data were used as model training information, and the excluded record was used for model evaluation [76]. Finally, for species with more than 25 occurrence records, we applied the get.checkerboard1 function in the “ENMeval” package [77]. In this method, the database was divided into two sets using a checkerboard pattern across the study area. Occurrences were separated according to their position on the board. We defined the set with more records as training data and the set with less information as the evaluation data [78].

A series of candidate models were calibrated with the training database for each species via the “kuenm” package [79]. This package allows for varying the parameter combinations of the MaxEnt algorithm. We tested combinations of seven feature classes in MaxEnt (“l”, “q”, “p”, “lq”, “lqp”, “lp”, and “qp”, where “l” is linear, “q” is quadratic, and "p" is product), eight regularization multipliers (0.10, 0.25, 0.50, 0.75, 1, 2, 3 and 4), and the five groups of environmental variables described above (thus, the maximum possible number of variables used to build any model is full set of “repository/depth” if no correlation between all variables were found). This resulted in 560 candidate models per species (spp × 2 resolutions × 7 Fs × 8 RMs × 5 groups of variables). Then we selected using the test database the best subset of models that met the following criteria hierarchically: 1) statistical significance, noting that due to limitations in SDMs and ENMs related to the sensitivity of the area used in model construction, we did not rely on the traditional AUC for model evaluation; instead, we utilized the partial AUC ratio, which penalizes the quantity of area involved in the estimation and is a more informative metric, widely accepted in the literature [80]; we retained those models that were better than expected by chance, depending on the proportion of bootstrap replicates with ratios of the partial area >1 [81]; 2) predictive capacity, where we selected models that were also able to predict at least 90% of the evaluation records (i.e., models with an omission rate–OR–≤ 0.10); and 3) complexity, by using Akaike information criterion corrected for small samples (ΔAICc) we selected as best combinations those ≤ 2 units, that are also the models that present a better fit and fewer parameters [82, 83]. Based on the best parameter combinations (full set of best combinations are available in S2 Table), we built a final set of models using the entire database (training and evaluation data) with the bootstrap functionality of MaxEnt performing 10 replicates. In each iteration, we randomly divided the presence records into 80% for training and 20% for evaluation, and defined Cloglog as output format with 10,000 background points (masked by our calibration area). Finally, we calculated the median and range of the predicted values across the total replicates (10 replicates × final parameter combination) to represent the consistency and variation in the predictions.

Importance of variables

To analyze the relative importance of environmental variables in explaining sea snakes’ distributions, we first classified the variables into ten groups based on their identity. For example, the group “Temperature (Tem)” included seven variables that represented the median of the mean, maximum and minimum values of the sea surface and benthic temperature (Table 1).

We used two measures to assess the relative importance of the environmental variables as estimated by MaxEnt: percentage of contribution (PC) and permutation of importance (PI). PC measures the gain of a model built with a single specific variable divided by the total gain of a model built with all variables. PI is obtained by randomizing the values of each variable so that this variable is not informative and then measuring the resulting drop in the area under the curve [84]. To identify the consistency or variability in our results, we performed this procedure at three levels: “lineage”, where we analyzed the variables for all species regardless of subfamily; “subfamily”, where we divided the results into the subfamilies Hydrophiinae and Laticaudinae [46–48]; and “genus”, where we repeated the same analysis for each of the four genera. We performed this analysis at both resolutions (5 and 10 arcmin). We also separated the identity variables into two groups (high relative importance and low relative importance) based on their relative importance using the natural Jenks thresholds calculated with the getJenksBreaks function of the “BAMMtools” package [85].

To determine if the results depended on spatial resolution, we evaluated the consistency of the importance of each variable between the two spatial resolutions for the two MaxEnt metrics (PC and PI) through a non-parametric Mann-Whitney-Wilcoxon test using the wilxicon.test function in “stats” package [63]. Overall, this approach allowed us to identify the most important environmental variables driving the distribution of sea snakes and to assess the consistency of these results across different metrics of importance, taxonomic levels, and spatial resolutions.

Asymmetry of the realized thermal niche (RTN)

To assess asymmetry in the RTNs, we extracted the mean, maximum, and minimum temperature values for the five groups depending on the repository and depth (Bsurf, Bma, Bme, Bmi, and Msurf) associated with each post processed presence record (see above in “2.1 Presence records”) of each species at 5 and 10 arcmin of spatial resolution. For Bio-Oracle, we directly used the mean, maximum, and minimum temperature variables at each depth. For MARSPEC, we used the surface temperature of the warmest month to represent the maximum temperature and the surface temperature of the coldest month to represent the minimum temperature. Using these values, we conserved and grouped the minimum and maximum values from each species to characterize the lower and upper RTN limits respectively.

Finally, we tested the asymmetry in the RTN limits hypothesis (asymmetry in the variances with more variance in the lower limit) by testing the homogeneity of the variances between the lower and upper realized thermal limits. This process was repeated for each variable group at each temperature measurement at the three taxonomic levels and two spatial resolutions by applying a non-parametric Fligner-Killeen test [86] (5 groups × 3 temperatures × 3 taxonomic levels × 2 resolutions). Importantly, at the genus level, we excluded Laticauda as it contains the identical species as the Laticaudinae subfamily.

Results

Of our 5,023 spatially unique presence records for 51 of the >70 recognized sea snake species [49, 60, 87], most were distributed in the Indo-Pacific Ocean, with Hydrophis platurus being the only species found in the eastern Pacific Ocean of the New World.

Regarding model evaluation, we obtained a median partial AUC ratio >1 for all species (i.e., all our models were better than expected by chance). The ORs in the final models were ≤ 0.10, with the exception of three species when using the variables at 5 arcmin resolution and six at 10 arcmin (Table 2).

thumbnail
Download:
Table 2. Summary of inputs and evaluations from ecological niche modeling.

https://doi.org/10.1371/journal.pone.0310456.t002

Importance of variables

We present all results of relative importance for each taxonomic level (lineage, subfamily, and genus) in Fig 1 (the results for each species are shown in S1, S2 Figs in S1 File). A summary of the relatively high important variables based on PC and PI as estimated by MaxEnt at 5 and 10 arcmin of spatial resolution is presented in Fig 2.

thumbnail
Download:
Fig 1. Variable contribution estimated by MaxEnt percentage of contribution (PC) and permutation of importance (PI) for each taxonomic level at both spatial resolutions.

A) = 5 arcmin, B) = 10 arcmin. Darker colors denoted higher contribution while lighter colors less contribution.

https://doi.org/10.1371/journal.pone.0310456.g001

thumbnail
Download:
Fig 2. Chord diagram for the relative important variables under the percentage of contribution (PC) and permutation of importance (PI) estimated by MaxEnt and their representation in different taxonomic and spatial resolution levels.

A) = PC at 5 arcmin, B) = PI at 5 arcmin, C) = PC at 10 arcmin, D) = PI at 10 arcmin, Nit = nitrates, Pho = phosphates, Sal = salinity, Sil = silicates, Tem = temperature. More connected variables represent that are relatively important variables in more taxonomic and spatial resolution levels comparing to variables less connected.

https://doi.org/10.1371/journal.pone.0310456.g002

Lineage.

