Intraguild interactions and abiotic conditions mediate occupancy of mammalian carnivores: co-occurrence of coyotes-fishers-martens
Cite this dataset
Twining, Joshua; Jensen, Paul; Fuller, Angela; Brazeal, Jen (2024). Intraguild interactions and abiotic conditions mediate occupancy of mammalian carnivores: co-occurrence of coyotes-fishers-martens [Dataset]. Dryad. https://doi.org/10.5061/dryad.pc866t1rz
Abstract
The widespread eradication of large carnivores and subsequent expansion of top mesopredators have the potential to impact species and community interactions with ecosystem-wide implications. An example of these trophic dynamics is the widespread establishment of coyotes following the extirpation of wolves and mountain lions in eastern North America. Here, we examined the occupancy of three carnivores in northern New York considering both environmental/habitat factors and interspecific interactions. We estimated the co-occurrence of coyotes, fishers, and martens from a landscape-scale winter camera trap survey repeatedly annually for three years. Martens occurred independently of both coyotes and fishers, while fishers and coyotes displayed positive intraguild interactions that were constant across the landscape. Both marten and fisher first-order occupancy was driven by a combination of biotic and abiotic factors, with both species displaying positive associations with forest cover but antithetical responses to average snow depth. The integral and antithetical role of snow depth in driving the occurrence of martens (positive) and fishers (negative) in the landscape indicates that future climatic warming could reduce the availability of current spatial refuges for martens created by severe winter conditions. Climate-driven alterations to established competitive interactions and co-existence patterns between marten and fishers have critical implications for the species' survival and conservation. We provide correlational evidence consistent with the potential for positive top-down effects of dominant mesocarnivores on subordinate species, with fisher occupancy increasing conditional on the presence of coyotes across the landscape. These findings align with the hypothesis that under certain conditions, coyotes may facilitate certain subordinate carnivores. The evidence produced here is consistent with hypotheses on the dynamic nature of trophic niches. We demonstrate the need to consider the interplay between climate, habitat, and interspecific interactions to understand wildlife occupancy patterns and inform wildlife management in a rapidly changing world.
README: Intraguild interactions and abiotic conditions mediate occupancy of mammalian carnivores: co-occurrence of coyotes-fishers-martens
https://doi.org/10.5061/dryad.pc866t1rz
The data stems from a large-scale camera trapping project conducted for 3-weeks annually at approximately 195 sites over 3 years in the Adirondacks and Tug Hill region of New York State, USA. The target species were fisher, marten, and coyote. Camera traps were secured to trees approximately 1.0-1.5 m above ground. A bait station was placed on a tree opposite the camera trap and secured to the tree using wire mesh. The bait used varied and included beaver, moose, white-tailed deer, muskrat, raccoon, cow, and chicken.
Description of the data and file structure
The data folder
This folder has ten files.
1. 7dayoccDetectionscoyoteNZ.csv
This file contains detection/non-detection data for coyotes between the years of 2016-2018 in New York state. 0 is not detected, and 1 is detected. Each row is a site, each column is a weekly occasion, and each cell is a detection/non-detection record.
2. 7dayoccDetectionsfisherNZ
This file contains detection/non-detection data for fishers between the years of 2016-2018 in New York state. 0 is not detected, and 1 is detected. Each row is a site, each column is a weekly occasion, and each cell is a detection/non-detection record.
3. 7dayoccDetectionsmartenNZ
This file contains detection/non-detection data for American marten between the years of 2016-2018 in New York state. 0 is not detected, and 1 is detected. Each row is a site, each column is a weekly occasion, and each cell is a detection/non-detection record.
4. NZ.3yrcamsite.6km.buffer.alllandscapecovariates.csv
This file contains all of the summarized spatial covariate data at a 6km scale used in the analysis. Each row is a different 6km 2 pixel in New York State, each column is a covariate, and each cell is a value.
The acronyms used in this dataset are WSI = Winter Severity Index, ADK = Adirondacks, TH = Tug Hill, FID = Field ID, and NZ = Northern Zone).
