We assessed reproductive success for the Adelie penguin colonies breeding on the Windmill Islands (Whitney Point, Odbert Island, Shirley Island, Blakeney Point 66°15' S, 110°33' E) from the 2011/12 – 2020/21 breeding seasons. We modelled nest survival against a range of potential environmental and behavioural drivers assessed over the same time period which are described below.
Sites:
We used four different sites with five camera locations for this study: Odbert Island = Odbt1, Shirley Island = Shrl1, Blakeney Point = Blak1, Whitney Point camera site 1 = Whit1, Whitney Point camera site 2 = Whit2.
Individual nest identifier:
For each nest, we developed a unique nest identifier (Year_Site_Nest). This includes the breeding season and camera site the nest was observed in, and also the number of the nest (1-10) observed.
Reproductive success data:
We used two metrics of reproductive success, breeding success and nest survival to creche. Breeding success was recorded as 0, 1, or 2 for each nest (0 = no chicks, 1 = one chick, 2 = two chicks). Nest survival was measured as binomial data and collected as 0,1 where 0 = nest failure and 1 = at least one chick survived to creche (Success.Fail). Nest failure occurred when either the parents abandoned the nest, both eggs failed to hatch, or chicks died. Failure and creche dates were recorded and the number of days since October 1 were calculated (Nest_Fail_date and Creche_Date). The breeding phase when failure occurred was noted for failed nests (Fail_Period). See BehaviouralData.csv.
Environmental covariates:
Southern Oscillation Index (SOI) and Southern Annular Mode (SAM)
SOI and SAM values were averaged for the October – January (SOI_OctJan and SAM_OctJan) and the April to October (SOI_JulyJune, SAM_JulyJune) periods to correlate with the Adelie penguin reproductive cycle and the winter season respectively. The NOAA Climate Prediction Centre https://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/aao/monthly.aao.index.b79.current.ascii.table provided the SAM values, while the Australian Bureau of Meteorology http://www.bom.gov.au/climate/enso/soi/ provided the SOI data. SOI values range from ~-35 - +35, and the value is quoted as a whole number, and we used monthly values to obtain our reproductive cycle and winter value averages. SAM values are collected on an index where positive and negative values relate to different polar pressures, and similarly to SOI, values we used for reproductive cycle and winter season averages were collected monthly. See file BehaviouralData.csv.
Fast ice extent (Fast_Ice)
Shapefiles of fast-ice distribution in the Windmill Islands were obtained from the Natice website www.natice.noaa.gov. We measured the shortest distance between the four study sites and the fast-ice edge in ArcGIS during mid-November for each breeding season during the study (2011/12 – 2020/21). See file BehaviouralData.csv.
Potential foraging overlap (Foraging_Intensity)
Here, values were based on those presented in Southwell et al. (2021). This data is estimated using the ‘foraging radius approach’ developed by Critchley et al. (2018) and Handley et al. (2021) and recent population estimates for the region’s breeding sites, the distance between penguin breeding sites, and foraging distances from colonies to estimate the degree of potential overlaps between local breeding penguins when foraging at sea. See file BehaviouralData.csv.
Windchill (WC)
Mean windchill (°C) was calculated for incubation, guard, and creche for each breeding season. These are calculated from ambient air temperature, wind speed, and relative humidity recorded at Casey Station using the formula provided by the Australian Bureau of Meteorology http://www.bom.gov.au/info/ thermal_stress/. We calculated values for incubation, guard, and creche during each breeding season, and also for the period of time prior to incubation when penguins are arriving at the nest (arrival windchill: AWC; incubation: IWC; guard windchill: GWC; creche windchill: CWC). See file BehaviouralData.csv.
Wind speed (MWS, km/h), Air Temperature (MAT, °C), Sum of snow days (SMS)
Mean wind speed, mean air temperature, and sum of mean snow days were all collected through the Bureau of Meteorology at Casey Station. Means were calculated for incubation (IMWS: Incubation mean wind speed; IMAT: Incubation mean air temperature; ISMS: Incubation sum of mean snow days; GMWS: Guard mean wind speed; GMAT: Guard mean air temperature; GSMS: Guard sum of mean snow days; CMWS: Creche mean wind speed; CMAT: Creche mean air temperature; CSMS: Creche sum of mean snow days; AMWS: Arrival mean wind speed; AMAT: Arrival mean air temperature; ASMS: Arrival sum of mean snow days). See file BehaviouralData.csv.
Snow cover and ground moisture
These covariates were recorded daily throughout each breeding season in categories of 0 -3. 0 is no snow cover or dry ground, 1 is light snow cover or low moisture, 2 is a moderate snow cover or moderate moisture, and 3 is a heavy snow cover or high moisture. These values were obtained from assessing the immediate ground surrounding the nest and also within the nest (~30 cm range outside of the nest). See EnvironmentalData.csv. Using these values, the individual nest’s propensity to have snow cover or be wet was calculated by determining the proportion of days during the observation period when each nest had snow or was moist (proportion of snow days = Prop.snowdays; Proportion of days with moisture = Prop.moist.days). Alongside this, we also calculated the number of days of snow cover or ground moisture for each category (1, 2, or 3) during incubation, guard, and creche (Number of days with snow cover = Num.snowdays; Number of days with ground moisture = Num.Moist.days). Here, we also took note of the number of nest observations per nest (Num.nest.obs). See Moisture_Snow_propensity_Data.csv.
Behavioural covariates
Nest occupation date
NOD was recorded for individual nests for each breeding season. It was recorded as the first day of a continuous period of five days with a penguin present at the nest and the number of days since October 1 and this date was calculated (NOD). See file BehaviouralData.csv.
First foraging trip (FFT)
FFT was recorded as the date when the female left the nest for more than five consecutive days. The first day the female was first absent from the nest was the date recorded and the number of days since October 1 and this date was calculated (FFT). See file BehaviouralData.csv.
Nest structure (NS)
NS was recorded on three separate dates throughout each breeding season on December 1, January 1, and January 15, representing the beginning of incubation (NSDec1), guard (NSJan1), and creche (NSJan15) respectively. NS was assessed on whether the stones were stacked to create a nest bowl with a score of 1, 2, or 3. 1 represents a nest with few or no stones, 2 a nest with some stones but not stacked off the ground, and 3 where the nest bowl is off the ground. See file BehaviouralData.csv.
Nest location (NL)
NL was recorded on three separate dates throughout each breeding season on December 1, January 1, and January 15, representing the beginning of incubation (NLDec1), guard (NLJan1), and creche (NLJan15) respectively. This was recorded as the number of active nests from the periphery of the colony, where 1 is the edge of the colony, and the location number increases towards the colony centre as a continuous variable. See file BehaviouralData.csv.
