An Annotated Bibliography

This paper aimed to correlate energetics of a harbor seal (Phoca vitulina) population with dive and foraging patterns. The study was conducted in a coastal archipelago near Sandoy, Norway. 13 individual animals were tagged and tracked via VHF radio tracking, and data from 13 other individuals from a nearby area was used to establish typical dive patterns. Empirical data of fishery locations was gathered, as well as dietary composition of the seals (via otoliths).

Study objectives included simulating energetics for the seal populations, identify common prey species, describe relationship between foraging to fishing operations, and relate this data to seal-fisheries interactions. A 'GIS-based habitat model' was used to assimilate and analyze spatial data. The results of the study demonstrated that harbor seals in the area forage heavily on fisheries species (50% of their diet was determined to be Norway pout). They also fed on haddock, herring, saithe, and various codfish.

This study was interesting to me because it demonstrated that pinniped/fisheries interactions are of international concern, and we can utilize the tools of GIS and radio telemetry to solve some of these issues. GIS allowed the incorporation of spatial data from fishery activities and pinniped foraging, which was not examined prior to this study.

Cooper et al. examined the potential effects of sea ice loss on mother/calf seperation in the Arctic Ocean. They were able to utilize GIS in many aspects of this study, including bathymetry, sea surface temperature, and water depths. Using interpolations of specific data points, they were able to correlate sea ice retreat with mother/calf seperations.

This paper is important because it explains how physical oceanographic changes can significantly alter the reproductive success and behavior of a pinniped species. The events discussed in the study are apparently novel, and are correlated with a changing ecosystem. Walruses are important predators in the Arctic Ocean, and we are likely to see many behavioral alterations for walruses in the next few decades, as they struggle to adapt to a physically-changing ecosystem.

This particular paper studied spatial and temporal extents of harp seal (Phoca groenlandica) and hooded seal (Cystophora cristata Erxleben) strandings across the Gulf of Maine. They were able to utilize stranding data from NOAA's Marine Mammal Stranding Network to determine characteristics such as species, location, time, age, and condition of the animal (alive, diseased, etc.).

GIS vector data was used to determine location, intertidal shore type, and off-shore variables. GIS was also used to map strandings. Additionally, human-related variables were also incorporated, including political boundaries, public lands, and census data.

This study concluded that hooded seals tend to strand further north than harp seals, in 50-100 meter water depths, and in a variety of shoreline types. Additionally, hooded seals were more likely to strand alive (78% versus 55%). Temporally, both species tended to strand further north in early winter and spring, and further south in late winter. These patterns were likely due to seasonal movements of populations.

For those who participate and coordinate marine mammal stranding networks, data such as these can be important to predicting frequencies and locations of strandings, and also correlating stranding behaviors to annual climatic events. GIS is a fantastic tool for this, as it can not only incorporate temporal and spatial data, but also attributes of stranded animals for future record.

The Kuhn et al. paper focused strongly on the use of GIS for foraging behavior analysis in northern fur seals. The study was conducted from 2007 to 2008 in St. Paul Island, Alaska. 42 adult female fur seals were tagged and tracked over the study period. Dive patterns were analyzed using IKNOS-DIVE software, and mapped according to geographical coordinates.

Specific characteristics of dives included dive time, depth, surface interval, descent and ascent rate, vertical distance and excursions, and dive efficiency. Dives were classified via the criteria of area-restricted search (ARS). Finally, using these characteristics, individuals were classified as epipelagic, benthic, or mixed strategy divers.

This study was able to characterize different foraging strategies based on GPS-based data. It demonstrates that using GIS systems, we can begin to establish short- and long-term patterns in pinniped behavior that would otherwise be inaccessible for study.

The objective of this study was to establish spatial models of harbor seal haul-outs in a tidal estuary in Southern Alaska (Cook Inlet). Aerial surveys were conducted from 2003 to 2008 to determine the location of haul-out sites throughout the estuary. In order to incorporate different environmental characteristics of each haul-out site, a GIS model was created by the research team. Also included in the model were number of seals at each haul-out site, substrate type and wave exposure, distance to water, distance to seabed, human disturbance, and access to anadromous fish populations.

Findings of this study included the fact that characteristics of popular haul-out sites differ seasonally. Characteristics that may be important in April (such as substrate type and proximity to salmon runs) are not equally as important in October or other months. However, harbor seals do tend to select rocky substrate types for haul-outs more often than sandy substrates. Other important factors in haul-out site selection included anthropogenic disturbances, access to deep water, and proximity to anadromous fish streams.

What I found most interesting about this study is the factor of temporal variance in habitat preference of marine mammals. Instead of relying on the old expectations that few large-scale factors determine habitat preferences, we must recognize that many integrated factors- both spatial and temporal- affect behavioral choices in marine mammals.

S.A. Norman presents us with the first review of the utilizations in GIS technology in disease epidemiology and spatial patterns. The author demonstrates the possible uses for GIS technology for predicting and understanding clusters of disease outbreaks. Because pinnipeds are gregarious when they haul out, the potential always exists for rapid disease and parasite transmission.

Norman describes several specific instances where GIS can be used to map disease outbreaks, including phocine distemper, leptospirosis, or morbillivirus. Her perspective is a new one, in that most GIS applications for pinnipeds lie within the realms of dive behavior, hauling out characteristics, and foraging bouts. It is clear that GIS will have increasing importance in pinniped studies (and marine mammal studies) as we begin to collect more knowledge about these animals.

This paper also lists a brief review of many studies that utilize GIS to analyze pinniped behavioral patterns.