The most important variables for sea snakes as a single lineage at a resolution of 5 arcmin, based upon highest PC, were Pho, Sal, and Tem. However, when the importance of variables was permuted (PI), Nit, Sil, and Tem emerged as the most important factors. For both PC and PI at 10 arcmin, Sil and Tem were important variables, while Nit was also important for the PI. Statistically comparing PC against PI, we found differences only in Sal at both resolutions (S3A, S3B Table). When comparing PC between resolutions we found that Sil was statistically different, while comparisons in PI did not show any differences between resolutions (S3C, S3D Table).

Subfamily.

The patterns of variable importance differed at the subfamily level. For the Hydrophiinae subfamily at 5 arcmin, the most important variables for the PC were Sal and Tem, while the most important for the PI were Nit, Sil and Tem. At 10 arcmin, Sil and Tem were the most important variables for the PC and Nit and Sil for the PI. We note that statistical differences in Sal were recorded between PC and PI at both 5 and 10 arcmin (S3A, S3B Table). For the Laticaudinae subfamily at 5 arcmin, the variables with highest PC were Nit, Pho, and Sal, while the variables with highest PI also included Tem. At 10 arcmin, Sil and Tem were the most important variables for both the PC and PI. Meanwhile, Tem differed between PI at both resolutions (S3D Table).

Genus.

At the genus level, for Aipysurus at 5 arcmin, Cal, Doxy, Iro, Pho, Sal and Tem were the most important variables for the PC, while Doxy, Iro, Pho and Tem were the most important for the PI. At 10 arcmin, Sil was the most important variable for the PC and Doxy, Pho and Sil for the PI. In Emydocephalus at 5 arcmin for the PC, Doxy, Sal, Sil and Tem were the most important variables, while Sil and Tem were the most important for the PI. At 10 arcmin, Cvel, Sal, Sil, and Tem were the most important variables for the PC and the same variables omitting Sal for the PI.

For Hydrophis at 5 arcmin, Nit, Pho, Sal, and Tem were the most important variables for the PC, while Nit and Tem were the most important for the PI. At 10 arcmin, Nit, Sil, and Tem were the most important variables for both the PC and PI, though Pho also proved important for PC. Additionally, we found differences between Sal when comparing PC against PI at both 5 and 10 arcmin (S3A, S3B Table). For Laticauda, the results were similar to those obtained at the subfamily level (containing all species as Laticaudinae; see 3.1.2).

Variation in the RTN limits

At the lineage level, our results reveal significant differences in the variance of the realized lower thermal limit compared to the realized upper thermal limit across all temperature measurements (mean, maximum, and minimum), regardless of the group of variables used (Bma, Bme, Bmi, Bsurf, and Msurf) or cell resolution (Fig 3A, 3B; S4A, S4B Table). The same pattern was observed in the Hydrophiinae subfamily (Fig 3C, 3D; S4A, S4B Table), whereas for the Laticaudinae subfamily, we found asymmetry with all temperatures in Bmi at both resolutions, only in the minimum temperature measure in Bsurf at both resolutions, and in Bma in minimum temperature at 10 arcmin (Fig 3E, 3F; S4A, S4B Table).

thumbnail
Download:
Fig 3. Lower and upper thermal limits of sea snakes at lineage and subfamily taxonomic levels, three temperatures measures, five variable groups and two spatial resolutions.

A) = Sea snakes’ lineage at 5 arcmin, B) = Sea snakes’ lineage at 10 arcmin, C) = Hydrophiinae subfamily at 5 arcmin, D) = Hydrophiinae subfamily at 10 arcmin, E) = Laticaudinae subfamily at 5 arcmin, F) = Laticaudinae subfamily at 10 arcmin, Max = maximum temperature, Mean = mean temperature, Min = minimum temperature, Bma = maximum benthic Bio-Oracle, Bme = mean benthic Bio-Oracle, Bmi = minimum benthic Bio-Oracle, Bsurf = superficial Bio-Oracle, Msurf = superficial MARSPEC. p values are in S4 Table.

https://doi.org/10.1371/journal.pone.0310456.g003

Statistically comparing the lower and upper RTN limits at the genus level for Aipysurus, we did not observe differences, except in mean and minimum surface temperatures at both resolutions. For Emydocephalus, we did not find significant differences between the lower and upper limits at any temperature, repository/depth, or spatial resolution. Conversely, for Hydrophis, all temperatures and repository/depths at both spatial resolutions showed statistical differences, which was the same pattern as for the Hydrophiinae subfamily and sea snake lineage as a whole (Fig 4; S4A, S4B Table).

thumbnail
Download:
Fig 4. Lower and upper thermal limits of sea snakes at genus taxonomic levels, three temperatures measures, five variable groups and two spatial resolutions.

A) Aipysurus genus at 5 arcmin, B) = Aipysurus genus at 10 arcmin, C) = Emydocephalus genus at 5 arcmin, D) = Emydocephalus genus at 10 arcmin, E) = Hydrophis genus at 5 arcmin, F) = Hydrophis genus at 10 arcmin, Max = maximum temperature, Mean = mean temperature, Min = minimum temperature, Bma = maximum benthic Bio-Oracle, Bme = mean benthic Bio-Oracle, Bmi = minimum benthic Bio-Oracle, Bsurf = superficial Bio-Oracle, Msurf = superficial MARSPEC. p values are in S4 Table.

https://doi.org/10.1371/journal.pone.0310456.g004

Discussion

Overview

Our study adds to the growing body of literature on the ecology of sea snakes and provides further insights into the factors that may influence their distribution patterns. This work utilized correlative models to determine the relationships between sea snakes at multiple taxonomic levels and environmental variables. We examined numerous proximal (direct) predictors, including temperature, salinity, dissolved molecular oxygen, pH, benthic current velocity, and several marine nutrients: calcite, iron, nitrate, phosphate, and silicate, and we found that particular environmental variables correspond with the presence of sea snakes. Previous studies with similar correlative methods for several taxa showed temperature, as well the more distal (indirect) predictors of bathymetry, precipitation and distance to shore to be the critical factors in determining the distribution of marine organisms [3, 20]. We did not include distal variables in our study because they often represent a combination of factors influencing a single variable (e.g., distance to shore is likely associated with temperature, salinity, etc.), which limits their explanatory power for species distribution [see 88]. Instead, we focused on environmental variables that have a more direct relationship with the regulatory and resource factors that define a taxon’s range [see 7]. Of course, certain variables appear to be relevant to almost all taxa, while different faunal groups may have different environmental preferences [3, 6, 20, 89, 90]. When looking specifically at sea snakes, our study confirmed temperature, but also salinity (and certain marine nutrients), to be significant variables in determining their distribution. This is in agreement with the findings on Hydrophis platurus [91] in the Tropical Indian and Pacific Oceans, with H. platurus xanthos in Costa Rica [56], and with sea snakes in Australia [55, 92].

