Modeling pygmy sperm whale (Kogia breviceps, De Blainville 1838) strandings along the southeast coast of the United States from 1992 to 2006 in relation to environmental factors.
Pygmy sperm whales are the second most commonly stranded marine mammal in the Southeastern Unites States (SEUS). They most often strand alive and the causes of these events remain largely unknown. Generalized linear models were built to identify potential relationships among environmental factors an...
| Main Authors: | , , |
|---|---|
| Format: | Text |
| Language: | unknown |
| Published: |
U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Ocean Service
2015
|
| Subjects: | |
| Online Access: | https://dx.doi.org/10.7289/v5/tm-nos-nccos-203 https://repository.library.noaa.gov/view/noaa/12941 |
| Summary: | Pygmy sperm whales are the second most commonly stranded marine mammal in the Southeastern Unites States (SEUS). They most often strand alive and the causes of these events remain largely unknown. Generalized linear models were built to identify potential relationships among environmental factors and the occurrence of pygmy sperm whale strandings in the SEUS. Two methods were used to model environmental parameters depending on the nature of the data. One method used data from NOAA buoys compiled over a week before a stranding event. Predictor variables included hourly wind direction and speed, wave height, average wave period, barometric pressure, and water temperature. The other method used Sea Surface Temperature data from satellite images compiled monthly, monthly Multivariate El Niño Southern Oscillation Index (MEI), and bathymetric data. Frontal features were extracted from the images using ArcMap Geographic Information System and landscape metrics were computed on these images in FRAGSTATS. The model compiled from buoy data was relatively stronger (AIC = 497.5) at predicting strandings. It indicated that more strandings occurred when there were sustained high wind speeds, low barometric pressures, and swell waves in the week before stranding events. While the other model was relatively weaker (AIC = 718.7), it showed that less numerous fronts and high MEI index were generally associated with a higher number of strandings. This study is a step toward appreciating which environmental factors may contribute to the observed marine mammal stranding patterns as well as the distribution of pygmy sperm whales. It is an attempt at building predictive statistical models that could be useful for the management of cetaceans. |
|---|