Figure 1.
Overview map of right whale habitat.
Overview map of the study area with the 9 geographic areas labeled at the approximate centroid for each region. Inset map highlights the regions contained within the greater Gulf of Maine. Abbreviations are as follows: NRTH = North region, BOF = Bay of Fundy, JL = Jeffreys Ledge, GOM = Gulf of Maine, RB = Roseway Basin, NE = Northeast, GSC = Great South Channel, MIDA = Mid-Atlantic, and SEUS = Southeastern US.
Figure 2.
Plate showing examples of the different classes of body condition judged by the evaluation of the dorsal back profile in the post-blowhole region: a) good, b) fair, and c) poor. Note the contrast between the level to convex nuchal area in a) and the concavity where the back drops off behind a pronounced hump in c). Photographs were taken under permits from the National Marine Fisheries Service (#15415) and the Department of Fisheries and Oceans, Canada. Photo credit: New England Aquarium.
Figure 3.
Photo plate depicting whale #1621, an adult male last seen in 2000. Photo in panel a) was taken in 1997, and panel b) was taken in 1999. Photos show examples from four of the visual health parameters: 1) entanglement severity, 2) presence of orange cyamids around the blowholes, 3) poor skin condition, and 4) rake marks forward of the blowholes. There are multiple ordinal classes within each health parameter. For example, in a) skin condition received a score of 2– good skin condition, while in b) skin condition would receive a score of 1– poor skin condition.
Figure 4.
Graphical depiction of the statistical model.
Graphical model depicting the dependency structure. We have observation models for the visual health parameters H, for survey effort E, and for sightings Y (top panel). The middle panel comprises two process models for the latent states of health h, and movement, z. Lastly, survival, s, is estimated as a function of latent health and movement.
Figure 5.
Movement transitions for male right whales.
Posterior estimates of transitions made by male right whales between regions over the course of the year. Size of the circles in each region at each month correspond to the actual number of male right whales observed. Lines connecting regions indicate probability of transition, pjkt >0.25. Magnitude of probability is depicted by line thickness.
Figure 6.
Movement transitions for female right whales.
Posterior estimates of transitions made by female right whales between regions over the course of the year. Lines and circles as in Figure 5. In contrast to males, females spend more time in BOF, and have more estimated transitions to SEUS at the end of the year.
Figure 7.
Movement transitions for whale #3911.
Heat map depicting Pr(zi,t = k) for the entire sighting record of #3911. Darker blue represents higher probability; darkest blue indicates animal was sighted in that region, i.e., #3911 was observed in August and September of 2009 in BOF. In the latter half of 2010, probable transitions are from BOF to GOM to MIDA to SEUS where the animal was observed in December. Vertical black lines indicate the start of 2009, 2010, and 2011.
Table 1.
Estimates of the beta parameters in the regression for health, ht. Health at the previous time-step has a strong relationship to health at current time-step.
Figure 8.
Health time series for whale #1333.
Time series of health observations H for skin condition, body condition, cyamids and rake marks (circles), estimates with uncertainty of health h (thick line and dashed lines), and estimates of survival s (height rectangle at bottom) for #1333. Photographic observations of H are color and size coded by class. For a visual health parameter with three categories, e.g., body condition, green is the best category, orange is fair, and purple is poor. For a visual health parameter with two categories, e.g., skin condition, green is the best category, while orange is poor. #1333 had a gradual decline in health over a period of years, and was observed dead in 1998.
Figure 9.
Health time series for whale #1245.
Photographic observations of H are color and size coded as in Figure 8. Additional health observations H, include periods of entanglement, and calving status. #1245 has experienced several periods of compromised health, but has recovered from each, and is currently imputed to be alive.
Figure 10.
Health time series for whale #3911.
Photographic observations of H are color and size coded as in Figure 8. #3911 was only alive briefly, had declining health towards the end her life as a result of a severe entanglement, and was observed dead in February of 2011.
Figure 11.
Health time series for whale #1077.
#1077 has not been seen since 2004. At the time of the last sighting its health status was poor. Estimates of health since the last sighting have declined to 0, and the animal is currently presumed to be dead.
Table 2.
Estimates for the parameters for survival as a function of health and location.
Figure 12.
Graphical representation of the parameters in matrix c, which relate observed health H to true health h. The solid curves represent the estimated probability with uncertainty (dashed lines) that given an estimate of true health h, the animal would be seen in a given health class for each of the 6 health parameters. Priors are depicted with dotted lines. Estimates for the parameters governing body condition, rake marks, skin condition, and cyamids show a clear relationship between true health status and observed health. In contrast, the relationship between calving and entanglement and true health is not clear.