Based on our models, temperature appears to be an important predictor at all taxonomic levels and for most species. Ectothermic organisms, such as reptiles, are particularly sensitive to thermal changes due to their reliance on external sources of heat for regulating body temperature [93–95]. Temperature influences metabolic rates, growth rates, and reproductive success of ectothermic organisms [96], and can ultimately influence species distribution [95, 97]. As such, temperature plays a crucial role in shaping the occurrence patterns of sea snakes. The ranges of sea snakes may even expand during warmer periods and contract during cooler ones [98]. Both elevated and lower temperatures affect sea snake ecology and behavior. Elevated temperatures (33–35°C) reduce survival time and disrupt reproductive cycles; breeding often coincides only with warmer periods [98]. Although sea snakes possess genetic sex determination (GSD) rather than temperature-dependent sex determination (TSD) like many reptiles, the thermal environment may still affect functional sex ratios, as observed in terrestrial snakes [98]. Lower temperatures (~7–16°C) generally decrease feeding efficiency, swimming abilities, and orientation capabilities [98]. Also, those changes generate more frequent surfacing and less efficient breathing patterns [58, 99]. These factors could compromise the ability of true sea snakes to acclimate to long-term increases in water temperature [58, 98, 100], while the wider thermal tolerance sensitivity of amphibious sea kraits could prove beneficial than true sea snakes [57, 98].

Salinity was also an influential variable in our environmental niche modeling. Like temperature, it had an almost-universal effect across lineages. Such results are not surprising, since salinity has been indicated to have a direct effect on the transition to marine life, driving evolution in taxa such as sea snakes [53, 101, 102]. Marine snakes developed specialized salt glands modified from sub-lingual glands to regulate salt efficiently [101, 103, 104], though they maintain a critical dependency on fresh water, including rainfall lenses at the ocean surface [91, 105, 106]. Laticauda spp. drink from fresh water sources found on land [105]. Based on their physiological requirements and those of their prey, sea snakes must distribute themselves in areas that facilitate osmotic balance [101], in certain cases avoiding extremely low-saline [100] or high-saline [54] waters. Because in this way salinity can influence their movements [91] it may act as a barrier, while also impacting population health and species diversity [107].

Our sea snake models further indicate that combinations of nutrients such as calcite, iron, nitrate, phosphate, and silicate have relative importance across various taxonomic levels. Nutrient availability is likely to influence the distribution and abundance of sea snake prey [108], as well as the coral reefs that provide habitat for multiple species [98, 109, 110]. This interconnected scenario (availability of nutrients influencing prey types and their habitats) contributes to the diversity of trophic specialization and feeding strategies observed in sea snakes [60, 111, 112], and sometimes even a rapid evolution of head size and body shape [60].

Although our findings provide important insights, caution must be exercised when interpreting the results of correlative models because they cannot establish causality. Some correlations may reflect the direct effects of environmental variables, while others may represent indirect relationships. We used MaxEnt, which can handle collinearity among explanatory variables; however, collinearity may still pose challenges when estimating the importance of explanatory variables [113]. The programmers of MaxEnt incorporated two different metrics (PC and PI) to evaluate the contribution of predictors to the model [84], although there is no consensus in the literature on which is more informative. Most studies use the PC, possibly because it appears first in the algorithm’s results. Halvorsen [114] concluded that the PC is most informative, whereas Searcy and Shaffer [115] suggested that the PI is better for describing the biological properties of species. Further research is necessary to establish causal relationships between environmental factors and the abundance of sea snakes and to gain a definitive understanding of their ecology and behavior.

Variance between taxonomic levels.

Overall, it appears that variable importance for the presence sea snakes varies depending on which level is evaluated, which species are included, and which methods (including raster resolution) are used. In our model outputs for different lineage and subfamily levels, metrics, and resolutions, temperature consistently showed the greatest influence, followed by salinity and the nutrients silicate, nitrate and phosphate. Another study also found distance to the coast and bathymetry to be relevant variables for predicting the presence of Laticaudinae [108].

When looking at the genus level, our Aipysurus models showed dissolved oxygen and iron to be more important than for other groups, though the nutrients calcite, phosphate, and silicate, as well as salinity and temperature, were also influential. Evidence indicates that environmental factors are important in shaping the behavior and distribution of A. laevis [116]. These factors have been linked to sexual dimorphism [117, 118], potentially related to ecological niche differentiation. Males and females may have adapted morphological differences in head size and body shape through dietary or feeding niche separation; specifically, females exhibit larger body cavities and smaller head sizes, which are associated with a decrease in intra-specific competition [117]. For Emydocephalus, we found temperature, silicate, salinity, current velocity and dissolved oxygen to be the most important variables. The relative importance of the different predictors may be influenced by this genus’ trend to occur in discrete patches of suitable habitat (sometimes referred to as “metapopulations”), which are affected by differences in prey abundance, habitat quality, and complexity of reef habitats [119]. Hence, negative trends in population abundance may be correlated to declines in environment resources resulting from anthropogenic modification of the ecosystem [120]. Regarding the importance of temperature, we discovered a discrepancy with a previous study of E. annulatus. It reported no significant effects of annual thermal variation on population densities, but found that water temperature significantly contributed to the species’ distribution [121]. Further research is necessary to resolve these inconsistencies and to enhance our understanding of the importance of environmental variables, such as salinity and other nutrient variables in the distribution and population dynamics of Emydocephalus. The Hydrophis genus was found to have almost identical results to the whole lineage, and as indicated above, the genus Laticauda is synonymous with the subfamily Laticaudinae; hence, these are not discussed separately.

RTN limits

According to our research findings, sea snakes demonstrate greater consistency in the variance of their upper thermal limits than in their lower thermal limits. This aligns with the results of previous studies on terrestrial species [122–124] and marine ectotherms [28, 125, 126]. Such results also align with existing evidence that cold tolerance evolves more rapidly than heat tolerance in both endothermic and ectothermic organisms. This phenomenon is attributed to different adaptations related to climatic extremes throughout the evolutionary history of taxa [127]: there is a consistent pattern of heat tolerance, whereas cold tolerance shows significant variability even within species [39, 122, 127]. Sea snakes may too have developed greater plasticity in their lower thermal limits through evolutionary processes [128, 129], resulting in less variation in their upper thermal limits. Moreover, physiology and biophysics may contribute to this asymmetry in tolerances limits and influence RTN limits.

If sea temperatures continue to rise due to climate change, the reduced variability in the upper thermal limits could have significant consequences for sea snakes [122, 130]. Indeed, marine ectotherms have a low physiological acclimatization rate, which makes them particularly susceptible to the effects of climate change [123, 131]. With elongated bodies, sea snakes facilitate rapid heat exchange due to their greater surface in comparison to body mass [132]; however, previous studies have indicated that sea snakes live near their upper thermal limits [98] and are unlikely to evolve increased heat tolerances [122]. Additionally, these thermal niche boundaries are linked to changes in diving and breathing behaviors, leading to more frequent surfacing and less efficient breathing patterns [58, 99]. These factors could compromise the ability of true sea snakes to acclimate to long-term increases in water temperature [58, 98, 100], while greater metabolic thermal tolerance in amphibious sea kraits could prove beneficial to that group [57, 98]. Although our study observed this pattern at the lineage level, we obtained mixed results at the subfamily and genus levels, potentially owing to limited sample size. Other factors, including biotic interactions, geographic barriers, and dispersal processes, may restrict species distribution and affect the capacity to measure complete thermal niches [39, 123, 124].

