Biogeography of herpetofaunas in the West Indies and alien invasion

Date

2017-05

Authors

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

I evaluated regional biogeographical patterns of West Indian herpetofauna and assessed: (1) the small island effect (SIE); (2) nestedness patterns; (3) multiple species–area relationship (SAR) models; (4) C- and Z-values, typically interpreted to represent insularity or dispersal ability; and (5) the average diversity of islands, among-island heterogeneity, γ-diversity, and the contribution of area effect towards explaining among-island heterogeneity by using additive diversity partitioning approach. Moreover, I used a mixed-effects modeling approach to determine whether human population, island size, congener richness, and establishment experience in the US were correlated with the establishment success of introduced herpetofaunas in the West Indies.

I found piecewise regression with three segments performed best, suggesting the SARs possess three different patterns that resulted from two area thresholds: a first one, delimiting the SIE, and a second one, delimiting evolutionary processes. Taxa with lower resource requirement, higher dispersal ability, and stronger adaptation to the environment generally displayed lower corresponding threshold values, indicating superior taxonomic groups could earlier end the SIE period and start in situ speciation as the increase of island size. Anti-nestedness occurred in the entire system, whereas high degree of nestedness could still occur in portions within the region. SARs were best modeled using the Cumulative Weibull and Lomolino relationships. The Cumulative Weibull and Lomolino regressions displayed both convex and sigmoid curves. The Cumulative Weibull regressions were more conservative than Lomolino at displaying sigmoid curves within the range of island size studied. The Z-value of all herpetofauna was overestimated by Darlington in 1957, and Z-values were ranked: (1) native > nonnative; (2) reptiles > amphibians; (3) snake > lizard > frog > turtle > crocodilian; (4) increased from lower to higher level taxonomic groups. Additive diversity partitioning showed that area had a weaker effect on explaining the among-island heterogeneity for nonnative species than for native species. I found human population has no obvious impact on the establishment success of amphibians but a negative effect on reptiles and all species; island size has no obvious impact for each group; congener richness likely contributes negatively for amphibians but positively for reptiles and all species; establishment experience in the US has no obvious impact for amphibians but a positive impact for reptiles and all species. I suggest that we found support for Darwin’s naturalization hypothesis for amphibians whereas pre-adaptation hypothesis for reptiles.

Although piecewise regressions with two segments have been widely used in SIE detection studies, they cannot clearly delimit three SAR patterns and may cause poor estimations for both slope and threshold value of the SIE. My findings suggest previous SIE detection studies conducted by the two-segment piecewise regression method should be reanalyzed. No matter the doubts about the existence of the SIE, the threshold value, where the slope changes, may be important for a successful application of island theory to conservation biogeography. Apart from area, it offers opportunity to assess variables such as habitat diversity, productivity, island age, energy and environmental heterogeneity that may predict species richness within the limits of the first threshold value. On the other hand, speciation may become the dominant process adding to the species richness of assemblages beyond the limits of the second threshold value, so the identification of such size threshold shines light on conservation biogeography over evolutionary time scales. Moreover, the comparison of threshold values will help evaluate resource requirement, dispersal ability, as well as environmental adaptation among taxa. And this in turn will help set up taxon-specific conservation planning.

A strong degree of nestedness implies that most species could be represented by conserving the largest (habitat or true) island. However, the low degree of nestedness shown in our result is consistent with the findings of our studies on the same set of 1668 islands, which indicate that species richness of the largest island fail to reach half the number of species pool. Contrary to the concept of protecting the largest reserve, I conclude that an array of reserves of different size and endemism could contribute to the maximal diversity in a region.

The Cumulative Weibull and Lomolino models are recorded to be sigmoid. However, in my study I found that they can display both convex and sigmoid curves, suggesting their flexibility has been underappreciated. Z-values may be overestimated if small islands are excluded from calculation. Species diversity is generated among islands, and within-island speciation rather than island area is the main source of new native species in this region.

The complex geological history of the West Indies offers many opportunities for dispersal and vicariance to affect biotas. Also, human activities are having a profound impact on local biotas, including the herpetofaunas. The comparisons of Z-values and the contribution of area effect towards explaining among-island heterogeneity between native and nonnative species reflect human activities accelerating the rate of over-water dispersal and weakening the area effect within the region. The contrast between small average diversity and low Z-values paradoxically reveals: (1) the reduction of area leads to a significant loss of species especially on small islands; (2) there are both extinctions and increases in among-island heterogeneity on small islands under natural and human-mediated conditions, respectively; and (3) many herpetofaunal species have strong over-water dispersal ability especially on islands of intermediate size. Human activities can temporarily enhance species richness on small islands, but also can disturb habitats and introduce predators and competitors, consequently increasing extirpations of populations and disrupting the complex but often fragile communities on large islands.

My results imply that areas to target for early detection of nonnative reptiles are those that host closely related species, and areas to target for early detection of nonnative amphibians are those that host distantly related species. Moreover, nonnative reptiles that have a successful establishment experience in the US are more likely to establish in novel environments. These findings may help wildlife conservationists to target high-risk areas and species and develop effective management strategies for amphibians and reptiles respectively.

Description

Keywords

Additive diversity partitioning, Bayesian analysis, Biological invasions, Congener richness, Establishment success, Generalized linear mixed models, Herpetofauna, Naturalization hypothesis, Nestedness, Non-linear regression, Small island effect, Species–area relationship, Threshold value, Pre-adaptation hypothesis, West Indies, Z-values

Citation