The application of GIS and spatiotemporal analyses to investigations of unusual marine mammal strandings and mortality events

• Published in: Marine Mammal Science
• Main Authors: Norman, Stephanie A., Huggins, Jessie, Carpenter, Tim E., Case, James T., Lambourn, Dyanna M., Rice, James M., Calambokidis, John, Gaydos, Joseph K., Hanson, M. Bradley, Duffield, Deborah A., Dubpernell, Sandra, Berta, Susan, Klope, Matt
• Format: Text
• Language: unknown
• Published: PDXScholar 2011
• Subjects: Geographic information systems > Applications to marine research; Marine mammals > Stranding > Causes; Harbor porpoise > Mortality > Evaluation; Pacific; Phocoena phocoena
• Online Access: https://pdxscholar.library.pdx.edu/bio_fac/2; https://doi.org/10.1111/j.1748-7692.2011.00507.x; https://pdxscholar.library.pdx.edu/context/bio_fac/article/1001/viewcontent/theapplicationofgisandspatiotemporalanalyses.pdf

In 2006–2007, an unusually high number of harbor porpoises (Phocoena phocoena) stranded along the Washington and Oregon coastlines. Spatiotemporal analyses were used to examine their ability to detect clusters of porpoise strandings during an unusual mortality event (UME) in the Pacific Northwest using stranding location data. Strandings were evaluated as two separate populations, outer coast and inland waters. The presence of global clustering was evaluated using the Knox spatiotemporal test, and the presence of local clusters was investigated using a spatiotemporal scan statistic (space–time permutation). There was evidence of global clustering, but no local clustering, supporting the hypothesis that strandings were due to more varied etiologies instead of localized causes. Further analyses at subregional levels, and concurrently assessing environmental factors, might reveal additional geographic distribution patterns. This article describes the spatial analytical tools applied in this study and how they can help elucidate the spatiotemporal epidemiology of other UMEs and assist in determining their causes. More than one spatial analytical technique should be used if the study objective is to detect and describe clustering in time and space and to generate hypotheses regarding causation of marine mammal disease and stranding events.