PREDICTING SUITABLE MARINE HABITAT FOR PINK-FOOTED SHEARWATERS (<i>ARDENNA CREATOPUS</i>) IN THE WATERS ALONG THE PACIFIC COAST OF CANADA
<p dir="ltr">This set of files contains the R code and input data used to run a <b>Generalized Additive Model (GAM)</b> for <b>Pink-footed Shearwaters (PFSH)</b> in Canadian Pacific waters. The model uses standardized at-sea survey data (1992–2019) to predict...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <p dir="ltr">This set of files contains the R code and input data used to run a <b>Generalized Additive Model (GAM)</b> for <b>Pink-footed Shearwaters (PFSH)</b> in Canadian Pacific waters. The model uses standardized at-sea survey data (1992–2019) to predict species presence and relative abundance in relation to key environmental covariates.</p><p dir="ltr">The <code>.R</code> script includes data preparation, model fitting, validation, and prediction steps. The accompanying <code>.csv</code> file provides the standardized 2 km² observation dataset, and the shapefile defines the study area boundary used for mapping model outputs.</p><p dir="ltr"><br></p><p dir="ltr">Manuscript Abstract: Anthropogenic activities are threatening global marine ecosystems, with seabirds representing a vulnerable group that has experienced pronounced population declines in recent decades. The ability to identify important marine areas for vulnerable seabirds is fundamental to conservation initiatives. The Pink-footed Shearwater (<i>Ardenna creatopus</i>; listed as Endangered in Canada) breeds only in Chile, but during the non-breeding season it ranges northward to waters off Canada’s Pacific coast and the northern Gulf of Alaska. Utilizing at-sea survey data spanning from 1992 to 2019, we examined the relationship between the species’ distribution and environmental variables using a two-step Generalized Additive Model approach. Cross-validation with out-of-sample testing showed high predictive accuracy for occurrence (AUC = 0.94) and moderate performance for abundance predictions (Spearman’s rank correlation = 0.32, RMSE = 3.92, MAE = 0.45) at a 4 km² resolution. The results give us confidence in the model’s ability to identify areas suitable for Pink-footed Shearwaters. Distribution was strongly associated with several oceanographic and geographic factors, particularly latitude and distance to the continental shelfbreak. The findings of this study may help inform marine conservation efforts within Canada’s Pacific Exclusive Economic Zone and beyond.</p><p dir="ltr"><br></p> |
|---|