Using stable isotope analysis to understand the migration and trophic ecology of northeastern Pacific white sharks (Carcharodon carcharias)

Abstract

The white shark (Carcharodon carcharias) is a wide-ranging apex predator in the northeastern Pacific (NEP). Electronic tagging has demonstrated that white sharks exhibit a regular migratory pattern, occurring at coastal sites during the late summer, autumn and early winter and moving offshore to oceanic habitats during the remainder of the year, although the purpose of these migrations remains unclear. The purpose of this study was to use stable isotope analysis (SIA) to provide insight into the trophic ecology and migratory behaviors of white sharks in the NEP. Between 2006 and 2009, 53 white sharks were biopsied in central California to obtain dermal and muscle tissues, which were analyzed for stable isotope values of carbon (δ(13)C) and nitrogen (δ(15)N). We developed a mixing model that directly incorporates movement data and tissue incorporation (turnover) rates to better estimate the relative importance of different focal areas to white shark diet and elucidate their migratory behavior. Mixing model results for muscle showed a relatively equal dietary contribution from coastal and offshore regions, indicating that white sharks forage in both areas. However, model results indicated that sharks foraged at a higher relative rate in coastal habitats. There was a negative relationship between shark length and muscle δ(13)C and δ(15)N values, which may indicate ontogenetic changes in habitat use related to onset of maturity. The isotopic composition of dermal tissue was consistent with a more rapid incorporation rate than muscle and may represent more recent foraging. Low offshore consumption rates suggest that it is unlikely that foraging is the primary purpose of the offshore migrations. These results demonstrate how SIA can provide insight into the trophic ecology and migratory behavior of marine predators, especially when coupled with electronic tagging data.

Figure 1. White shark focal areas from satellite tag data from Jorgensen et al. .

Regions used in the mixing model are indicated (California, Pelagic, Hawaii), see text for details. White shark aggregation sites in central California where tissue collection occurred are designated with the star. Mean chlorophyll-a concentration for 2006 (Aqua MODIS, http://oceanwatch.pfeg.noaa.gov) showing productivity gradients which isotopic gradients generally follow. Inset shows conceptual diagram showing how the seasonal migration of white sharks take them between regions with isotopically distinct prey.

Figure 2. Probability of white sharks occurring in different focal areas by time of year based on satellite tag data from Jorgensen et al. .

Figure 3. Relationship between white shark length (TL) and δ¹³C and δ¹⁵N of white shark muscle.

The two ellipses contain the δ¹³C and δ¹⁵N values of two of the smallest sharks that had the highest δ¹³C and δ¹⁵N values.

Figure 4. δ¹³C and δ¹⁵N values of white shark tissues and potential prey items from different focal areas.

Regions and shark tissues are designated with different symbols and colors. For regions, smaller symbols show mean (SD) isotopic values of individual species and larger symbols indicating mean regional average (SD). 4a: δ¹³C and δ¹⁵N values for different prey, regions and white sharks tissues. Individual species are labeled according to Table 1. 4b: Mixing polygon used to estimate contribution of different regions to white shark tissue. Regional (and prey) values are adjusted to account for discrimination factor (Kim et al. [67]). Error in discrimination factors was propagated into error of mean regional values. Muscle (est) is the predicted mean stable isotope composition of white shark muscle if sharks foraged at the same rate in the different regions. Dermis (corr) is the mean dermal stable isotope composition adjusted to resemble muscle by accounting for differences in discrimination factors between the tissues; the adjustment is based on Hussey et al. . Note that the closer the tissue values are to source (region) values, the higher the contribution of that source to the tissue.

Figure 5. Overall contribution of different focal areas to white shark tissues estimated using the spatially explicit Bayesian mixing model.

Results show posterior model estimates (median, interquartile range and max/min values) of source (CA: California, PEL: Pelagic, HI: Hawaii) contribution to muscle. The estimated contribution of the different focal areas incorporates electronic tag data from Jorgensen et al. and is estimated for three different tissue incorporation rates (SB: juvenile sandbar shark, BA: allometrically scaled bird, FA: allometrically scaled fish, see text for details).

Figure 6. Rate of consumption in offshore focal areas (PEL: Pelagic, HI: Hawaii) relative to California (CA) estimated using the spatially explicit Bayesian mixing model.

Results show posterior model estimates (median, interquartile range and max/min values). The green dashed line designates the relative consumption rate in the California focal area. Consumption rate is estimated using three different tissue incorporation rates for both offshore focal areas (SB: juvenile sandbar shark, BA: allometrically scaled bird, FA: allometrically scaled fish, see text for details).