Co-occurence between snow leopard and Eurasian lynx in southern Mongolia
Data files
Dec 05, 2024 version files 72.88 MB
-
HexGrid_points.csv
791.57 KB
-
HexGrid_pointsExp.zip
621.35 KB
-
Lynx_SLData.csv
75.67 KB
-
Occ_Analysis.RData
71.37 MB
-
README.md
7.19 KB
-
Site_Cov_Raw.csv
17.93 KB
Abstract
Interspecific competition, a fundamental ecological process characterized by negative interactions between species, plays a vital role in shaping ecological communities. Despite the co-occurrence of the snow leopard (Panthera uncia) and the Eurasian lynx (Lynx lynx) across vast landscapes in Asia, their interactions remain poorly understood. In this study, we investigated how the presence of snow leopards affected site-use by lynx and whether presence of snow leopards resulted in behavioral adaptations by the lynx.
Between 2017 and 2022, we conducted camera trap-based surveys across six sites in Southern Mongolia and evaluated species co-occurrence by snow leopards and lynx using the occupancy framework. We assumed snow leopard to be the dominant species while using topographical and land cover variables as covariates. Our results show that the presence of snow leopards influenced site use by lynx, leading to a shift in space use when snow leopards were present. Specifically, lynx used the entire range of ruggedness and did not select for shrubby areas in the absence of snow leopards whereas they avoided rugged areas and had a strong preference for shrubby areas when snow leopards were present.
Our findings emphasize the influence a larger predator can have on the space use of a smaller predator, and how presence of snow leopard can result in altering space use of lynx. Understanding these interactions and behavioral adaptations can be useful for developing effective conservation strategies in the region.
README: Co-occurence between snow leopard and Eurasian lynx in southern Mongolia
https://doi.org/10.5061/dryad.pnvx0k6x0
Description of the data and file structure
The data was collected with camera traps placed in mountain ranges in southern Mongolia.
Files and variables
File: Lynx_SLData.csv
Description: The encounter history data for lynx and snow leopard collected from the study area denoting all camera traps were operational irrespective of whether they detected one of the two species or not.
Variables
- ID: Unique ID for each row.
- Area: The sub-area within the study area that the particular camera trap and particular site corresponds to.
- CT_ID: Camera trap ID denoting the unique id of a camera trap (note, each camera trap may have been deployed multiple times, so these IDs don't always represent unique locations
- Latitude: Latitude of the camera trap rounded off to two decimals
- Longitude: Longitude of the camera trap rounded off to two decimals
- Loc_type: Microhabitat representing where the camera trap was installed. Categorical data represented by saddle, canyon, cliff and valley.
- Date_installed: The date when a particular camera trap was installed
- Date_stopped: The date when a particular camera trap stopped working
- Enc_Date: The date when a snow leopard or lynx were detected on the particular camera trap. Note that in case the species of interest were detected multiple times on the same camera, a new row has been created to denote each unique encounter.
- Enc_Time: Time when the snow leopard or lynx was detected on the camera
- Lynx: Binary data denoting if lynx was detected (1) or not (0)
- Snow leopard: Binary data denoting if snow leopard was detected (1) or not (0)
- Unique_ID: Unique ID of the camera taking into consideration the Area, camera trap ID (CT_ID) and Year. This was important because the same camera traps were installed at different locations during the study period. Unique ID represent each of the unique 270 camera trap locations used for sampling during the study
File: Site_Cov_Raw_2.csv
Description: Site covariate data used to model the heterogeneity in the probability with which lynx and snow leopards used sites across the study area.
Variables
Probability with which lynx and snow leopards interacted and used the study area was modelled as a function of one or more spatial covariates. Each sampling unit was characterized by a circular polygon of radius 2.5km around each camera trap. The dataset provides spatial covariates for the 270 sampling units. The dataset has the following fields:
1. Area: The sub-area within the study area that the particular camera trap and particular site corresponds to.
2. CT_ID: ID of the camera traps used during the study period
3. Unique_ID: Unique ID of the camera taking into consideration the Area, camera trap ID (CT_ID) and Year. This was important because the same camera traps were installed at different locations during the study period. Unique ID represent each of the unique 270 camera trap locations used for sampling during the study
4. Forest: Binary data that denotes if the sampling site, i.e. circular area with 2.5km radius around the camera trap had shrub forest or not.
5. DEM: Mean altitude of the sampling site, i.e. circular area within 2.5km radius around the camera trap.
6. Rgd: Mean ruggedness of the sampling site, i.e. circular area within 2.5km radius around the camera trap.
7. NDVI: Mean value of the Normalized Differential Vegetation Index (NDVI) representing plant biomass at the sampling site, i.e. circular raea within 2.5km radius around the camera trap.
File: HexGrid_pointsExp.zip
Description: Hexgrid file denoting the entire study area as a mesh of hexgrids nearly the same size as the sampling unit (circles with radius 2.5km)
The file is used to map occupancy outputs in the analysis file to plot probability of sites being used by the two species across the entire study area.
File: Lynx_SLData.csv
Description: The encounter history data for lynx and snow leopard collected from the study area denoting all camera traps were operational irrespective of whether they detected one of the two species or not. The original data transcribed from the field survey and encounters obtained during the study period. The coordinates have been rounded off to two decimals to protect it being misused for the threatened species such as the snow leopard and lynx. Should detailed coordinates be required for research, the authors can be reached out and a data sharing agreement be signed before making the coordinates available.
Variables
- ID: Unique ID for each row.
- Area: The sub-area within the study area that the particular camera trap and particular site corresponds to.
- CT_ID: Camera trap ID denoting the unique id of a camera trap (note, each camera trap may have been deployed multiple times, so these IDs don't always represent unique locations
- Latitude: Latitude of the camera trap rounded off to two decimals
- Longitude: Longitude of the camera trap rounded off to two decimals
- Loc_type: Microhabitat representing where the camera trap was installed. Categorical data represented by saddle, canyon, cliff and valley.
- Date_installed: The date when a particular camera trap was installed
- Date_stopped: The date when a particular camera trap stopped working
- Enc_Date: The date when a snow leopard or lynx were detected on the particular camera trap. Note that in case the species of interest were detected multiple times on the same camera, a new row has been created to denote each unique encounter.
- Enc_Time: Time when the snow leopard or lynx was detected on the camera
- Lynx: Binary data denoting if lynx was detected (1) or not (0)
- Snow leopard: Binary data denoting if snow leopard was detected (1) or not (0)
- Unique_ID: Unique ID of the camera taking into consideration the Area, camera trap ID (CT_ID) and Year. This was important because the same camera traps were installed at different locations during the study period. Unique ID represent each of the unique 270 camera trap locations used for sampling during the study
File: HexGrid_points.csv
Description: Covariate data along with coordinates used to plot site-use probabilities in the R script for the entire study area. Not used in the paper.
Variables
- GRID_ID: Unique Grid ID
- dem_hex: Average value of elevation above msl for each hexagonal polygon
- forest_hex: Binary covariate denoting 1 if there is any shrubs within the hexagonal polygon
- rgd_hex: Average value for ruggedness for each hexagonal polygon
- ndvi_hex: Average value of the normalized differential vegetation index (NDVI) for each hexagonal polygon
File: Occ_Analysis.RData
Description: A binary file that stores the entire R workspace used and created while running the analysis file Lynx_SL2.R. The file can be used to reload R objects efficiently, preserving their structure, attributes, and metadata.
Code/software
The files can be viewed using R
Methods
The data was collected by depolying camera traps in several mountain ranges across Southern Mongolia. Data has been analysed in an occupancy framework