The consequences of the observed asymmetric realized thermal limits for the physiological performance of sea snakes and their ecological interactions with other species are ecologically significant. Sea snakes are critical predators in marine ecosystems, and alterations in their distribution or abundance can have cascading effects on food webs [133]. Thus, the observed thermal limits of sea snakes have implications not only for their own survival but also for the health and functioning of marine ecosystems. Indeed, our research highlights the importance of conservation efforts and marine management policies that consider the potential vulnerability of sea snakes to changes in sea temperature, including identifying areas that provide thermal refugia for sea snakes, monitoring populations for changes in their distribution or abundance, and implementing strategies that can respond to changing conditions [130], for the benefit of biodiversity. From a research perspective, we note that sample sizes for some subfamilies and genera were relatively small, which may have affected our results, and we encourage future studies to explore the potential for sea snakes to acclimate or adapt to warming temperatures and investigate the mechanisms underlying the observed pattern of asymmetric realized thermal limits. Integrating physiological and genetic data could also help provide a more comprehensive understanding of the thermal ecology of sea snakes and their ability to adapt to climate change.

Conclusions

Temperature, salinity, and some nutrients (such as nitrates, phosphates, and silicates) appear significant in shaping the distribution of sea snakes. Most sea snakes also demonstrate asymmetry in their realized thermal limits, which is to say, they have accumulated greater plasticity in their lower thermal limits through evolutionary processes with less variation in their upper thermal limits. This pattern is consistent with previous research on ectotherms and other endothermic species [122, 130] and highlights the potentially critical implications for sea snakes under scenarios of sea warming. Marine environmental variables have clear bearing on the lives of sea snakes and ENMs that incorporate them can provide valuable insights for conservation management strategies. They can even help guide searches for previously unknown, potentially endangered, populations [130]. Our findings emphasize the importance of generating scientifically robust ecological niche models at multiple scales, which can guide future marine policies for the protection of these vulnerable and ecologically important reptiles.

Supporting information

S1 Table. Access species-specific data via GBIF using the provided DOI references.

This table contains all the DOI facilitated by GBIF to referencing their data.

https://doi.org/10.1371/journal.pone.0310456.s001

(PDF)

S2 Table. Final model parameterizations are outlined for each species based on spatial resolution.

This table contains all the model parametrizations for each species.

https://doi.org/10.1371/journal.pone.0310456.s002

(PDF)

S3 Table. Non-parametric Wilcoxon test comparing the percentage contribution and permutation importance metrics calculated by MaxEnt at 5–10 arc-minute resolution.

This table contains the results of the non-parametric Wilcoxon test of MaxEnt metric across taxa level and spatial resolution.

https://doi.org/10.1371/journal.pone.0310456.s003

(PDF)

S4 Table. Summary of p values for homogeneity of variances of lower and upper realized thermal limits by the minimum, mean, and maximum temperatures across lineage, family, and genus levels at 5 arcmin by non-parametric Fligner-Killeen test.

This table contains the results of the homogeneity of variances of thermal limits by the non-parametric Fligner-Killeen test. (PDF).

https://doi.org/10.1371/journal.pone.0310456.s004

(PDF)

S1 File. Variable contribution estimated by MaxEnt percentage of contribution (PC) and permutation of importance (PI) for each species at 5 and 10 arc-minutes.

These figures show the contribution of each species estimated by MaxEnt metrics at both resolutions.

https://doi.org/10.1371/journal.pone.0310456.s005

(PDF)

Acknowledgments

We appreciate the facilities provided by the Universidad Nacional Autónoma de México (UNAM) and the Laboratorio de Ecología Geográfica (ECOLGEO).

