Temperature-driven density gradients of two congeneric felids reveal contrasting responses to climate change at a range margin
Data files
Nov 11, 2025 version files 5.08 MB
-
Bobcat_CH_Summer_2023_5_Day.csv
4.41 KB
-
Detectors_Condensed_5_Day_Occ_Dryad.csv
6.03 KB
-
GEDI_Canopy_Height.tif
4.77 MB
-
Lynx_CH_Summer_2023_5_Day.csv
2.74 KB
-
Mean_June_to_August_Max_Temp_SSP_126_2041_2070.tif
51.58 KB
-
Mean_June_to_August_Max_Temp_SSP_370_2041_2070.tif
51.58 KB
-
Mean_June_to_August_Max_Temp_SSP_585_2041_2070.tif
51.58 KB
-
Mean_Summer_June_to_August_Max_Temp.tif
51.58 KB
-
README.md
8.08 KB
-
SCR_Summer_2023_Lynx_Bobcats.R
41.24 KB
-
Snowshoe_Hare_Abundance_SWE_Canopy_Cover_Canopy_Height_Summer_2023.tif
46.90 KB
Abstract
Climate change causes divergent range shifts in cold versus warm-tolerant species, potentially reshuffling biotic interactions at range margins. Yet, outside of coarse distributional metrics, little information exists regarding the ecology of species along range peripheries. Here, we use camera traps and spatially-explicit capture-recapture (secr) modeling to examine how climatic gradients influence current and future patterns of density, abundance, and density overlap between two congeneric felids - cold-adapted Canada lynx (Lynx canadensis) and warm-adapted bobcats (Lynx rufus) - at a range margin in Washington, United States. Temperature drove density patterns along the range margin, with lynx densities declining and bobcat densities increasing as a function of temperature. Future abundances, obtained via projection of current-day models onto future climate scenarios, declined for lynx but were stable for bobcats, with both species experiencing upward elevational shifts. Areas of the landscape with high-lynx and low-bobcat densities declined in the future, but areas with low-lynx and high-bobcat densities increased, with only limited high-elevation refugia for lynx from expanding bobcat populations. Our approach reveals how temperature gradients shape density patterns of cold and warm-tolerant mammals and could be applied to other species and montane systems to better understand mammalian population trajectories and spatial associations at range edges.
Dataset DOI: 10.5061/dryad.g1jwstr3p
Description of the data and file structure
Temperature-Driven Density Gradients of Two Congeneric Felids Reveal Contrasting Responses to Climate Change at a Range Margin
This dataset contains .csv files, .tif files, and the associated R code that implements the .csv and .tif files to conduct spatially explicit capture recapture analyses for Canada lynx and bobcats.
Two of the .csv files (i.e., "Bobcat_CH_Summer_2023_5_Day" & "Lynx_CH_Summer_2023_5_Day") represent the capture histories of individual bobcats and lynx. These files include information regarding which camera trap station (i.e., "Detector") individuals were detected at, as well as when these individuals were detected (i.e., "Occasion").
The third .csv file (i.e., "Detectors_Condensed_5_Day_Occ_Dryad") contains information on the locations of camera trap stations (in UTM coordinates), as well as values for Effort (i.e., the number of days a camera trap station was active per 5-day occasion), Prey (i.e., the total number of snowshoe hare detections at a camera station divided by the total number of days that camera station was active during the duration of the study), and Cattle (i.e., the total number of cattle detections at a camera station divided by the total number of days that camera station was active during the duration of the study). Prey and Cattle were utilized as covariates for g0 (i.e., detection probability) during analyses. We note that the locations of camera trap stations provided here are rounded to the nearest 2,000 m in order to protect the precise locations of lynx, which are listed as Threatened under the U.S. endangered species act, and considered an Endangered species by the state of Washington. Consequently, attempts to reproduce analyses using the provided data may not produce the same results as presented in the manuscript. Please contact the corresponding author if any questions arise.
Three of the .tif files (i.e., "GEDI_Canopy_Height", "Snowshoe_Hare_Abundance_SWE_Canopy_Cover_Canopy_Height_Summer_2023", and "Mean_Summer_June_to_August_Max_Temp") represent rasters from which covariate values for density were extracted from. The remaining three .tif files (i.e., "Mean_June_to_August_Max_Temp_SSP_126_2041_2070", "Mean_June_to_August_Max_Temp_SSP_370_2041_2070", "Mean_June_to_August_Max_Temp_SSP_585_2041_2070") represent rasters from which future temperature values were extracted from in order to predict future density/abundance of lynx and bobcats.
