A penalized likelihood for multi‐species occupancy models improves predictions of species interactions

SummaryCamera traps and acoustic recording devices are essential tools to quantify the distribution, abundance and behavior of mobile species. Varying detection probabilities among device locations must be accounted for when analyzing such data, which is generally done using occupancy models. We introduce a Bayesian Time-dependent Observation Model for Camera Trap data (Tomcat), suited to estimate relative event densities in space and time. Tomcat allows to learn about the environmental requirements and daily activity patterns of species while accounting for imperfect detection. It further implements a sparse model that deals well will a large number of potentially highly correlated environmental variables. By integrating both spatial and temporal information, we extend the notation of overlap coefficient between species to time and space to study niche partitioning. We illustrate the power of Tomcat through an application to camera trap data of eight sympatrically occurring duiker Cephalophinae species in the savanna - rainforest ecotone in the Central African Republic and show that most species pairs show little overlap. Exceptions are those for which one species is very rare, likely as a result of direct competition.

Download Full-text

Predicting fish species richness and habitat relationships using Bayesian hierarchical multispecies occupancy models

Canadian Journal of Fisheries and Aquatic Sciences ◽

10.1139/cjfas-2019-0125 ◽

2020 ◽

Vol 77 (3) ◽

pp. 602-610

Author(s):

Shannon White ◽

Evan Faulk ◽

Caleb Tzilkowski ◽

Andrew Weber ◽

Matthew Marshall ◽

...

Keyword(s):

Species Richness ◽

Fish Community ◽

Fish Assemblages ◽

Similar Species ◽

Occupancy Models ◽

Imperfect Detection ◽

Modeling Framework ◽

Bayesian Hierarchical ◽

Recreation Area ◽

Species Occupancy

Understanding how stream fishes respond to changes in habitat availability is complicated by low occurrence rates of many species, which in turn reduces the ability to quantify species–habitat relationships and account for imperfect detection in estimates of species richness. Multispecies occupancy models have been used sparingly in the analysis of fisheries data, but address the aforementioned deficiencies by allowing information to be shared among ecologically similar species, thereby enabling species–habitat relationships to be estimated for entire fish communities, including rare species. Here, we highlight the utility of hierarchical multispecies occupancy models for the analysis of fish community data and demonstrate the modeling framework on a stream fish community dataset collected in the Delaware Water Gap National Recreation Area, USA. In particular, we demonstrate the ability of the modeling framework to make inferences at the species-, guild-, and community-levels, thereby making it a powerful tool for understanding and predicting how environmental variables influence species occupancy probabilities and structure fish assemblages.

Download Full-text

Model selection and assessment for multi-species occupancy models

Ecology ◽

10.1890/15-1471.1 ◽

2016 ◽

Vol 97 (7) ◽

pp. 1759-1770 ◽

Cited By ~ 54

Author(s):

Kristin M. Broms ◽

Mevin B. Hooten ◽

Ryan M. Fitzpatrick

Keyword(s):

Model Selection ◽

Occupancy Models ◽

Species Occupancy ◽

Selection And Assessment

Download Full-text

A dynamic occupancy model for interacting species with two spatial scales

10.1101/2020.12.16.423067 ◽

2020 ◽

Author(s):

Eivind Flittie Kleiven ◽

Frederic Barraquand ◽

Olivier Gimenez ◽

John-André Henden ◽

Rolf Anker Ims ◽

...

Keyword(s):

Spatial Structure ◽

Species Interactions ◽

Spatial Scales ◽

Species Interaction ◽

Camera Traps ◽

Occupancy Models ◽

Occupancy Model ◽

Interacting Species ◽

Multi Scale ◽

Extinction Probabilities

1AbstractOccupancy models have been developed independently to account for multiple spatial scales and species interactions in a dynamic setting. However, as interacting species (e.g., predators and prey) often operate at different spatial scales, including nested spatial structure might be especially relevant in models of interacting species. Here we bridge these two model frameworks by developing a multi-scale two-species occupancy model. The model is dynamic, i.e. it estimates initial occupancy, colonization and extinction probabilities - including probabilities conditional to the other species’ presence. With a simulation study, we demonstrate that the model is able to estimate parameters without bias under low, medium and high average occupancy probabilities, as well as low, medium and high detection probabilities. We further show the model’s ability to deal with sparse field data by applying it to a multi-scale camera trapping dataset on a mustelid-rodent predator-prey system. The field study illustrates that the model allows estimation of species interaction effects on colonization and extinction probabilities at two spatial scales. This creates opportunities to explicitly account for the spatial structure found in many spatially nested study designs, and to study interacting species that have contrasted movement ranges with camera traps.

