Satellite tracking
of elephant seals

Southern elephant seals (Mirounga leonina) spend about ten months at sea, interrupted by two periods on land of around 3-4 weeks each: in November when they breed, and between January and March when they moult. The largest moult and breeding site is at South Georgia in the South Atlantic. In order to understand how they find and exploit food resources we need first to be able to describe their movements and dive behaviour. To achieve this we have designed and built a Satellite Relay Data Logger (SRDL).

SRDLs

While at sea, only ten percent of an elephant seal's time is spent at the surface. This imposes a severe restriction on the uplink rate. Thus data, rather than being sent in their raw form, must be processed and compressed. Data from the srdl depth and submergence sensors are used to determine the activity of the seal: either a 'dive' (deeper than 6 m for at least 6 s), a 'haulout' (dry for at least 240 s), or else at the 'surface'. Distance swum is determined by a turbine odometer. Individual dive records include information on maximum depth, depth profile, distance swum, and dive and surface duration. Dive and haulout records stored in memory are selected for transmission, such that those times of day when the Argos satellites are unavailable are adequately represented. The SRDLs are attached with a two-part rapid setting epoxy resin on the back of the neck just behind the head, so that the aerial emerges when the seal surfaces.

System Performance

Twelve seals were tagged on South Georgia between 1990 and 1994 and each produced an average of 119 days of data. Their tracks are shown in Fig. 1 (McConnell & Fedak, 1996). Females either travelled eastward, up to 3000 km away to the open Southern Ocean, or to the continental shelf on or near to the Antarctic Peninsula. Males either stayed close to South Georgia or used South Georgia as a base for shorter trips.

On average 16 uplinks were received per day. These provided detailed information on an average of 21 dives per day, approximately 35% of the seals' total number of dives. On average 4.4 locations were estimated per day. After passing through a location-filtering algorithm (McConnell et al., 1992) 3.5 locations per day remained. 87% of these were class zero locations.

Argos Enhancements

The current Argos system has provided us with a unique insight into seal behaviour, in spite of the limitations imposed by the narrow effective bandwidth. This limitation is the result of the minimal time spent by the animals at the surface, the satellite coverage and the acceptable transmission rate. To optimise the effectiveness of srdls, we would like to make each bit of information sent from the animal a useful one, receive as many bits as possible and know which bits have been received so as to minimise redundancy. This would allow biologists to reduce the energy cost of sending useful information, allowing srdls to be smaller or last longer.

The proposed enhancements to the Argos system will help to do this. Higher data transfer rates would allow more of the behavioural data gathered by the srdls to be relayed home. Improved satellite receiver sensitivity would help to keep energy costs down. More uplinks would be obtained providing more data and locations. Alternatively transmissions could be at lower power levels, allowing either smaller batteries, or increased battery life. Two-way handshaking between srdl and satellites (acknowledgement of the srdl message) would reduce the necessary redundancy of re-transmitting the same data, increasing the effective data transfer rate.

Finally downlink messaging would allow the srdl program to be modified in the light of its location and the animal's behaviour. For example, if the srdl appeared to malfunction, biologists could command it to send diagnostic information or reset some of its parameters. These could change program control functions to work around data collection problems or reset registers. Or, for example, if the seal were diving near a frontal system (important to this work because such oceanographic features may affect prey distribution) biologists could reset the srdl to relay detailed temperature/depth data.

References:

• McConnell, B.J., Chambers, C., and Fedak, M.A. (1992) Foraging ecology of southern elephant seals in relation to the bathymetry and productivity of the Southern Ocean. Antarctic Sci. 4(4): 393-398.

• McConnell, B.J., and Fedak, M.A. (1996) Movements of southern elephant seals. Can. J. Zool. 74: 1485-1496.