Effects of urbanization and habitat composition on site occupancy of two snake species using regional monitoring data from southern California

The influence of species life history and distribution characteristics on species responses to habitat fragmentation in an urban landscape

Fragmentation within urbanized environments often leads to a loss of native species diversity; however, variation exists in responses among-species and among-populations within species. We aimed to identify patterns in species biogeography in an urbanized landscape to understand anthropogenic effects on vertebrate communities and identify species that are more sensitive or resilient to landscape change. We investigated patterns in species richness and species responses to fragmentation in southern Californian small vertebrate communities using multispecies occupancy models and determined factors associated with overall commonness and sensitivity to patch size for 45 small vertebrate species both among and within remaining non-developed patches. In general, smaller patches had fewer species, with amphibian species richness being particularly sensitive to patch size effects. Mammals were generally more common, occurring both in a greater proportion of patches and a higher proportion of the sites within occupied patches. Alternatively, amphibians were generally restricted to larger patches but were more ubiquitous within smaller patches when occupied. Species range size was positively correlated with how common a species was across and within patches, even when controlling for only patches that fell within a species' range. We found sensitivity to patch size was greater for more fecund species and depended on where the patch occurred within a species' range. While all taxa were more likely to occur in patches in the warmer portions of their ranges, amphibians and mammals were more sensitive to fragmentation in these warmer areas as compared to the rest of their ranges. Similarly, amphibians occurred at a smaller proportion of sites within patches in drier portions of their ranges. Mammals occurred at a higher proportion of sites that were also in drier portions of their range while reptiles did not differ in their sensitivity to patch size by range position. We demonstrate that taxonomy, life history, range size and range position can predict commonness and sensitivity of species across this highly fragmented yet biodiverse landscape. The impacts of fragmentation on species communities within an urban landscape depend on scale, with differences emerging among and within species and populations.

The behaviour of sea snakes (Emydocephalus annulatus) shifts with the tides

Tidal cycles are known to affect the ecology of many marine animals, but logistical obstacles have discouraged behavioural studies on sea snakes in the wild. Here, we analyse a large dataset (1,445 observations of 126 individuals) to explore tidally-driven shifts in the behaviour of free-ranging turtle-headed sea snakes (Emydocephalus annulatus, Hydrophiinae) in the Baie des Citrons, New Caledonia. Snakes tended to move into newly-inundated areas with the rising tide, and became more active (e.g. switched from inactivity to mate-searching and courting) as water levels rose. However, the relative use of alternative habitat types was largely unaffected by tidal phase.

Comparative survey techniques for a cryptic Australian snake (Hoplocephalus bitorquatus)

Ecologists endeavour to develop survey techniques that are cost-effective for the species they target and robust enough for statistical analysis. Using time as a measure of effort, we compared visual encounter surveys with artificial cover objects (strapped to trees), targeting an arboreal elapid, the pale-headed snake (Hoplocephalus bitorquatus) and its potential prey (geckos). Within a red gum forest vegetation community with relatively high snake density, capture rates were 0.6 ± 0.1 (s.e.) snakes/person-hour using visual encounter surveys, compared with only 0.1 ± 0.1 snakes/person-hour using cover objects. The probability of detection of pale-headed snakes was estimated from occupancy modelling at 0.70 ± 0.06 in visual encounter surveys and 0.19 ± 0.09 in cover object surveys. Gecko capture rates (among all vegetation communities) were significantly greater (P < 0.001) using cover objects. The probability of detection of geckos was estimated from occupancy modelling at 0.77 ± 0.05 in visual encounter surveys and 0.97 ± 0.02 in cover object surveys. Geckos favoured (P < 0.001) cover objects facing south during all seasons except winter. Artificial cover objects may provide some value in detecting pale-headed snakes in vegetation communities where habitat resources are limited; however, where resources are plentiful, visual encounter surveys are likely to remain the most cost-effective survey option.