Download Full-text

Designing an occupancy framework to monitor an endemic rainforest duck: Methodological and modelling considerations

10.1101/2020.02.03.931543 ◽

2020 ◽

Author(s):

Shah Nawaz Jelil ◽

Murchana Parasar ◽

Laura Cancino ◽

Kimberly Cook

Keyword(s):

Detection Probability ◽

Site Occupancy ◽

Optimal Number ◽

Occupancy Models ◽

Occupancy Detection ◽

Habitat Factors ◽

Long Term Ecological Research ◽

Species Occupancy ◽

Future Population

AbstractUnderstanding species trend, decline or growth, is vital to further conservation efforts. Species-habitat relationship studies are equally important for conservation as it helps in understanding the habitat a particular species depends upon, i.e. habitat conservation. However, rare and endemic species are inherently difficult to study and occupancy models are especially useful in such cases. We conducted the first detection, non-detection survey for the white winged duck in Dehing Patkai Wildlife Sanctuary, India to assess site occupancy and test habitat factors that explain its occupancy. We found that white winged duck occupancy was low (0.27 ± 0.21 SE) and detection probability was 0.44 ± 0.30 SE. We found that increasing tree richness and decreasing elevation increased species occupancy. Detection probability was influenced by our effort in that detection increased with increasing number of survey hours. Using two standard approaches, we estimated the optimal number of sites and replicate surveys for future occupancy studies. We further present considerations for future surveys. Considering the sporadic and fragmented information available, we recommend long-term ecological research to better understand the present and future population trends of the species.

Download Full-text

Dynamic multi-species occupancy models of birds of high-altitude grasslands in eastern South Africa

10.7287/peerj.preprints.26932 ◽

2018 ◽

Author(s):

David H Maphisa ◽

Hanneline Smit_Robinson ◽

Res Altwegg

Keyword(s):

High Altitude ◽

Habitat Preferences ◽

Bird Species ◽

Bird Diversity ◽

Grass Cover ◽

Occupancy Models ◽

Occupancy Dynamics ◽

Pumped Storage ◽

Species Occupancy ◽

High Altitude Grasslands

Moist, high-altitude grasslands of eastern South African harbour rich avian diversity and endemism. This area is also threatened by increasingly intensive agriculture and land conversion for energy production. This conflict is particularly evident at Ingula, an Important Bird and Biodiversity Area located within the least conserved high-altitude grasslands and which is also the site of a new Pumped Storage Scheme. The new management seeks to maximise biodiversity through manipulation of the key habitat variables: grass height and grass cover through burning and grazing to make habitat suitable for birds. However, different species have individual habitat preferences, which further vary through the season. We used a dynamic multi-species occupancy model to examine the seasonal occupancy dynamics of 12 common grassland bird species and their habitat preferences. We estimated monthly occupancy, colonisation and persistence in relation to grass height and grass cover throughout the summer breeding season of 2011/12. For majority of these species, at the beginning of the season occupancy increased with increasing grass height and decreased with increasing grass cover. Persistence and colonisation decreased with increasing grass height and cover. However, the 12 species varied considerably in their responses to grass height and cover. Our results suggest that management should aim to provide plots which vary in grass height and cover to maximise bird diversity. We also conclude that the decreasing occupancy with increasing grass cover and low colonisation with increasing grass height and cover is a results of little grazing on our study site. We further conclude some of the 12 selected species are good indicators of habitat suitability more generally because they represent a range of habitat needs and are relatively easy to monitor.

Download Full-text