Vagility: The Neglected Component in Historical Biogeography

The conceptual gap between ecological and historical biogeography is wide, although both disciplines are concerned with explaining how distributions have been shaped. A central aim of modern historical biogeography is to use a phylogenetic framework to reconstruct the geographic history of a group in terms of dispersals and vicariant events, and a number of analytical methods have been developed to do so. To date the most popular analytical methods in historical biogeography have been parsimony-based. Such methods can be classified into two groups based on the assumptions used. The first group assumes that vicariance between two areas creates common patterns of disjunct distributions across several taxa whereas dispersals and extinctions generate clade specific patterns. The second group of methods assumes that passive vicariance and within-area speciation have a higher probability of occurrence than active dispersal events and extinction. Typically, none of these methods takes into account the ecology of the taxa in question. I discuss why these methods can be potentially misleading if the ecology of the taxon is ignored. In particular, the vagility or dispersal ability of taxa plays a pivotal role in shaping the distributions and modes of speciation. I argue that the vagility of taxa should be explicitly incorporated in biogeographic analyses. Likelihood-based methods with models in which more realistic probabilities of dispersal and modes of speciation can be specified are arguably the way ahead. Although objective quantification will pose a challenge, the complete ignorance of this vital aspect, as has been done in many historical biogeographic analyses, can be dangerous. I use worked examples to show a simple way of utilizing such information, but better methods need to be developed to more effectively use ecological knowledge in historical biogeography.     

Differential response of mammals to agricultural fences — The effects of species vagility and body size

An increase in wolf populations during the 1990s in North-Eastern Israel compelled livestock growers to establish fenced calving enclosures to minimize cattle predation. We hypothesized that such fences form barriers to animals traversing the landscape, to which various wildlife species may respond differentially. In order to test this, an array of line transects were established near six enclosures at which scat pellets were monitored for 23 months. We identified 1496 pellets, belonging mainly to medium or large mammal species. To estimate pellet abundance and quantify mammal activity levels we used N-mixture models. We tested the performance of seven models assuming two abundance distribution functions: Poisson and zero-inflated Poisson (ZIP) distribution models which evaluated different combinations of explanatory variables: effects of time, enclosure and transect. Model selection was performed using the Akaike Information Criterion (AIC). Pellet counts and mean estimated abundance were greater in transects external to the enclosures, for all species combined, mountain gazelle (Gazella gazella) and wild boar (Sus scrofa). Additionally, two behavioral responses to the presence of protective fences were observed. Outside the enclosures a U-shaped response of decrease in abundance at a distance of 50 m from the fence followed by an increase towards the distant transects at 200–700 m from the fence. Inside the calving enclosures a decreasing abundance response was observed, which was negatively correlated with distance from the fence. This study identifies the spatial effects emerging from the presence of protective fences as habitat-fragmenting agents. It suggests that a protective fence imposes a habitat-independent behavioral filter at the landscape level which could be related to species vagility and body size. Inconsistent activity responses of the different species to the presence of fences are repeatedly observed, providing support for our initial hypothesis.     

Evaluation of fast-acquisition GPS in stationary tests and fine-scale tracking of green turtles

Fastloc GPS (FGPS) is a variant of Global Positioning System (GPS) technology that offers important new utility for investigating fine-scale movements of marine animals like green turtles that surface too briefly for effective use of standard GPS. I report here on the accuracy and efficiency of this novel technology, compare it with two alternative methods, namely boat-based ultrasonic tracking and Argos Platform Transmitter Terminals (PTTs), and provide new data on the vagility and habitat selection of green turtles in shallow coastal foraging habitat. I used a combined FGPS receiver and PTT transmitter (Sirtrack, Havelock North, New Zealand) mounted together with an ultrasonic transmitter and time-depth recorder in a tether-attached housing that allowed automatic detachment and subsequent retrieval of the equipment without the requirement to recapture turtles. With this equipment I conducted short deployments (4.5 to 16.8 d) on 3 free-living adult-size green turtles in coastal foraging habitat in Queensland, Australia. In addition, stationary tests in air and afloat were conducted at the same site. FGPS location error (mean ± SD) increased as the number of satellites used in each computation decreased, from 26 m ± 19.2 (8 satellites) to 172 m ± 317.5 (4 satellites). During live tracking the frequency of FGPS locations greatly exceeded Argos PTT, such that screened data comprised about 50 times more FGPS locations despite a much tighter screening threshold for FGPS (250 m) than for Argos PTT (1000 m). FGPS locations showed the three study turtles used modest short-term activity ranges with Minimum Convex Polygon area mean ± SD 662 ha ± 293.9. They all remained within < 4.7 km of their capture-release locations and favoured shallow water, with 86% of locations at charted depths ≤ 3 m and the deepest location at 5.9 m. Fine-scale movements of each turtle varied from day to day with respect to tortuosity and areas traversed. Statistically significant day-night differences were evident in average rates of movement (greater by day) and in habitat selection, where diurnal locations had greater seagrass density while nocturnal locations featured deeper bathymetry. Individual turtles revisited some of their centres of activity (identified from 50% fixed kernel utilisation distributions) on multiple occasions but none of the study turtles travelled consistently between the same day-night pair of sites as has been reported elsewhere. Such disparity and the day-to-day variation in movements revealed by these short-term findings highlight the need for detailed tracking over longer periods at multiple locations. Fastloc GPS technology proved an effective new tool for this area of research.