The landscape covariates used in the analysis are:
| Column name / covariate | Description | Source |
|---|---|---|
| Deciduous | Proportion of a 6 km2 buffer around the detector made up of deciduous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Coniferous | Proportion of a 6 km2 buffer around the detector made up of coniferous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Mixed | Proportion of a 6 km2 buffer around the detector made up of mixed forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Pasture | Proportion of a 6 km2 buffer around the detector made up of pasture | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Cultivated.Crops | Proportion of a 6 km2 buffer around the detector made up of cultivated crops | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| RoadDen | Mean number of km of road per km2 in each 6 km2 buffer | calculated from primary and secondary roads raster provided by the NYSDEC, hosted on GitHub |
| snowdepth | Mean daily snow depth(m) of the 6 km2 buffer around each detector across the sampling period | National Operational Hydrologic Remote Sensing Centre, 2004. Snow data assimilation system (SNODAS) products (https://doi.org/10.7265/N5TB14TC). |
| forest.edge | Edge density of combined class of all forest in each 6 km2 buffer around detectors | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| min_temp | The mean daily minimum temperature over the sampling period at each detector (degree C). | PRISM Climate data, Oregon State University (http://prism.oregonstate.edu). |
| GPP | The amount of carbon captured by plants (kg C MJ-1) in at the detectors | MODIS Land Satellite, 2017 (https://lpdaac.usgs.gov/products/mod17a2hv006/) |
5. NZ.3yrcamsite.20km.buffer.alllandscapecovariates.csv
This file contains all of the summarized spatial covariate data at a 20km scale used in the analysis. Each row is a different 20km 2 pixel in New York State, each column is a covariate, and each cell is a value.
The acronyms used in this dataset are WSI = Winter Severity Index, ADK = Adirondacks, TH = Tug Hill, FID = Field ID, and NZ = Northern Zone).
The landscape covariates used in the analysis are:
| Column name / covariate | Description | Source |
|---|---|---|
| Deciduous | Proportion of a 20 km2 buffer around the detector made up of deciduous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Coniferous | Proportion of a 20 km2 buffer around the detector made up of coniferous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Mixed | Proportion of a 20 km2 buffer around the detector made up of mixed forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Pasture | Proportion of a 20 km2 buffer around the detector made up of pasture | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Cultivated.Crops | Proportion of a 20 km2 buffer around the detector made up of cultivated crops | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| RoadDen | Mean number of km of road per km2 in each 20 km2 buffer | calculated from primary and secondary roads raster provided by the NYSDEC, hosted on GitHub |
| snowdepth | Mean daily snow depth(m) of the 20 km2 buffer around each detector across the sampling period | National Operational Hydrologic Remote Sensing Centre, 2004. Snow data assimilation system (SNODAS) products (https://doi.org/10.7265/N5TB14TC). |
| forest.edge | Edge density of combined class of all forest in each 20 km2 buffer around detectors | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| min_temp | The mean daily minimum temperature over the sampling period at each detector (degree C). | PRISM Climate data, Oregon State University (http://prism.oregonstate.edu). |
| GPP | The amount of carbon captured by plants (kg C MJ-1) in at the detectors | MODIS Land Satellite, 2017 (https://lpdaac.usgs.gov/products/mod17a2hv006/) |
6. NZ.3yrcamsite.15km.buffer.alllandscapecovariates.csv
This file contains all of the summarized spatial covariate data at a 15km scale used in the analysis. Each row is a different 15km 2 pixel in New York State, each column is a covariate, and each cell is a value.
The acronyms used in this dataset are WSI = Winter Severity Index, ADK = Adirondacks, TH = Tug Hill, FID = Field ID, and NZ = Northern Zone).