References

  1. 1. Walther G-R, Post E, Convey P, Menzel A, Parmesan C, Beebee TJC, et al. Ecological responses to recent climate change. Nature. 2002;416(6879):389–95. pmid:11919621
  2. 2. Walther GR, Roques A, Hulme PE, Sykes MT, Pysek P, Kühn I, et al. Alien species in a warmer world: risks and opportunities. Trends Ecol Evol. 2009;24(12):686–93. Epub 2009/08/29. pmid:19712994.
  3. 3. Bosch S, Tyberghein L, Deneudt K, Hernandez F, De Clerck O. In search of relevant predictors for marine species distribution modelling using the MarineSPEED benchmark dataset. Diversity and Distributions. 2018;24(2):144–57. https://doi.org/10.1111/ddi.12668.
  4. 4. Gaston KJ, Fuller RA. The sizes of species’ geographic ranges. Journal of Applied Ecology. 2009;46(1):1–9.
  5. 5. Sexton JP, McIntyre PJ, Angert AL, Rice KJ. Evolution and Ecology of Species Range Limits. Annual Review of Ecology, Evolution, and Systematics. 2009;40(1):415–36.
  6. 6. Wiens JJ. The niche, biogeography and species interactions. Philosophical transactions of the Royal Society of London Series B, Biological sciences. 2011;366(1576):2336–50. Epub 2011/07/20. pmid:21768150; PubMed Central PMCID: PMC3130432.
  7. 7. Guisan A, Thuiller W. Predicting species distribution: offering more than simple habitat models. Ecology Letters. 2005;8(9):993–1009. pmid:34517687
  8. 8. Gause GF. The Influence of Ecological Factors on the Size of Population. The American naturalist. 1931;65(696):70–6.
  9. 9. Hutchings JA, Myers RA. Timing of cod reproduction: interannual variability and the influence of temperature. Marine Ecology Progress Series. 1994;108(1):21–32.
  10. 10. Kordas RL, Harley CDG, O’Connor MI. Community ecology in a warming world: The influence of temperature on interspecific interactions in marine systems. Journal of Experimental Marine Biology and Ecology. 2011;400(1):218–26. https://doi.org/10.1016/j.jembe.2011.02.029.
  11. 11. Fraser D. Animal ethics and animal welfare science: bridging the two cultures. Applied Animal Behaviour Science. 1999;65(3):171–89. https://doi.org/10.1016/S0168-1591(99)00090-8.
  12. 12. Contreras-Díaz RG, Falconi M, Osorio-Olvera L, Cobos ME, Soberón J, Townsend Peterson A, et al. On the relationship between environmental suitability and habitat use for three neotropical mammals. Journal of Mammalogy. 2022;103(2):425–39.
  13. 13. Gillespie TW, Foody GM, Rocchini D, Giorgi AP, Saatchi S. Measuring and modelling biodiversity from space. Progress in Physical Geography: Earth and Environment. 2008;32(2):203–21.
  14. 14. Pearson RG, Dawson TP. Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global Ecology and Biogeography. 2003;12(5):361–71. https://doi.org/10.1046/j.1466-822X.2003.00042.x.
  15. 15. Bentlage B, Peterson AT, Barve N, Cartwright P. Plumbing the depths: extending ecological niche modelling and species distribution modelling in three dimensions. Global Ecology and Biogeography. 2013;22(8):952–61. https://doi.org/10.1111/geb.12049.
  16. 16. Peterson A, Soberón J. Species Distribution Modeling and Ecological Niche Modeling: Getting the Concepts Right. J Natureza & Conservação. 2012;10:102–7.
  17. 17. Randin CF, Jaccard H, Vittoz P, Yoccoz NG, Guisan A. Land use improves spatial predictions of mountain plant abundance but not presence-absence. Journal of Vegetation Science. 2009;20(6):996–1008.
  18. 18. Colwell RK, Rangel TF. A stochastic, evolutionary model for range shifts and richness on tropical elevational gradients under Quaternary glacial cycles. Philosophical Transactions of the Royal Society B: Biological Sciences. 2010;365(1558):3695–707. pmid:20980317
  19. 19. Eme D, Malard F, Colson-Proch C, Jean P, Calvignac S, Konecny-Dupré L, et al. Integrating phylogeography, physiology and habitat modelling to explore species range determinants. Journal of Biogeography. 2014;41(4):687–99. https://doi.org/10.1111/jbi.12237.
  20. 20. Bradie J, Leung B. A quantitative synthesis of the importance of variables used in MaxEnt species distribution models. Journal of Biogeography. 2017;44(6):1344–61.
  21. 21. Wiens JA, Stralberg D, Jongsomjit D, Howell CA, Snyder MA. Niches, models, and climate change: Assessing the assumptions and uncertainties. Proceedings of the National Academy of Sciences. 2009;106(Supplement 2):19729. pmid:19822750
  22. 22. Ashcroft MB, French KO, Chisholm LA. An evaluation of environmental factors affecting species distributions. Ecological Modelling. 2011;222(3):524–31. https://doi.org/10.1016/j.ecolmodel.2010.10.003.
  23. 23. Lee-Yaw JA, Kharouba HM, Bontrager M, Mahony C, Csergő AM, Noreen AME, et al. A synthesis of transplant experiments and ecological niche models suggests that range limits are often niche limits. Ecology Letters. 2016;19(6):710–22. pmid:27111656
  24. 24. Bicego KC, Barros RCH, Branco LGS. Physiology of temperature regulation: Comparative aspects. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology. 2007;147(3):616–39. pmid:16950637
  25. 25. Clarke A. Costs and consequences of evolutionary temperature adaptation. Trends in Ecology & Evolution. 2003;18(11):573–81. https://doi.org/10.1016/j.tree.2003.08.007.
  26. 26. Gvoždík L. Just what is the thermal niche? Oikos. 2018;127(12):1701–10.
  27. 27. Magnuson JJ, Crowder LB, Patricia AM. Temperature as an Ecological Resource. American Zoologist. 1979;19(1):331–43.
  28. 28. Stuart-Smith RD, Edgar GJ, Bates AE. Thermal limits to the geographic distributions of shallow-water marine species. Nature Ecology & Evolution. 2017;1(12):1846–52. pmid:29062125
  29. 29. Collin R, Rebolledo AP, Smith E, Chan KYK. Thermal tolerance of early development predicts the realized thermal niche in marine ectotherms. Functional Ecology. 2021;35(8):1679–92. https://doi.org/10.1111/1365-2435.13850.
  30. 30. Pörtner H. Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals. Naturwissenschaften. 2001;88(4):137–46. pmid:11480701
  31. 31. Pörtner HO, Farrell AP. Physiology and Climate Change. Science. 2008;322(5902):690–2.
  32. 32. Nguyen KDT, Morley SA, Lai C-H, Clark MS, Tan KS, Bates AE, et al. Upper Temperature Limits of Tropical Marine Ectotherms: Global Warming Implications. PLOS ONE. 2011;6(12):e29340. pmid:22242115
  33. 33. Pinsky ML, Eikeset AM, McCauley DJ, Payne JL, Sunday JM. Greater vulnerability to warming of marine versus terrestrial ectotherms. Nature. 2019;569(7754):108–11. pmid:31019302
  34. 34. Vinagre C, Dias M, Cereja R, Abreu-Afonso F, Flores AAV, Mendonça V. Upper thermal limits and warming safety margins of coastal marine species–Indicator baseline for future reference. Ecological Indicators. 2019;102:644–9. https://doi.org/10.1016/j.ecolind.2019.03.030.
