Data from: Experimental and field evidence indicate that islet-nesting tundra birds experience reduced nest predation and benefit indirectly from high snow goose densities
Data files
Jun 18, 2026 version files 41.02 KB
-
cacg_nests_2018-2023.csv
8.76 KB
-
cleaned_triads_2022_2023_2024.csv
21.73 KB
-
glgu_nests_2018-2023.csv
4.18 KB
-
README.md
6.35 KB
Abstract
Landscape features can shape the occurrence and strength of predator-prey interactions by influencing predation risk and prey distribution. In the High Arctic, some bird species select nesting sites with physical features that impede access for their main terrestrial predator, the Arctic fox. We investigated how nest microhabitat characteristics and prey availability modulate nest survival in tundra birds that select pond and lake islets as breeding sites. Over four summers, we analyzed survival of 132 cackling goose and 55 glaucous gull nests located on islets or pond and lake shores within a 150 km2 area encompassing a snow goose colony on Bylot Island, Nunavut, Canada. Over three years, we also assessed the effect of microhabitat on predation risk using artificial nests. We found that islets act as partial prey refuges, with higher nest survival on islets than on pond and lake shores. Nest survival generally increased with islet distance from shore, but we found little evidence of this effect for cackling geese and glaucous gulls, which rarely nested on islets near shore. Moreover, water depth surrounding islets had little to no influence on any nest type. Nest survival on islets was much lower in a year with relatively low snow goose nest density, suggesting that geese may confer a short-term positive indirect effect on other bird species nesting on islets within or near the main goose nesting area. Since the Arctic fox was virtually the sole predator of artificial nests, our findings suggest that annual variation in nest survival on islets was driven by a shift in fox foraging behavior in response to changes in prey availability across the landscape. Our study, which integrates multi-year monitoring and field experiments, highlights the interplay between microhabitat selection and predator-multi-prey dynamics in the Arctic tundra.
Dataset DOI: 10.5061/dryad.9w0vt4bx5
Description of the data and file structure
Files and variables
File: cacg_nests_2018-2023.csv
Description: The cacg_nests_2018-2023.csv dataset contains processed data on cackling goose nest survival on islets and on pond and lake shores during the 2018, 2019, 2022, and 2023 reproductive seasons on Bylot Island, Nunavut, Canada. Each row represents a monitored nest. These processed values (as described in the Methods section and Supporting information of the article) are used to analyze cackling goose nest survival in relation to microhabitat and main prey densities, as well as in relation to islet physical characteristics and main prey densities.
Variables
-
Nest_Unique_ID: Unique identifier for each nest monitored
-
Visit_Year: Year of observation
-
Species: Species to which the nest belongs
-
Nest_Site_Location: Location of the nest microhabitat (categorical)
Islet = nest located on an islet
Shore = nest located on the shoreline
-
Fate: Nest survival success (binary variable)
1 = at least one egg hatched or was still being incubated by a parent at the end of the monitoring
0 = complete failure (all eggs depredated)
-
exposure: For successful nests, exposure corresponds to the number of days from discovery to either the hatch date (observed or estimated) or to the last monitoring date if eggs were still incubating. For depredated nests, exposure corresponds to the number of days from discovery to the midpoint between the last active date and the date it was determined to be depredated.
-
Cluster: Nesting zone unique identification number
-
Island_ID: Islet unique identifier number
-
Dist_m: Islet distance to the nearest shore, in meters
-
Depth_cm: Deepest water depth measured along the shortest distance to the islet, in centimeters
-
D_Goose: Estimated annual snow goose nest density in the study area. For details of the method, see Moisan et al. 2025.
-
D_lem_june_july: Estimated lemming density (individuals per km2). We estimated lemming density with capture-recapture. For details of the method, see Fauteux et al. 2015. We estimated annual lemming density as the mean of June and July densities for two lemming species, a period which corresponds to the incubation period for both cackling goose and glaucous gull
File: glgu_nests_2018-2023.csv
Description: The glgu_nests_2018-2023.csv dataset contains processed data on glaucous gull nest survival on islets and on pond and lake shores during the 2018, 2019, 2022, and 2023 reproductive seasons on Bylot Island, Nunavut, Canada. Each row represents a monitored nest. These processed values (as described in the Methods section and Supporting information of the article) are used to analyze glaucous gull nest survival in relation to islet physical characteristics and main prey densities.
Variables
-
Nest_Unique_ID: Unique identifier for each nest monitored
-
Visit_Year: Year of observation
-
Species: Species to which the nest belongs
-
Nest_Site_Location: Location of the nest microhabitat (categorical)
Islet = nest located on an islet
Shore = nest located on the shoreline
-
Island_ID: Islet unique identifier number
-
Dist_m: Islet distance to the nearest shore, in meters
-
Depth_cm: Deepest water depth measured along the shortest distance to the islet, in centimeters
-
Fate: Nest survival success (binary variable)
1 = at least one egg hatched or was still being incubated by a parent at the end of the monitoring
0 = complete failure (all eggs depredated)
-
exposure: For successful nests, exposure corresponds to the number of days from discovery to either the hatch date (observed or estimated) or to the last monitoring date if eggs were still incubating. For depredated nests, exposure corresponds to the number of days from discovery to the midpoint between the last active date and the date it was determined to be depredated.
-
D_Goose: Estimated annual snow goose nest density in the study area. For details of the method, see Moisan et al. 2025.
-
D_lem_june_july: Estimated lemming density (individuals per km2). We estimated lemming density with capture-recapture. For details of the method, see Fauteux et al. 2015. We estimated annual lemming density as the mean of June and July densities for two lemming species, a period which corresponds to the incubation period for both cackling goose and glaucous gull
File: cleaned_triads_2022_2023_2024.csv
Description: The cleaned_triads_2022_2023_2024.csv dataset contains processed data on artificial nest predation from the 2022, 2023, and 2024 bird reproductive seasons on Bylot Island, Nunavut, Canada. Each row represents a monitored nest, grouped by experimental unit (Unique_ID), which corresponds to a triad of artificial nests (one on an islet and two on the pond/lake shore). These processed values (as described in the Methods section) are used to analyze artificial nest survival in relation to microhabitat and year.
Variables
-
Unique_ID: Unique identifier for each experimental unit (i.e., triad of artificial nests)
-
Year: Year this experiment was conducted
-
Microhabitat: Location of the nest microhabitat (categorical)
islet = nest located on an islet
shore = nest located on the shoreline
-
Islet_nb: Islet unique identifier number
-
Dist_m: Islet distance to the nearest shore, in meters
-
Depth_cm: Deepest water depth measured along the shortest distance to the islet, in centimeters
-
Exposure_h: Number of hours (converted from days) the nest was deployed before at least on nest from the experimental unit was predated
-
Predation: Nest predation (binary variable)
YES = at least one egg was predated, presumably by an Arctic fox
NO = all eggs were intact
Cited references
Fauteux, D. et al. 2015. Seasonal demography of a cyclic lemming population in the Canadian Arctic. - J. Anim. Ecol. 84: 1412–1422.
Moisan, L. et al. 2025. Long-Term Abundance Time-Series of the HighArctic Terrestrial Vertebrate Community of Bylot Island, Nunavut. Ecology 106 (10): e70223
These datasets contain the data underlying the analyses presented in the article "Experimental and field evidence indicate that islet-nesting tundra birds experience reduced nest predation and benefit indirectly from high snow goose densities". The variables included are the processed values described in the Methods section and Supporting Information and were used to model the influence of nest microhabitat characteristics and prey availability on natural and artificial nest survival.