Files and variables
File: Bobcat_CH_Summer_2023_5_Day.csv
Description: Capture history file for bobcats.
Variables
- Session: Only one session (i.e., Summer2023) used in this analysis.
- ID: "Name" of individual bobcat.
- Occasion: Cameras were active for 97 total days. We split detections of individuals to any one of twenty 5-day "occasions" to use in this analysis.
- Detector: Camera trap station ID. This number corresponds with the "Detector" variable in the file "Detectors_Condensed_5_Day_Occ_Dryad".
File: Lynx_CH_Summer_2023_5_Day.csv
Description: Capture history file for lynx.
Variables
- Session: Only one session (i.e., Summer2023) used in this analysis.
- ID: "Name" of individual lynx.
- Occasion: Cameras were active for 97 total days. We split detections of individuals to any one of twenty 5-day "occasions" to use in this analysis.
- Detector: Camera trap station ID. This number corresponds with the "Detector" variable in the file "Detectors_Condensed_5_Day_Occ_Dryad".
File: Detectors_Condensed_5_Day_Occ_Dryad.csv
Description: File containing information about camera trapping stations along with the covariate used to model effort and detection probability during SECR analyses. We note that the locations of camera trap stations provided here are rounded to the nearest 2,000 m in order to protect the precise locations of lynx, which are listed as Threatened under the U.S. endangered species act, and considered an Endangered species by the state of Washington. Consequently, attempts to reproduce analyses using the provided data may not produce the same results as presented in the manuscript. Please contact the corresponding author if any questions arise.
Variables
- Detector: Camera trap station ID.
- X: Location of camera trap station in UTM coordinates.
- Y: Location of camera trap station in UTM coordinates.
- Effort: The number of days a camera trap station was active per 5-day occasion.
- /: Formatting column needed in order to include detection covariates in SECR analyses.
- Prey: Detection covariate. The total number of snowshoe hare detections at a camera station divided by the total number of days that camera station was active during the duration of the study.
- Cattle: Detection covariate. The total number of cattle detections at a camera station divided by the total number of days that camera station was active during the duration of the study
File: Snowshoe_Hare_Abundance_SWE_Canopy_Cover_Canopy_Height_Summer_2023.tif
Description: Raster layer containing snowshoe hare abundance values based on a Royle/Nichols occupancy model incorporating camera trapping data collected from this study during the summer of 2023. For more details on this calculation see Appendix 2 in the manuscript. Values of snowshoe hare abundance from this layer were used as a covariate to predict current density of lynx and bobcats.
File: Mean_Summer_June_to_August_Max_Temp.tif
Description: Raster layer containing mean maximum temperatures between June-August 2016. This data was downloaded from CHELSA (https://doi.org/10.16904/envidat.159, https://doi.org/10.1038/s41597-020-00587-y), and used as a covariate to predict current density of lynx and bobcats.
File: GEDI_Canopy_Height.tif
Description: Raster layer containing canopy height values for 2019. This data was downloaded from GEDI (https://doi.org/10.1016/j.rse.2020.112165), and used as a covariate to predict current density of lynx and bobcats.
File: Mean_June_to_August_Max_Temp_SSP_126_2041_2070.tif
Description: Raster layer containing future (2041-2070) mean maximum temperatures between June-August under the SSP1-2.6 climate change scenario. This data was downloaded from CHELSA (https://doi.org/10.1038/sdata.2017.122), and used to predict future density of lynx and bobcats.
File: Mean_June_to_August_Max_Temp_SSP_370_2041_2070.tif
Description: Raster layer containing future (2041-2070) mean maximum temperatures between June-August under the SSP3-7.0 climate change scenario. This data was downloaded from CHELSA (https://doi.org/10.1038/sdata.2017.122), and used to predict future density of lynx and bobcats.
File: Mean_June_to_August_Max_Temp_SSP_585_2041_2070.tif
Description: Raster layer containing future (2041-2070) mean maximum temperatures between June-August under the SSP5-8.5 climate change scenario. This data was downloaded from CHELSA (https://doi.org/10.1038/sdata.2017.122), and used to predict future density of lynx and bobcats.
File: SCR_Summer_2023_Lynx_Bobcats.R
Description: R code incorporating all the other attached files (except the README) to run SECR models and predict current and future density, abundance, and patterns of spatial overlap between lynx and bobcats along a range edge environment.
Code/software
R studio was used to analyze all the provided data. All packages needed are provided at the start of the R script provided. The workflow essentially follows the script from top to bottom.