  35. 35. Dong Y, Somero GN. Temperature adaptation of cytosolic malate dehydrogenases of limpets(genus Lottia): differences in stability and function due to minor changes in sequence correlate with biogeographic and vertical distributions. Journal of Experimental Biology. 2009;212(2):169–77. pmid:19112135
  36. 36. Helmuth B, Broitman BR, Blanchette CA, Gilman S, Halpin P, Harley CDG, et al. Mosaic patterns of thermal stress in the rocky intertidal zone: implications for climate change. Ecological Monographs. 2006;76(4):461–79. https://doi.org/10.1890/0012-9615(2006)076[0461:MPOTSI]2.0.CO;2.
  37. 37. Sánchez-Fernández D, Aragón P, Bilton DT, Lobo JM. Assessing the Congruence of Thermal Niche Estimations Derived from Distribution and Physiological Data. A Test Using Diving Beetles. PLOS ONE. 2012;7(10):e48163. pmid:23133560
  38. 38. Stillman JH, Somero GN. A Comparative Analysis of the Upper Thermal Tolerance Limits of Eastern Pacific Porcelain Crabs, Genus Petrolisthes: Influences of Latitude, Vertical Zonation, Acclimation, and Phylogeny. Physiological and Biochemical Zoology. 2000;73(2):200–8. pmid:10801398
  39. 39. Bennett JM, Calosi P, Clusella-Trullas S, Martínez B, Sunday J, Algar AC, et al. GlobTherm, a global database on thermal tolerances for aquatic and terrestrial organisms. Scientific Data. 2018;5(1):180022. pmid:29533392
  40. 40. Huff DD, Hubler SL, Borisenko AN. Using Field Data to Estimate the Realized Thermal Niche of Aquatic Vertebrates. North American Journal of Fisheries Management. 2005;25(1):346–60.
  41. 41. Kuo ESL, Sanford E. Geographic variation in the upper thermal limits of an intertidal snail: implications for climate envelope models. Marine Ecology Progress Series. 2009;388:137–46.
  42. 42. Sánchez-Fernández D, Rizzo V, Cieslak A, Faille A, Fresneda J, Ribera I. Thermal niche estimators and the capability of poor dispersal species to cope with climate change. Scientific Reports. 2016;6(1):23381. pmid:26983802
  43. 43. Descombes P, Wisz MS, Leprieur F, Parravicini V, Heine C, Olsen SM, et al. Forecasted coral reef decline in marine biodiversity hotspots under climate change. Global change biology. 2015;21(7):2479–87. Epub 2015/01/23. pmid:25611594.
  44. 44. Gaüzère P, Jiguet F, Devictor V. Rapid adjustment of bird community compositions to local climatic variations and its functional consequences. Global change biology. 2015;21(9):3367–78. Epub 2015/03/04. pmid:25731935.
  45. 45. Quintero I, Wiens JJ. Rates of projected climate change dramatically exceed past rates of climatic niche evolution among vertebrate species. Ecology Letters. 2013;16(8):1095–103. pmid:23800223
  46. 46. Lukoschek V, Keogh JS. Molecular phylogeny of sea snakes reveals a rapidly diverged adaptive radiation. Biological Journal of the Linnean Society. 2006;89(3):523–39.
  47. 47. Rasmussen AR, Sanders KL, Guinea ML, Amey AP. Sea snakes in Australian waters (Serpentes: subfamilies Hydrophiinae and Laticaudinae)—a review with an updated identification key. Zootaxa. 2014;3869(4):351–71. pmid:25283923
  48. 48. Smith MA. Monograph of the sea-snakes (Hydrophiidae): London: British Museum (Natural History); 1926.
  49. 49. The Reptile Database [Internet]. 2023 [cited Last accessed 08 March 2024]. Available from: http://www.reptile-database.org.
  50. 50. Sanders KL, Lee MSY, Mumpuni , Bertozzi T, Rasmussen AR. Multilocus phylogeny and recent rapid radiation of the viviparous sea snakes (Elapidae: Hydrophiinae). Molecular phylogenetics and evolution. 2013;66(3):575–91. pmid:23026811
  51. 51. Wallach V, Williams KL, Boundy J. Snakes of the world: a catalogue of living and extinct species: CRC press; 2014.
  52. 52. Dunson WA. The biology of sea snakes 1975.
  53. 53. Heatwhole H. Sea Snakes: University of New South Wales Press, Sidney, Australia; 1999.
  54. 54. Lillywhite HB, Sheehy CM, III, Heatwole H, Brischoux F, Steadman DW. Why Are There No Sea Snakes in the Atlantic? BioScience. 2018;68(1):15–24.
  55. 55. Udyawer V, Oxenham K, Hourston M, Heupel M. Distribution, fisheries interactions and assessment of threats to Australia’s sea snakes. Report to the National Environmental Science Program Tasmania: Marine Biodiversity Hub. 2020.
  56. 56. Bessesen BL, Garrido-Cayul C, González-Suárez M. Habitat suitability and area of occupancy defined for rare New World sea snake. Conservation Science and Practice. 2023;5(1):e12865. https://doi.org/10.1111/csp2.12865.
  57. 57. Dabruzzi T F., Sutton M A., Bennett W A. Metabolic Thermal Sensitivity Optimizes Sea Krait Amphibious Physiology. Herpetologica. 2012;68(2):218–25.
  58. 58. Udyawer V, Simpfendorfer CA, Heupel MR, Clark TD. Coming up for air: thermal dependence of dive behaviours and metabolism in sea snakes. Journal of Experimental Biology. 2016;219(21):3447–54. pmid:27591306
  59. 59. Burbrink FT, Grazziotin FG, Pyron RA, Cundall D, Donnellan S, Irish F, et al. Interrogating Genomic-Scale Data for Squamata (Lizards, Snakes, and Amphisbaenians) Shows no Support for Key Traditional Morphological Relationships. Systematic Biology. 2020;69(3):502–20. pmid:31550008
  60. 60. Sherratt E, Rasmussen AR, Sanders KL. Trophic specialization drives morphological evolution in sea snakes. Royal Society Open Science. 2018;5(3):172141. pmid:29657807
  61. 61. Zaher H, Murphy RW, Arredondo JC, Graboski R, Machado-Filho PR, Mahlow K, et al. Large-scale molecular phylogeny, morphology, divergence-time estimation, and the fossil record of advanced caenophidian snakes (Squamata: Serpentes). PLOS ONE. 2019;14(5):e0216148. pmid:31075128
  62. 62. Robertson MP, Visser V, Hui C. Biogeo: an R package for assessing and improving data quality of occurrence record datasets. Ecography. 2016;39(4):394–401. https://doi.org/10.1111/ecog.02118.
  63. 63. RCoreTeam. R: A language and environment for statistical computing. Foundation for Statistical Computing, Vienna, Austria. 2013.
    • 64. Hijmans RJ, Phillips S, Leathwick J, Elith J, Hijmans MRJJC. Package ‘dismo’. 2017;9(1):1–68.
    • 65. Assis J, Tyberghein L, Bosch S, Verbruggen H, Serrão EA, De Clerck O. Bio-ORACLE v2.0: Extending marine data layers for bioclimatic modelling. Global Ecology and Biogeography. 2018;27(3):277–84.
    • 66. Sbrocco EJ, Barber PH. MARSPEC: ocean climate layers for marine spatial ecology. Ecology. 2013;94(4):979–.
    • 67. Connor T, Viña A, Winkler JA, Hull V, Tang Y, Shortridge A, et al. Interactive spatial scale effects on species distribution modeling: The case of the giant panda. Scientific Reports. 2019;9(1):14563. pmid:31601927
    • 68. Soberon J, Peterson AT. Interpretation of Models of Fundamental Ecological Niches and Species’ Distributional Areas. Biodiversity Informatics. 2005;2(0):10.
