Evaluating consequences of stressors on vital rates in marine mammals is of considerable interest to scientific and regulatory bodies. Many of these species face numerous anthropogenic and environmental disturbances. Despite its importance as a critical form of mortality, little is known about disease progression in air-breathing marine megafauna at sea.  We examined the movement, diving, foraging behaviour, and physiological state of an adult female northern elephant seal (Mirounga angustirostris) who suffered from an infection while at sea. Comparing her to healthy individuals, we identified abnormal behavioural patterns from high-resolution biologging instruments that are likely indicators of diseased and deteriorating condition. We observed continuous extended (3–30 min) surface intervals coinciding with almost no foraging attempts (jaw motion) during two weeks of acute illness early in her post-breeding foraging trip. Elephant seals typically spend ~2 min at the surface. There were less frequent but highly extended (30–200 min) surface periods across the remainder of the trip. Dive duration declined throughout the trip rather than increasing. This seal returned in the poorest body condition recorded for an adult female elephant seal (18.3% adipose tissue; post-breeding trip average is 30.4%). She was immunocompromised at the end of her foraging trip and has not been seen since that moulting season. The timing and severity of the illness, which began during the end of the energy-intensive lactation fast, forced this animal over a tipping point from which she could not recover. Additional physiological constraints to foraging, including thermoregulation and oxygen consumption, likely exacerbated her already poor condition. These findings improve our understanding of illness in free-ranging air-breathing marine megafauna, demonstrate the vulnerability of individuals at critical points in their life-history, highlight the importance of considering individual health when interpreting biologging data, and could help differentiate between malnutrition and other causes of at-sea mortality from transmitted data.

Adult female diving and foraging behaviour was measured using biologging instruments deployed on ~20 individuals per year during the post-breeding trip from 2004–2020 for a total of 289 deployments. Each animal was sedated following standard protocols (Robinson et al., 2012) and equipped with a time-depth recorder (TDR) programmed to collect depth data at least every 8 seconds and with a satellite tag providing either Argos or GPS locations (Wildlife Computers, Seattle, WA, USA; Sea Mammal Research Unit, St. Andrews University, UK). Upon returning to shore, individuals were sedated again for instrument recoveries. 

Body composition and energy gain values were calculated using established methods (Robinson et al., 2012). Morphometric measurements were collected during deployment and recovery sedations, including weight, blubber depths, lengths, and girths. Body composition was calculated using the truncated cones method (Gales and Burton, 1987) and calibrated to body water measurements (Webb et al., 1998). Blubber thickness was measured at six evenly spaced locations along the length of the body along a dorsal line and again along a lateral line (12 measurements total) using a backfat ultrasound meter (Scanoprobe, Ithaca, NY, USA) at deployment and a Signos handheld ultrasound (Signostics, Thebarton, AUS) at recovery. Both instruments have previously been used to measure blubber thickness in elephant seals and produce comparable data. Mass was measured each time animals were sedated by suspending the seal from a hanging scale using a tripod and sling. To determine mass upon arrival and departure from the colony, measured masses were corrected to account for time spent onshore fasting before (Simmons et al., 2010) and after measurements were taken.

The surface area to volume ratio (SA:V) was calculated for post-breeding seals in 2017 using morphometrics measured at deployment and recovery. Each seal was represented geometrically by seven circular truncated cones from the ankle to the ears (divided at the same six locations along the length of the body where blubber thickness was measured) and a normal circular cone from the ears to the nose for the head of the seal. This method was chosen as the least likely to over- or underestimate the dimensions of the seal. Surface area and volume were calculated as the sum of the lateral surface areas and volumes, respectively, of each of these geometric shapes. For comparison to our best estimate methodology, a minimum and maximum SA:V were also calculated using differing geometric representations of the seal. The minimum estimate was calculated without the normal cone representing the head of the seal (only truncated cones from ankles to ears). The maximum estimate was calculated with the addition of normal circular cones for both the head (ears to nose) and tail (ankles to tail tip) to the truncated cones from ankles to ears.

The body composition and morphometric data provided are in CSV format.  Files included:

• DiseaseAtSea_ForagingSuccess_PB2004-2020.csv - includes all foraging success metrics for post-breeding adult female elephant seals from 2004 to 2020 included in this paper (Mass Gain, Mass Gain Rate, % Mass Gain, Energy Gain, Energy Gain Rate, Departure %Adipose, Arrival %Adipose, Trip Duration; Table 2, Fig. 1).
• DiseaseAtSea_LengthMass_PB2013-2020.csv - includes the standard length and mass, measured in the field at deployment for all post-breeding females from 2013-2020. These data were used to produce S1. 
• DiseaseAtSea_AllMorphs_PB2017.csv - includes all morphometric measurements collected at deployment and recovery for post-breeding 2017 animals.  These metrics were used to calculate surface-area-to-volume ratios.
• DiseaseAtSea_PB2017_SAV_V3.csv - contains the surface area and volume calculations completed using the script "DiseaseAtSea_SA_V_calc.R" and the data from DiseaseAtSea_AllMorphs_PB2017.csv.  These values were used to produce Fig. 8.

All code is available at  https://doi.org/10.5281/zenodo.7864665