Research Question

My research objective was to investigate the spatial scale and locations of surface feeding sites of humpback whales (Megaptera novaeangliae) in the Stellwagen Bank National Marine Sanctuary (SBNMS) in the southern Gulf of Maine, USA, and to investigate movement parameters of the whales.

Dataset

The existing data for this analysis stems from a long-term study investigating humpback whale behavior and ecology in the southern Gulf of Maine, USA, (for more details, see Friedlaender et al. 2009). Almost every summer since 2004, whales were equipped with non-invasive tags that recorded detailed information on the underwater movement of the whales or collected video-footage of the behavior of the tagged animal and associated whales. During daylight deployments that usually lasted for up to 8 hours, focal follows were conducted from a small boat following the tagged whale, during which detailed information on the behavior of the tagged whale at the water surface was collected. Because the tags did not contain a GPS, range and bearing information on the whale were also collected at least once when the whale was observed at the surface in between consecutive dives, usually resulting in the collection of one location point every 3-5 min (temporal extent: 117-260 min), with a spatial resolution of tens or hundreds of meters (less when the whale is foraging, more when it is traveling) (extent: 10³-10⁴ m). Continuous GPS locations of the boat were automatically collected. Based on the time stamps of the range/bearing data and the boat GPS data, the GPS locations collected at or close to the time the range and bearing data was collected were identified and together this data was used to calculate the locations of the whale.

Hypotheses

• Surface feeding areas (FPT radius) are larger than mean length of prey schools (139 m, Hazen et al. 2009)
• Size of surface feeding areas (FPT radius) positively correlated with group size (larger groups search larger area)
• Positive correlation between FPT (at FPT radius) and depth and slope (prey habitat preference)
• No autocorrelation in FPT
• No autocorrelation of absolute turning angles on spatial scale of surface feeding sites (unpredictable foraging movement)

Approaches

• Scan focal follow data for deployments during which surface feeding was major behavior
• For these deployments, use R package adehabitatLT (Calegne 2011) to implement first-passage time (FPT) metric
  • Based on the location points of an animal and the time spend at/between these points, FPT radius identifies the spatial scale on which foraging effort along an animal’s path is concentrated (Fauchald & Tveraa 2003)
  • The time spent within this radius of each location point is called FPT. Location points with high FPT identify important foraging areas along the animal’s path (Bailey & Thompson 2006)
• Compare FPT radius for each deployment to the group size during the deployment
• Map color-coded FPT (at FPT radius) for each deployment onto bathymetry and slope raster layer in ArcMap
• Extract depth and slope values for each location point
• Connect the location points for each deployment with lines to show path of the animal
• Buffer each location point by the value of the FPT radius for the respective deployment
• Dissolve buffer for better visibility of the range of foraging effort around the location points and relative to underlying depth and slope rasters
• Perform linear regression of FPT against depth/slope for each deployment and calculate r2 value
• Calculate autocorrelation in turning angles and first-passage time using Moran’s I coefficient of autocorrelation in the pgirmess R-package

Results

Three deployments (188a, 188b_f, 195b) were identified for which suitable location data existed and that contained a large amount of surface feeding. Because deployment 188b_f originally also contained a large amount of traveling, here, only the part of the path with surface feeding activity was included in the analysis. Table 1 summarizes the results of the analyses.

Table 1: FPT radius, group size and correlation coefficient r2 for the regression of FPT against depth and slope are shown for each of the three deployments.