    • 69. Spalding MD, Fox HE, Allen GR, Davidson N, Ferdaña ZA, Finlayson M, et al. Marine Ecoregions of the World: A Bioregionalization of Coastal and Shelf Areas. BioScience. 2007;57(7):573–83.
    • 70. Dambach J, Rödder D. Applications and future challenges in marine species distribution modeling. Aquatic Conservation: Marine and Freshwater Ecosystems. 2011;21(1):92–100. https://doi.org/10.1002/aqc.1160.
    • 71. Jiménez LM, Simões N, Yáñez-Arenas C. Biodiversity and conservation of “solar-powered” sea slugs from the Western Atlantic under climate change scenarios. Marine Ecology. 2022;43(3):e12706. https://doi.org/10.1111/maec.12706.
    • 72. Feng X, Park DS, Liang Y, Pandey R, Papeş M. Collinearity in ecological niche modeling: Confusions and challenges. Ecology and Evolution. 2019;9(18):10365–76. pmid:31624555
    • 73. Mateo RG, Vanderpoorten A, Muñoz J, Laenen B, Désamoré A. Modeling Species Distributions from Heterogeneous Data for the Biogeographic Regionalization of the European Bryophyte Flora. PLOS ONE. 2013;8(2):e55648. pmid:23409015
    • 74. Osorio-Olvera L, Lira-Noriega A, Soberón J, Peterson AT, Falconi M, Contreras-Díaz RG, et al. ntbox: An r package with graphical user interface for modelling and evaluating multidimensional ecological niches. Methods in Ecology and Evolution. 2020;11(10):1199–206. https://doi.org/10.1111/2041-210X.13452.
    • 75. Phillips SJ, Anderson RP, Schapire RE. Maximum entropy modeling of species geographic distributions. Ecological Modelling. 2006;190(3):231–59. https://doi.org/10.1016/j.ecolmodel.2005.03.026.
    • 76. Shcheglovitova M, Anderson RP. Estimating optimal complexity for ecological niche models: A jackknife approach for species with small sample sizes. Ecological Modelling. 2013;269:9–17. https://doi.org/10.1016/j.ecolmodel.2013.08.011.
    • 77. Muscarella R, Galante PJ, Soley-Guardia M, Boria RA, Kass JM, Uriarte M, et al. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods in Ecology and Evolution. 2014;5(11):1198–205.
    • 78. Brown JL. SDMtoolbox: a python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods in Ecology and Evolution. 2014;5(7):694–700.
    • 79. Cobos ME, Peterson AT, Barve N, Osorio-Olvera L. kuenm: an R package for detailed development of ecological niche models using Maxent. PeerJ 7:e6281. 2019:15. pmid:30755826
    • 80. Lobo JM, Jiménez-Valverde A, Real R. AUC: a misleading measure of the performance of predictive distribution models. Global Ecology and Biogeography. 2008;17(2):145–51. https://doi.org/10.1111/j.1466-8238.2007.00358.x.
    • 81. Peterson AT, Papeş M, Soberón J. Rethinking receiver operating characteristic analysis applications in ecological niche modeling. Ecological Modelling. 2008;213(1):63–72. https://doi.org/10.1016/j.ecolmodel.2007.11.008.
    • 82. Radosavljevic A, Anderson RP. Making better Maxent models of species distributions: complexity, overfitting and evaluation. Journal of Biogeography. 2014;41(4):629–43.
    • 83. Warren DL, Seifert SN. Ecological niche modeling in Maxent: the importance of model complexity and the performance of model selection criteria. Ecological Applications. 2011;21(2):335–42. pmid:21563566
    • 84. Phillips SJ. A brief tutorial on Maxent. AT&T Research, Princeton, NJ. 2011.
    • 85. Rabosky DL, Grundler M, Anderson C, Title P, Shi JJ, Brown JW, et al. BAMMtools: an R package for the analysis of evolutionary dynamics on phylogenetic trees. Methods in Ecology and Evolution. 2014;5(7):701–7. https://doi.org/10.1111/2041-210X.12199.
    • 86. Fligner MA, Killeen TJ. Distribution-Free Two-Sample Tests for Scale. Journal of the American Statistical Association. 1976;71(353):210–3.
    • 87. Zaher H, Murphy RW, Arredondo JC, Graboski R, Machado-Filho PR, Mahlow K, et al. Correction: Large-scale molecular phylogeny, morphology, divergence-time estimation, and the fossil record of advanced caenophidian snakes (Squamata: Serpentes). PLOS ONE. 2019;14(5):e0217959. pmid:31150527
    • 88. Merow C, Smith MJ, Edwards TC Jr, Guisan A, McMahon SM, Normand S, et al. What do we gain from simplicity versus complexity in species distribution models? Ecography. 2014;37(12):1267–81. https://doi.org/10.1111/ecog.00845.
    • 89. Sexton JP, McIntyre PJ, Angert AL, Rice KJ. Evolution and Ecology of Species Range Limits. Annual Review of Ecology, Evolution, and Systematics. 2009;40(Volume 40, 2009):415–36. https://doi.org/10.1146/annurev.ecolsys.110308.120317.
    • 90. Gillingham PK, Palmer SCF, Huntley B, Kunin WE, Chipperfield JD, Thomas CD. The relative importance of climate and habitat in determining the distributions of species at different spatial scales: a case study with ground beetles in Great Britain. Ecography. 2012;35(9):831–8. https://doi.org/10.1111/j.1600-0587.2011.07434.x.
    • 91. Dunson WA, Ehlert GW. EFFECTS OF TEMPERATURE, SALINITY, AND SURFACE WATER FLOW ON DISTRIBUTION OF THE SEA SNAKE PELAMIS1. Limnology and Oceanography. 1971;16(6):845–53. https://doi.org/10.4319/lo.1971.16.6.0845.
    • 92. Udyawer V, Cappo M, Simpfendorfer CA, Heupel MR, Lukoschek V. Distribution of sea snakes in the Great Barrier Reef Marine Park: observations from 10 yrs of baited remote underwater video station (BRUVS) sampling. Coral Reefs. 2014;33(3):777–91.
    • 93. Seebacher F, Franklin CE. Physiological mechanisms of thermoregulation in reptiles: a review. Journal of Comparative Physiology B. 2005;175(8):533–41. pmid:16047177
    • 94. Hudson SB, Lidgard AD, French SS. Glucocorticoids, energy metabolites, and immunity vary across allostatic states for plateau side-blotched lizards (Uta stansburiana uniformis) residing in a heterogeneous thermal environment. Journal of Experimental Zoology Part A: Ecological and Integrative Physiology. 2020;333(10):732–43. pmid:32959993
    • 95. Kearney M, Porter WP. MAPPING THE FUNDAMENTAL NICHE: PHYSIOLOGY, CLIMATE, AND THE DISTRIBUTION OF A NOCTURNAL LIZARD. Ecology. 2004;85(11):3119–31. https://doi.org/10.1890/03-0820.
    • 96. Kearney MR, Porter WP. NicheMapR–an R package for biophysical modelling: the ectotherm and Dynamic Energy Budget models. Ecography. 2020;43(1):85–96. https://doi.org/10.1111/ecog.04680.
    • 97. Kearney M, Porter W. Mechanistic niche modelling: combining physiological and spatial data to predict species’ ranges. Ecology Letters. 2009;12(4):334–50. pmid:19292794
    • 98. Heatwole H, Grech A, Monahan JF, King S, Marsh H. Thermal biology of sea snakes and sea kraits. Integrative and comparative biology. 2012;52(2):257–73. Epub 2012/06/07. pmid:22669175.
    • 99. Theresa D, Melanie AS, Nann AF, Wayne AB. Evidence for Control of Cutaneous Oxygen Uptake in the Yellow-Lipped Sea Krait Laticauda colubrina (Schneider, 1799). Journal of Herpetology. 2016;50(4):621–6.
