Assessing occupancy and activity of two invasive carnivores in two Caribbean islands: implications for insular ecosystems

Abstract

The introduction of exotic species is one of the major causes of the decline of global biodiversity. Tropical insular ecosystems, including many biodiversity hotspots, are particularly threatened by biological invasions. Two wild carnivores have been introduced in the Caribbean, the northern raccoon Procyon lotor and the small Indian mongoose Urva auropunctata. Understanding the spatial distribution and activity patterns of both species is crucial for conservation purposes. Here we used camera trap data to model single‐season occupancy and detection of these two species on two Caribbean islands, Guadeloupe and Martinique. Our survey highlighted the broad distribution of both species on these islands, with the exception of the northern raccoon population in Martinique which appears very limited. Moreover, spatio‐temporal co‐occurrence with other bird and mammal species revealed that the northern raccoon and the small Indian mongoose face few or no competitors. Finally, our models show that the occupancy of both species was not influenced by any variable tested (i.e. elevation, precipitation, temperature and land cover) and that the probability to detect small Indian mongooses was influenced by land cover and camera model. These results highlight the potential of both the northern raccoon and the small Indian mongoose to have a significant impact on the native ecosystems in these hotspots of biodiversity and demonstrate the necessity to develop conservation actions towards control and limitation of these invasive carnivores.

Spatial Occupancy, Local Abundance and Activity Rhythm of Three Ground Dwelling Columbid Species in the Forests of Guadeloupe in Relation to Environmental Factors

“…To this aim, 24 stations (each station involving 5 camera-traps operating over two separate 7-day periods) were surveyed between February and May 2019, including 6 dry forest stations, 6 flooded forest stations, and 12 tropical rainforest stations (Figure 1, Supplementary Table S2). Following Louppe et al [45], stations were not chosen using an aerial grid due to the very steep landscape and extensive urban areas characterizing Guadeloupe, but rather to best cover the islands high diversity of forest habitats and to obtain a representative sample of elevations [61,62]. At each station (with a distance between stations that ranged from 2.82 km to 62 km), we arranged five passive infrared camera-traps (Moultrie© M-40i, with a detection range of about 24 m over a 125° angle of view) in a straight line, with Following Louppe et al [45], stations were not chosen using an aerial grid due to the very steep landscape and extensive urban areas characterizing Guadeloupe, but rather to best cover the islands high diversity of forest habitats and to obtain a representative sample of elevations [61,62].…”

“…In Guadeloupe, the ZD and, to a lesser extent, the two Quail-Dove species are highly sought after by local hunters for personal food use [43], although the precise impact of hunting pressure on population dynamics remains undocumented. The three species are likely be vulnerable to predation by several invasive mammal species [44,45]. However, little is known about the overlap in the distribution and activity rhythms between the three columbid species and their potential predators.…”

Determination of freedom-from-rabies for small Indian mongoose populations in the United States Virgin Islands, 2019–2020

“…Mongoose predation of sea turtle nests on protected beaches on St. Croix was observed during our sampling, and capture rates were 85% the first day at one heavily predated field site (Prune Bay). In Guadeloupe and Martinique in the Caribbean region, statistical modelling of small Indian mongoose populations revealed mongooses were well established throughout various habitats on these islands and had few or no competition for resources [27]. Because mongooses are well established throughout USVI, eradication is unrealistic and has met with limited success in other island nations [28].…”

Temporal niche partitioning between carnivores differs depending on the analytical method: comparison between temporal overlap and co-occurrence, and time-to-encounter analyses

“…In this study, we proposed the temporal co-occurrence analysis (Fig. 1b), adapted from the spatial co-occurrence analysis 44,45 , which is a method used to assess the spatial cooccurrence of multiple species by using presence-absence data (e.g., [46][47][48][49] ). Furthermore, we also performed the time-to-encounter analysis (Fig.…”