The landscape covariates used in the analysis are:
| Column name / covariate | Description | Source |
|---|---|---|
| Deciduous | Proportion of a 15 km2 buffer around the detector made up of deciduous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Coniferous | Proportion of a 15 km2 buffer around the detector made up of coniferous forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Mixed | Proportion of a 15 km2 buffer around the detector made up of mixed forest | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Pasture | Proportion of a 15 km2 buffer around the detector made up of pasture | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| Cultivated.Crops | Proportion of a 15 km2 buffer around the detector made up of cultivated crops | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| RoadDen | Mean number of km of road per km2 in each 15 km2 buffer | calculated from primary and secondary roads raster provided by the NYSDEC, hosted on the github |
| snowdepth | Mean daily snow depth(m) of the 15 km2 buffer around each detector across the sampling period | National Operational Hydrologic Remote Sensing Centre, 2004. Snow data assimilation system (SNODAS) products (https://doi.org/10.7265/N5TB14TC). |
| forest.edge | Edge density of combined class of all forest in each 15 km2 buffer around detectors | NLCD, 2019 (https://www.mrlc.gov/data/nlcd-2019-land-cover-conus) |
| min_temp | The mean daily minimum temperature over the sampling period at each detector (degree C). | PRISM Climate data, Oregon State University (http://prism.oregonstate.edu). |
| GPP | The amount of carbon captured by plants (kg C MJ-1) in at the detectors | MODIS Land Satellite, 2017 (https://lpdaac.usgs.gov/products/mod17a2hv006/) |
7. NZ7daycameras.csv
This file contains observation-level covariate data detailing the camera used on each date. Each row is a different site in New York State, each column is an occasion, and each cell is a value.
The acronyms used in this dataset are ADK = Adirondacks, TH = Tug Hill, and NZ = Northern Zone).
8. NZ7dayoccbait.csv
This file contains observation-level covariate data detailing the bait used on each date. Each row is a different site in New York State, each column is an occasion, and each cell is a value.
The acronyms used in this dataset are ADK = Adirondacks, TH = Tug Hill, and NZ = Northern Zone).
9. NZAvgTempByOcc.csv
This file contains observation-level covariate data detailing the mean temperature on each occasion at each site. Each row is a different site in New York State, each column is an occasion, and each cell is a value. The temperatures are in degrees C.
The acronyms used in this dataset are ADK = Adirondacks, TH = Tug Hill, Occ = Occasion, and NZ = Northern Zone).
10. NZAvgTempByOcc.csv
This file contains observation-level covariate data detailing the ordinal date on the first day of each occasion at each site. Each row is a different site in New York State, each column is an occasion, and each cell is a value.
The acronyms used in this dataset are ADK = Adirondacks, TH = Tug Hill, Occ = Occasion, and NZ = Northern Zone).
Description of software and packages required
All analysis was conducted in R, and code to fully reproduce the analysis is provided. The packages required to run this analysis in R include 'unmarked', 'aiCcmodavg', and 'MuMIn'.
Methods
We conducted camera trap surveys for carnivores from January-March 2016-2018 throughout the Adirondack and Tug Hill regions of northern New York State (Figure S1). The study area is an approximately 32,000 km2 region dominated by contiguous mature stands of deciduous, mixed, and coniferous forests along an elevational gradient ranging from 24 – 1448m. We used a stratified random sampling design to select 15-km2 sample units and a standardized methodology across all surveys which included the same 195 sample units in each of the three years (except for 13 sites that were not sampled in 2017). At each site, a camera trap was deployed randomly within the 15 km2 grid. Camera traps were secured to trees approximately 1.0-1.5 m above ground. A bait station was placed on a tree opposite the camera trap and secured to the tree using wire mesh. The bait used varied and included beaver (Castor canadensis), moose (Alces alces), white-tailed deer (Odocoileus virginianus), muskrat (Ondatra zibethicus), raccoon (Procyon lotor), beef (Bos taurus), and chicken (Gallus domesticus). At all sites, skunk-based call lures were applied. Cameras were deployed for 3 weeks (21 days) at each location after which cameras were retrieved. Cameras and bait were checked halfway through 3 weeks of sampling (generally day 10 or 11) with batteries and bait replaced and replenished as necessary. Detection records were created for each of the species sampled over the recording period. We used a weekly occasion length, with only one detection of each species being possible per weekly period to ensure independence of detections.
Funding
New York State Department of Environmental Conservation, Award: W-173-G