    • 100. Graham JB, Rubinoff I, Hecht MK. Temperature Physiology of the Sea Snake Pelamis platurus: An Index of Its Colonization Potential in the Atlantic Ocean. Proceedings of the National Academy of Sciences. 1971;68(6):1360–3. pmid:16591934
    • 101. Brischoux F, Tingley R, Shine R, Lillywhite HB. Salinity influences the distribution of marine snakes: implications for evolutionary transitions to marine life. Ecography. 2012;35(11):994–1003.
    • 102. Rash R, Lillywhite HB. Drinking behaviors and water balance in marine vertebrates. Marine Biology. 2019;166(10):122.
    • 103. Dunson WA, Dunson MK. Convergent evolution of sublingual salt glands in the marine file snake and the true sea snakes. Journal of comparative physiology. 1973;86(3):193–208.
    • 104. Brischoux F, Rolland V, Bonnet X, Caillaud M, Shine R. Effects of Oceanic Salinity on Body Condition in Sea Snakes. Integrative and comparative biology. 2012;52(2):235–44. pmid:22710931
    • 105. Lillywhite Harvey B, Babonis Leslie S, Sheehy Coleman M, Tu MC. Sea Snakes (Laticauda spp.) Require Fresh Drinking Water: Implication for the Distribution and Persistence of Populations. Physiological and Biochemical Zoology. 2008;81(6):785–96. pmid:18821840
    • 106. Lillywhite HB, Sheehy CM, III, Sandfoss MR, Crowe-Riddell J, Grech A. Drinking by sea snakes from oceanic freshwater lenses at first rainfall ending seasonal drought. PLOS ONE. 2019;14(2):e0212099. pmid:30730972
    • 107. Udyawer V, Read M, Hamann M, Simpfendorfer CA, Heupel MR. Effects of environmental variables on the movement and space use of coastal sea snakes over multiple temporal scales. Journal of Experimental Marine Biology and Ecology. 2015;473:26–34. https://doi.org/10.1016/j.jembe.2015.07.011.
    • 108. Gherghel I, Brischoux F, Papeş M. Using biotic interactions in broad-scale estimates of species’ distributions. Journal of Biogeography. 2018;45(9):2216–25.
    • 109. Veron JEN. Corals of the World 2000.
    • 110. Veron JEN. A reef in time: the Great Barrier Reef from beginning to end: Harvard University Press; 2008.
    • 111. Voris HK, Voris HH. Feeding Strategies in Marine Snakes: An Analysis of Evolutionary, Morphological, Behavioral and Ecological Relationships1. American Zoologist. 1983;23(2):411–25.
    • 112. Sanders KL, Rasmussen AR, Mumpuni , Elmberg J, de Silva A, Guinea ML, et al. Recent rapid speciation and ecomorph divergence in Indo-Australian sea snakes. Molecular Ecology. 2013;22(10):2742–59. pmid:23506038
    • 113. Smith AB, Santos MJ. Testing the ability of species distribution models to infer variable importance. Ecography. 2020;43:1–13.
    • 114. Halvorsen R. A strict maximum likelihood explanation of MaxEnt, and some implications for distribution modelling. Sommerfeltia. 2013;36(1):1–132.
    • 115. Searcy CA, Shaffer HB. Do Ecological Niche Models Accurately Identify Climatic Determinants of Species Ranges? The American naturalist. 2016;187(4):423–35. Epub 2016/03/31. pmid:27028071.
    • 116. Lynch TP, Alford RA, Shine R. Sex-based divergence in tidal, lunar and seasonal cycles of activity in the olive sea snake, Aipysurus laevis (Elapidae, Hydrophiinae). Coral Reefs. 2023.
    • 117. Borczyk B, Paśko Ł, Kusznierz J, Bury S. Sexual dimorphism and skull size and shape in the highly specialized snake species, Aipysurus eydouxii (Elapidae: Hydrophiinae). PeerJ. 2021;9:e11311. Epub 2021/05/13. pmid:33976986; PubMed Central PMCID: PMC8063874.
    • 118. Shine R, Shine TG, Brown GP, Udyawer V. Sexual dimorphism in aipysurine sea snakes (Elapidae, Hydrophiinae). Royal Society Open Science. 2023;10(12):231261. pmid:38094274
    • 119. Lukoschek V, Heatwole H, Grech A, Burns G, Marsh H. Distribution of two species of sea snakes, Aipysurus laevis and Emydocephalus annulatus, in the southern Great Barrier Reef: metapopulation dynamics, marine protected areas and conservation. Coral Reefs. 2007;26(2):291–307.
    • 120. Goiran C, Shine R. Decline in sea snake abundance on a protected coral reef system in the New Caledonian Lagoon. Coral Reefs. 2013;32(1):281–4.
    • 121. Shine R, Brown GP, Goiran C. Population dynamics of the sea snake Emydocephalus annulatus (Elapidae, Hydrophiinae). Scientific Reports. 2021;11(1):20701. pmid:34667211
    • 122. Araújo MB, Ferri-Yáñez F, Bozinovic F, Marquet PA, Valladares F, Chown SL. Heat freezes niche evolution. Ecology Letters. 2013;16(9):1206–19. pmid:23869696
    • 123. Sunday JM, Bates AE, Dulvy NK. Global analysis of thermal tolerance and latitude in ectotherms. Proceedings of Royal Society B Biological Sciences. 2011;278(1713):1823–30. pmid:21106582
    • 124. Sunday JM, Bates AE, Dulvy NK. Thermal tolerance and the global redistribution of animals. Nature Climate Change. 2012;2(9):686–90.
    • 125. Grady JM, Maitner BS, Winter AS, Kaschner K, Tittensor DP, Record S, et al. Metabolic asymmetry and the global diversity of marine predators. Science. 2019;363(6425):eaat4220. pmid:30679341
    • 126. Nishizaki MT, Barron S, Carew E. Thermal stress increases fluctuating asymmetry in marine mussels: environmental variation and developmental instability. Ecosphere. 2015;6(5):art85. https://doi.org/10.1890/ES14-00399.1.
    • 127. Bennett JM, Sunday J, Calosi P, Villalobos F, Martínez B, Molina-Venegas R, et al. The evolution of critical thermal limits of life on Earth. Nature Communications. 2021;12(1):1198. pmid:33608528
    • 128. Gunderson AR, Stillman JH. Plasticity in thermal tolerance has limited potential to buffer ectotherms from global warming. Proceedings of the Royal Society B: Biological Sciences. 2015;282(1808):20150401. pmid:25994676
    • 129. Verberk WCEP, Calosi P, Spicer JI, Kehl S, Bilton DT. Does plasticity in thermal tolerance trade off with inherent tolerance? The influence of setal tracheal gills on thermal tolerance and its plasticity in a group of European diving beetles. Journal of Insect Physiology. 2018;106:163–71. pmid:29278714
    • 130. Udyawer V, Somaweera R, Nitschke C, d’Anastasi B, Sanders K, Webber BL, et al. Prioritising search effort to locate previously unknown populations of endangered marine reptiles. Global Ecology and Conservation. 2020;22:e01013. https://doi.org/10.1016/j.gecco.2020.e01013.
    • 131. Deutsch CA, Tewksbury JJ, Huey RB, Sheldon KS, Ghalambor CK, Haak DC, et al. Impacts of climate warming on terrestrial ectotherms across latitude. Proceedings of the National Academy of Sciences. 2008;105(18):6668. pmid:18458348
    • 132. Heatwole H, Taylor JA. Ecology of reptiles: Surrey Beatty & Sons; 1987.
    • 133. Lukoschek V, Beger M, Ceccarelli D, Richards Z, Pratchett M. Enigmatic declines of Australia’s sea snakes from a biodiversity hotspot. Biological Conservation. 2013;166:191–202. https://doi.org/10.1016/j.biocon.2013.07.004.