Dahlheim et al.: Interactions, injuries, and mortalities of Orcinus orca during fishing operations 83 
code, number of whales present, and details summariz- 
ing whale behavior. 
Only those images that were in focus, had adequate 
lighting, and were of a quality high enough to accurately 
identify an individual killer whale or the ecotype involved 
(i.e., shape of dorsal fin or saddle patch, whale markings, 
and clearly visible scars) were used in the analyses. Deter- 
mination of killer whale ecotype was done by examining a 
whale’s external morphology (i.e., fin shape and size and 
shape of saddle patch), by comparing the photographic 
data to images of known Alaska killer whale ecotypes 
(i.e., images in Dahlheim, 1997, and in unpublished 
catalogs maintained by both the MML and the North 
Gulf Oceanic Society. In addition, different ecotypes of 
killer whales do not intermix. If unknown whales are 
seen with known members of a resident pod, those 
unknown whales (linked by association) are considered 
residents. Photographic data were also used to document 
injuries and mortalities of killer whales. Any dead killer 
whale (either killed directly in a fishing operation or 
found floating at sea) was brought on board the fishing 
vessel to be photographed and examined by the observer. 
Whenever possible, the dead animal’s sex, age class (i.e., 
calf, juvenile, or adult) was determined and external 
measurements were collected (AFSC*”). The results of 
a review of the observer’s photographs and associated 
interaction form provide information on the plausible 
cause of death. 
Tissue collection and genetic analyses 
In 1999, the Cetacean Tissue Collection project was ini- 
tiated by the MML through requests made to observers 
to collect tissue samples from any dead cetacean brought 
aboard a fishing vessel by using MML-provided sampling 
kits; written instructions were included in the observer 
manual (see AFSC‘). When possible, an 8-cm? sample of 
skin and blubber was obtained from each carcass. The 
samples were stored in a vial containing buffered salt 
solution with dimethyl sulfoxide and labeled with the 
observer’s cruise number and the date, location, and 
detailed circumstances related to the event (e.g., cause 
of death when possible, location on the whale’s body 
where sample was taken, whale measurements, and car- 
cass condition). With the exception of 2 tissue collections 
(biopsies) taken from live whales, all tissues collected 
were taken from dead whales that were brought on board 
the vessel. 
Genetic samples from killer whales were analyzed at 
the NMFS Southwest Fisheries Science Center. Genetic 
results include species verification, sex determination, 
and when possible, identification of the killer whale eco- 
type. Results of the analyses were returned to the MML 
to update the Cetacean Tissue Collection database and to 
assign the injured or dead animal to a particular stock. 
Genetic results were also forwarded to staff members of 
the NMFS Alaska Regional Office to provide them with 
vital information for assessing the effects to a specific 
killer whale ecotype or stock. 
Results 
Fishery interactions 
Between 2001 and 2016, observers reported 3245 (2415 in 
full-coverage strata) fishery interactions with killer whales 
from 16 NMFS reporting areas ranging from the Bering Sea, 
throughout the Aleutian Islands, and eastward into Southeast 
Alaska (Fig.1). There were 3026 hauls (2268 in full-coverage 
strata) during which at least one interaction occurred (Suppl. 
Table 1A) (online only) out of the 680,872 hauls (534,601 in 
full-coverage strata) that occurred on observed trips over the 
16-year period. The percentage of the hauls for which inter- 
actions with killer whales were reported varied by area, with 
the highest percentage of hauls with interactions occurring in 
the Bering Sea (Fig. 2, Suppl. Table 1B [online only]). Interac- 
tions with killer whales also varied by year within each area 
(Suppl. Table 1, A and B) (online only), by month or season (Fig. 
3), and by type of interaction (Fig. 4). 
Killer whales feeding on catch before the catch was 
landed, killer whales feeding on discards, and vessel crew 
members using varied methods of whale deterrence rep- 
resented 96% of all observed interactions recorded, and 
these interaction types consistently accounted for higher 
percentages of affected hauls than other interaction types 
(Suppl. Table 2, A and B) (online only). Feeding on catch 
that had not yet been landed was the most common inter- 
action documented during this study. Feeding on discards 
was known to occur prior to 2008, on the basis of records 
maintained by MML staff; however, this whale behavior 
was not formally documented by observers until 2008. 
Figure 4 shows that the use of deterrence methods was 
relatively low for the years prior to 2008 and that the fre- 
quency of deterrent use appears to have increased in more 
recent years. The areas in which these 3 types of interac- 
tions occurred by season is depicted in Figure 5. 
We found that 2527 hauls (84%) with at least one inter- 
action with killer whales occurred during longline opera- 
tions (Suppl. Table 3A) (online only). Longline operations in 
the Bering Sea and Aleutian Islands (BSAI) accounted for 
2130 hauls (84%) with interactions, whereas longline fish- 
eries in the GOA represented only 397 hauls (13%) with 
interactions (Suppl. Table 3A) (online only). There were 
56 hauls with interactions with killer whales that were 
assigned to the unknown fishery category (for 38 hauls 
of longlines and for 18 hauls of trawl gear, the collected 
data corresponds to none of the fisheries in the LOF). An 
additional 38 hauls could not be categorized to a specific 
fishery because the necessary data were not available. 
Interactions with killer whales were also reported for 
trawl operations in the BSAI and GOA, during which 
322 hauls with interactions (11%) were documented over 
the 16-year period (Suppl. Table 3A) (online only). The great- 
est number of interactions occurred in the BSAI flatfish 
trawl fishery (n=272). Few, if any, interactions occurred in 
the fishery that uses non-pelagic trawl gear. 
The operations of the pot gear fishery that targets sable- 
fish in the BSAI logged 82 hauls with at least one interac- 
tion over the entire period (Suppl. Table 3A) (online only). The 
