298 
Fishery Bulletin 112(4) 
To address this constraint, we specifically developed 
the life history transmitter (LHX tag; Wildlife Com- 
puters 1 , Redmond, WA; described in Horning and Hill, 
2005) to detect and determine causes of Steller sea lion 
mortality irrespective of time and location of mortality, 
or age of host. LHX tags were designed to remain im- 
planted through the entire life of the host (Horning et 
ah, 2008). Surgical implantation allows multiyear mon- 
itoring beyond the annual molt that limits the use of 
externally attached devices. LHX tags do not transmit 
while inside the host body, greatly extending battery 
life beyond 10 years. Tags liberated from decomposing, 
dismembered, or digested bodies send transmissions to 
the Argos telemetry system (CLS America Inc., Lan- 
tham, MD) aboard NOAA satellites with global cover- 
age (Soma and Tsutsumi, 1986). LHX tags, therefore, 
provide postmortem data on time and date, location, 
and causes of mortality with spatially and temporally 
unlimited resight effort (Horning and Mellish, 2009). 
Here, we provide indirect evidence (from LHX tag data) 
of predation by Pacific sleeper sharks on juvenile Stell- 
er sea lions. 
Materials and methods 
This study was carried out in strict compliance with 
all applicable animal care and use guidelines under 
the U.S. Animal Welfare Act and was approved as re- 
quired under the U.S. Marine Mammal Protection Act 
and the U.S. Endangered Species Act by the National 
Marine Fisheries Service (permit numbers 1034-1685, 
1034-1887, 881-1890, 881-1668, 14335, and 14336) and 
by the Institutional Animal Care and Use Commit- 
tees of the Alaska Sea Life Center (permit numbers 
02-015, 03-007, 05-002, 06-001, 08-005, and R10-09-04) 
and Texas A&M University (permit numbers 2003-181, 
2005-170, and 2006-37). All surgeries were performed 
under aseptic conditions and under full anesthesia ad- 
ministered by inhalant gas, and all efforts were com- 
pleted in a way meant to minimize pain and suffering. 
Life history transmitters 
We used satellite-linked LHX tags (Horning and Hill, 
2005) to determine survival rates and infer causes of 
mortality in juvenile Steller sea lions in the Gulf of 
Alaska. LHX tags are intraperitoneally implanted ar- 
chival transmitters that monitor sensor data through 
the life of host animals. Select recorded data are trans- 
mitted (uplinked) post mortem to the Argos system 
aboard NOAA satellites, after the positively buoyant 
tags are liberated from a decomposing, dismembered, 
or digested carcass. LHX tags provide 48 h of tempera- 
ture data across detected mortality events. In 30-min 
1 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
increments, 96 temperature values are recorded with 
an accuracy better than 0.1°C and with a nonlinear 
resolution from 0.2°C (at 0°C) to 0.25°C (at 40°C), for 
48 ante mortem and 48 postmortem values. Mortality 
is detected when temperature drops below the preset 
threshold of 24°C. In addition to a temperature sensor, 
LHX tags incorporate a radiometric irradiance sensor 
and a proprietary radio-frequency immersion sensor 
(see Horning and Hill, 2005) that allows the classifica- 
tion of the medium surrounding the antenna portion of 
a tag as tissue, saltwater, or air. 
Tags begin to transmit after they sense surrounding 
air or sufficient ambient light. Thereafter, current tem- 
perature (hereafter, called uplink temperature) and ir- 
radiance are included in each transmission. Additional 
diagnostic sensor data transmitted by the tags allow 
the classification of ambient light conditions from time 
of death to onset of transmissions. Light conditions 
are classified into the categories of complete dark- 
ness or light exceeding a preset irradiance threshold 
of 2.9xl0 -8 Wcm -2 (for ?i=527-583 nm). This threshold 
corresponds with the approximate penetration depth of 
radiant energy in the ocean that ranges from 0 m for a 
starlit night to 25 m for a night with a full moon, 100 
m for a dark, cloudy day, and 200 m for a bright day, 
as estimated from the optical properties of seawater 
reported for the Alaska Gyre (Sasaki et al., 2001). Tags 
perform a daily test of the irradiance sensor through a 
built-in LED light source, and any faults are reported. 
For short-term diagnostic monitoring of postoperative 
recovery, LHX tags were programmed to transmit tag 
temperatures every 15-60 min for periods ranging up 
to 1-2 weeks after implantation. When the animals 
were not submerged in the saltwater pools of the re- 
covery facility, these data were received with a nearby 
(<20 m) handheld uplink receiver. 
The Argos system provides approximate locations 
of uplinking tags with accuracy indicated within a 
range from 0.49 km to 4.18 km (Soma and Tsutsumi, 
1986; Costa et ah, 2010). However, LHX tags commence 
transmissions after extrusion only at local noon. This 
timing may result in the estimates of the first posi- 
tion occurring up to 24 h post mortem for immediate 
extrusion events or after longer periods for delayed ex- 
trusions. We previously have estimated event location 
accuracy at approximately 10 km (Horning and Mell- 
ish, 2009). 
Inferring causa mortis from temperatures 
The temperature, irradiance, and immersion data al- 
low the classification of detected mortality events as 
1) death at sea by massive trauma likely due to preda- 
tion or as 2) nonpredation death (Horning and Mellish, 
2009). The normal abdominal temperature of healthy 
pinnipeds — although not well studied — is thought to 
range from 36°C to 38°C (Whittow, 1987). Antemortem 
temperature patterns may reveal hypothermia or py- 
rexia (fever), the former indicative of chronic energetic 