Rogers et al. examined seasonal movements of leopard seals in the Antarctic in this study. They tagged 19 individuals of intermediate ages over several seasons. Capture and tagging occurred in the Antarctic Peninsula region. After individual movement patterns (swim speed, daily distance, total distance) were recorded, data were analyzed using GIS software (ArcView 3.2). GIS software allowed researchers to integrate independent data points into a spatial model of seal movements.

This study was interesting because the investigators were able to establish seasonal movements of a very difficult-to-track species. They also established basic parameters such as average swim speed, monthly variation in travel distance, and differences in swim speed throughout different seasons. Using GIS to integrate these data over longer time periods can allow us to collect an information database, and observe changes over long periods of time.

Isabelle Schmelzer examined monk seal movements in the northwestern Hawaiian Islands. The Hawaiian monk seal is an endangered species that is endemic to the NW Hawaiian islands. This being so, the main objective of this study was to examine spatial correlations in monk seal distributions in relation to the physical environment.

The first step of this experiment utilized GIS to create mesoscale maps of oceanographic and geographic characteristics such as sea surface temperature, productivity (chlorophyll concentrations), and water column structure. Next, habitats were divided and categorized based on their individual features. Finally, monk seal data were gathered via various population counts.

The author found that the NW Hawaiian Islands are strongly heterogenous environment, and monk seals show strong site fidelity to different areas. A large number of subpopulations has strong implications for ecosystem-based management as a method to protect this species from extinction. I found this paper really interesting because it demonstrates that in the case of conservation, it is very important to consider many multivariate factors in order to construct a management strategy for a species or population.

Another paper by Twiss et al., which uses GIS to predict expanding haul out sites within a growing population of grey seals. This study was conducted at the Isle of May, UK. The authors use a high-resolution Digital Terrain Model (DTM) to establish characteristics of sites, and then examine whether suitable sites are occupied by grey seals. The data used in this study was from the breeding seasons (fall) of 1994 and 1998.

Seal locations were digitised, and differentiated based on sex and age (adult or juvenile). After data from 1994 were collected and analyzed, they were used to predict haul-out behavior of grey seals in 1998. The model created from 1994 data correctly predicted the highest-density haul-out areas for calves observed in the 1998 season.

This paper demonstrates that GIS can be used to effectively predict population densities and distributions using data from previous years. This could have management and conservation implications for many pinniped species, since they often exhibit site fidelity year to year. Similar data could also be used to predict distributions of growing or shrinking breeding populations.

Twiss et al. utilize many aspects of GIS to estimate grey seal body mass and size by remote methods, namely aerial photography. Remote methods allow us to study pinnipeds without disturbing them, or using invasive methods. Therefore, using GIS and DEMs (Digital Elevation Models) are potential ways to collect physical data without capture and physical measures.

This study was based in the United Kingdom, at two study sites: the Isle of May, and North Rona. Photographs were taken of grey seals at these sites, and the graphic data was analyzed using ARCEdit. The methods used in this paper were novel and therefore merit some improvements if they are to be utilized effectively. The authors see potential for the development of more complex models to establish measurements, and reduce error due to shadows, seasonality and differing weather conditions.

This study describes the analysis of short term movements of harbor seals in coastal Massachusetts. Four individuals were tagged during the initial experiment from May 3-7, 1999 in Western Penobscot Bay, and 39 additional individuals were tagged between March 14 and April 20, 2001. In addition to tagging, the researchers collected data on size, weight, age, gender, and reproductive status of individual subjects.

Tracking of tagged individuals proceeded from May 1-14, 2001. 75% of study subjects were detected at least once. Seals generally stayed in approximately a 25km area of initial tagging, with two exceptional individuals traveling 60km during the tracking period.

The site fidelity of harbor seals in this study suggests some amount of stock structure within the population. Even though this was a somewhat small-scale study, it demonstrated seasonal movement patterns of different age classes through the Penobscot Bay area. We can use this type of information to establish management strategies, predict behavior, and better understand stock dynamics of pinniped species.

This paper examined spatial movement patterns of a species local to the California Current System, the California sea lion. In 2005, the CCS experienced anomalous warm conditions which likely resulted in the redistribution of prey items throughout the coastal system. Researchers tagged and tracked 25 male sea lions near Montgomery Bay, California. Using ArcGIS, diving positions and foraging patterns from 2004 and 2005 were cataloged and compared.

During 2005, adult sea lions spent much more time foraging offshore, and less time hauled out as compared to the normal year of 2004. In addition, foraging dives were of longer extent than those in the previous year. These behavioral changes are likely related to decreased prey availability near shore.

This paper was very interesting because it showed that climatic fluctuations, such as El Nino oscillations, or anomalous conditions, can have viable impacts on local populations of fauna- including upper trophic level predators. As noted by the authors, this paper demonstrated the necessity of pinnipeds to exhibit behavioral plasticity in response to environmental stochastic events. Additionally, I found this paper to have very clear figures, and was very well-written.

Fay G., A.E. Punt. 2006. Modeling spatial dynamics of Steller sea lions (Eumetopias jubatus) using maximum likelihood and Bayesian methods: evaluating causes for population decline. Sea Lions of the World, Fairbanks, AK: Alaska Sea Grant College Program, pp 405–433

Galimberti F., S. Sanvito. 1999. A very spatial relationship: GPS mapping aids understanding of elephant seal behavior. GPS World, 10: 22–28.

Guinet C., L. Dubroca, M.A. Lea, S. Goldsworthy, Y. Cherel, G. Duhamel, C. Bonnadona, J.P. Donnay. 2001. Spatial distribution of foraging in female Antarctic fur seals Arctocephalus gazella in relation to oceanographic variables: a scale-dependent approach using geographic information systems. Marine Ecology Progress Series, 219: 251–264.