Fig 1.
Simplified Western Antarctic Peninsula (WAP) food web, focused on connections with seabird species (Illustrations by Freya Hammar).
Sea ice algae and pelagic phytoplankton are displayed as separate components, although there may be overlap in species. This is to highlight the importance of sea ice in this ecosystem, as increases in sea surface and air temperatures increase there are impacts on sea ice along the WAP.
Fig 2.
Annual and summer averaged environmental data for the South Shetland Islands and surrounding Southern Ocean.
Air temperature data (A) were derived from monthly measurements from five stations around the South Shetland Islands [43]. Sea surface temperature (SST) data (B) were extracted from compiled data (HadISST 1˚ daily; [44]) from the surrounding region (60° to 64° S, 55° to 66° W). The annual air temperatures have increased in this region (R2 = 0.11, F(1,45)=5.71, p value = 0.021), while there is no significant trend in SST in the broader foraging region (R2= < 0.001, F(1,45)=0.028, p-value = 0.87).
Table 1.
Sources of the data used in this study.
Fig 3.
Map of the locations of the data from published reports and that were analyzed in this study (ESRI-derived map using data from NOAA National Centers for Environmental Information; International Bathymetric Chart of the Southern Ocean; General Bathymetric Charts of the Ocean).
Alongside each location, are the species that had data available and were used in this study. Due to lack of multi-species, publicly available data, our review did not extend beyond the South Shetland Islands (Illustrations by Freya Hammar).
Table 2.
The results of our multiple linear regressions and backward selection of best fit models. Species data (B) from each colony was the dependent variable with year (Y), summer sea surface temperature (S) and summer air temperature (A) as independent variables. Locations are listed from north to south. The number of years of the data (n), the model results (Adj R2, F statistic, and p value), and results on each independent variable (Coefficient/Coef., standard error/SE, and p value/p) are indicated. Significant trends (p < 0.05) are indicated in bold.
Fig 4.
Seabird data used in this study (
Table 1) compared with the three independent variables used in analysis; year, sea surface temperature, and air temperature. Plot include points of the raw data, along with the line of best fit and the 95% confidence interval.
Table 3.
The relationships, measured by Pearson’s correlation coefficient (r) between avian prey species (CHPE, ADPE, GEPE*) and avian predator species (BRSK, SPSK, SGPE*) data used in this study. This analysis was only done between species and data types that had 3 or more data points from the same year. We classified r between 0.40 and 0.60 as moderate (italicized) and over 0.60 as strong (bold) correlations. The Pearson correlation coefficient (r) measures how closely related two variables are, from −1 as a perfect negative linear relationship to +1 as a perfect positive linear relationship, and 0 indicates that there is no linear correlation between them.