Horning and Meliish: Predation on Eumetopias jubatus by Somniosus pacificus in the Gulf of Alaska 
305 
Time of day (h) 
Figure 5 
Postmortem temperatures indicating rapid cooling recorded for 3 mortalities of Steller 
sea lions ( Eumetopias jubatus ) and subsequently transmitted by life history transmit- 
ters (LHX tags) with delayed onset of transmissions. The 3 events occurred in the 
Gulf of Alaska in 2011. Temperatures were recorded by 5 tags in 30-min increments 
and subsequently transmitted through the Argos satellite system. Data were received 
from only 1 LHX tag for animal number TJ52. The left horizontal arrow indicates 
that this tag sensed light or air and began transmissions 11 days post mortem. Data 
were received from 2 LHX tags for animal TJ64 (line, circles). The middle horizontal 
arrow indicates that these tags sensed air or light and began to transmit 5 and 9 d 
post mortem. The right horizontal arrow indicates that one LHX tag from animal TJ63 
(line) sensed light or air and began transmissions 6 d post mortem. The vertical ar- 
row indicates when the second tag from the sea lion TJ63 (circles) began to transmit 
after sensing light or air. Gray diamonds indicate the regional estimates of sea-surface 
temperatures (for sources, see Table 1). The gray shaded area at the top of this graph 
indicates the normal range of abdominal temperatures for healthy sea lions (Fig. 1). 
24 h post mortem. Conversely, 5 tags without delayed 
transmissions from hosts that died close to noon and, 
therefore, also provided 24 h of postmortem tempera- 
tures recorded maximum fluctuations of 0.82°C, 1.43°C, 
1.83°C, 2.2°C, and 2.4°C (Table 1). 
Discussion 
We previously reported that survival rates based on 
data from LHX tags for juveniles of ages 14-60 months 
in our study were indistinguishable from control val- 
ues from a separate study based on resightings of indi- 
viduals that had been hot-iron branded (Horning and 
Meliish, 2012; see Materials and methods section). Our 
probability for event detection was very high at 98.4%, 
indicating that, most likely, no events went undetected. 
This high level of probability supports the hypothesis 
that LHX tags or surgeries do not affect postrelease 
survival to 60 months in implanted juvenile Steller sea 
lions. All 15 detected mortality events with data were 
classified as death by predation on the basis of pre- 
cipitous temperature declines (14 events) or the very 
low mass indicated by the algor mortis model (1 event). 
In combination, these findings support the notion that 
none of the events reported here were caused by im- 
planted tags or tagging procedures; after all, it would 
seem highly improbable that LHX tags could increase 
the susceptibility of sea lions to predation without a 
concurrent decrease in long-term survival rates below 
control values. 
Theoretically, the observed patterns of rapid drops 
in temperature could be associated with direct extru- 
sion of a tag by a live host animal. Noncatastrophic 
extrusion (i.e., percutaneous) is a slow process that 
should result in substantially different times for the 
