Using occupancy models to assess the direct and indirect impacts of agricultural expansion on species’ populations

Abstract

Land-use change is a global threat to biodiversity, but how land-use change affects species beyond the direct effect of habitat loss remains poorly understood. We developed an approach to isolate and map the direct and indirect effects of agricultural expansion on species of conservation concern, using the threatened giant anteater (Myrmecophaga tridactyla) in the Gran Chaco as an example. We reconstructed anteater occupancy change between 1985 and 2015 by fitting single-season occupancy models with contemporary camera-trap data and backcasting the models to 1985 and 2000 land-cover/use maps. Based on this, we compared the area of forest loss (direct effect of agricultural expansion) with the area where forests remained but occupancy still declined (indirect effect of agricultural expansion). Anteater occupancy decreased substantially since 1985, particularly after 2000 when agriculture expanded rapidly. Between 1985 and 2015, ~ 64,000 km2 of forest disappeared, yet occupancy declined across a larger area (~ 102,000 km2), extending far into seemingly untransformed habitat. This suggests that widespread sink habitat has emerged due to agricultural land-use change, and that species may lose their habitat through direct and indirect effects of agricultural expansion, highlighting the urgent need for broad-scale conservation planning in the Chaco. Appropriate management responses could proactively protect more habitat where populations are stable, and restore habitat or address causes of mortality in areas where declines occur. Our work also highlights how occupancy modelling combined with remote sensing can help to detect the direct and indirect effects of agricultural expansion, providing guidance for spatially targeting conservation strategies to halt extinctions.

Landscape Structure and Seasonality: Effects on Wildlife Species Richness and Occupancy in a Fragmented Dry Forest in Coastal Ecuador

“…NDVI allows for greater spatial explicitness and numeric quantification of vegetation productivity and seasonality [55]. Occupancy analyses account for different detection probabilities [24,25,27,30] and indicate that DF exhibits higher species occupancy. Many factors could contribute to this finding.…”

“…Advantages include the following: incorporation of uncertainty about detectability to overcome sampling biases related to the detection differences among species, improved occupancy estimations of rare or infrequent species, accurate calculations of community features by including potential undetected species and the inclusion of habitat covariates, such as detailed spatial and temporal information derived from remote sensing occupancy models are also ideal for dealing with small sample sizes. [27][28][29][30]. Remote sensing has been an important tool for wildlife management, especially for the conservation of large animals over extensive areas [31][32][33].…”

Local forest proportion and proximity to large forest patches are important for native mammal conservation in Dry Chaco agroecosystems

“…Though the collared peccary is more tolerant to anthropogenic pressures compared to the other two peccary species, it still is highly dependent on woodlands to persist, and it is strongly affected by hunting (Camino et al., 2020; Periago et al., 2017; Semper‐Pascual et al., 2019; Weiler et al., 2020). The giant anteater, on the other hand, is rarely hunted intentionally, and in the Chaco, it still occurs in fragmented and degraded areas, but it is also increasingly affected by agroecosystem expansion (Di Blanco et al., 2019; Quiroga et al., ,2016a, 2016b; Semper‐Pascual et al., 2020). Due to these and other threats, these two species are categorized as vulnerable in the Argentine Red List of mammals (SAyDS‐SAREM, 2019), and the Giant anteater is also considered as vulnerable by the IUCN (Miranda, Bertassoni & Abba, 2014).…”

How do habitat amount and habitat fragmentation drive time-delayed responses of biodiversity to land-use change?

“…Estimating detection probability requires sites to be visited on multiple occasions within a period closed to changes in occupancy [44]. For mammals, we defined a sampling occasion as seven consecutive camera-days [50,51]. We discarded sites surveyed less than 14 camera-days to have a minimum of two occasions per site and assumed communities to be closed (no site-level species extinction or colonization) for a maximum of 12 sampling occasions (84 days).…”