• There does not seem to be a fixed spatial scale of surface feeding as the values of FPT radius range between 90 and 347 m.
• It is difficult to investigate a potential relationship between FPT radius and group size as the foraging groups were very dynamic with changes in group size throughout the deployments. However, the smallest FPT radius was calculated for the only group that consisted of group size 1 for some part of the deployment, and the largest FPT radius was observed for the group that had the largest group size during some part of the deployment.
• Significant relationships between FPT and depth or slope were found for two of the three deployments. For deployment 188a, depth explained 17.2 % of the variability in FPT (p=0.001). For deployment 188b_f, depth explained 14.2 % (p<0.001) and slope 29.2 % (p<0.001) of the variability in FPT (Figures 1-3)
• Autocorrelation in FPT was found for all three deployments at scales greater than the FPT radius for each deployment (Figure 4)
• Autocorrelation in absolute turning angles was found for two of three deployments at spatial scales much greater than FPT radius.

Figures 1-3: Whale paths, FPT and FPT radius mapped on slope chart of SBNMS (USGS/SBNMS). Whale locations for each deployment (circles) are color-coded to represent high/low FPT (red/green) at FPT radius (purple buffer) and connected with lines to show temporal sequence of locations. From top to bottom: 188a, 188b_f, 195b.

Figure 4: Moran’s I autocorrelation coefficient calculating FPT autocorrelation plotted against lag distance for all deployments. Vertical lines represent FPT radius for each deployment. Red circles indicate significant autocorrelation in FPT.

Discussion and Significance

The three deployments investigated here showed a wide range of spatial scales of surface feeding. However, there is some indication that this variability could at least in part be explained by the differences in group size between the three groups, with larger groups having a greater spatial scale of surface feeding. The spatial scale of surface feeding is somewhat consistent with the mean size of prey schools. Bathymetry and bathymetric relief seem to have some influence on the locations of concentrated search effort (FPT). This is obvious from both the correlation coefficients of FPT and depth/slope as well as from the map. However, the results of this analysis are obscured by the autocorrelation in FPT. Since the deployments are opportunistic observations, no absence data was analyzed here and the sample size is small, it is unclear whether non-feeding whales also concentrate their activity on similar locations within SBNMS and whether surface feeding also occurs over other parts of SBNMS with different ranges of depth/slope. The spatial scale of autocorrelation in FPT (451-596 m) is more similar between the three whales than the spatial scale of foraging effort. This could suggest the existence of a prey searching mechanism that is similar for all three whales and works on a scale larger than the foraging mechanism captured by the FPT analysis.

A better understanding of the foraging mechanisms of the whales can help to improve management decisions aimed at mitigating ship strike and entanglement risks in the SBNMS.

Learning outcomes

• Working with spatial data in R
• Implementing FPT metric
• Calculating spatial autocorrelation
• Extracting raster values in ArcMap

 References

Bailey, H. & Thompson P. (2006). “Quantitative Analysis of Bottlenose Dolphin Movement Patterns and Their Relationship with Foraging: Movement Patterns and Foraging.” Journal of Animal Ecology 75 (2): 456–65.

Calenge, C. (2011). “Analysis of Animal Movements in R: The adehabitatLT Package.” Saint Benoist, Auffargis, France: Office Nationale de La Chasse. http://cran.gis-lab.info/web/packages/adehabitatLT/vignettes/adehabitatLT.pdf.

Fauchald, P. & Tveraa T. (2003). “Using First-Passage Time in the Analysis of Area-Restricted Search and Habitat Selection.” Ecology 84 (2): 282–88.

Friedlaender, A. S., Hazen, E. L., Nowacek, D. P., Halpin, P. N., Ware, C., Weinrich, M. T., Hurst, T., Wiley, D. (2009). Diel changes in humpback whale Megaptera novaeangliae feeding behavior in response to sand lance Ammodytes spp. behavior and distribution. Marine Ecology Progress Series 395: 91–100.

Hazen E.L., Friedlaender A.S., Thompson M.A., Ware C.R., Weinrich M.T., Halpin P.N., Wiley D.N. (2009). “Fine-Scale prey aggregations and foraging ecology of humpback whales Megaptera Novaeangliae.” Marine Ecology Progress Series 395: 75–89.

Jennifer Allen and Michael Thompson calculated and provided the whale locations based on boat GPS and range and bearing data. Thank you!