A trial of a non-statistical computer program for monitoring fuel reserves, response to wind and other details from GPS tracks of migrating geese

C. J. Pennycuick (1) , L. R. Griffin (2) , Kendrew Colhoun (3) , Richard Angwin (4)

1 School of Biological Sciences
2 Wildfowl and Wetlands Trust
3 Wildfowl and Wetlands Trust
4 Weather Desk, Broadcasting House

GPS transmitters were used with the Argos satellite system to track two Light-bellied Brent Geese () from Iceland to Arctic Canada, three Greenland White-fronted Geese () from Scotland to west Greenland, and two Barnacle Geese () from Scotland to Spitsbergen. Each goose's wing span was measured at the time of tagging, and its body mass and fat fraction were estimated at the time of departure. This was the starting point for the program's time-marching simulation, which is a non-statistical procedure based on flight mechanics. The ground speed was measured between each GPS fix and the previous one, and combined with a wind estimate to find the air speed. The program calculated the power, using the air density from the GPS altitude. The rates of consumption of fat and muscle tissue were calculated from the power, and used to update the body mass and composition. The total air distance flown by each bird was not consistently less than the ground distance, and there was no indication that the birds could select their routes, or even their timing, so as to bias the wind in their favour. Rates of climb when ascending the Greenland Ice Cap were very low in the Brent Geese, in relation to the maximum rate of climb of which the flight muscles were expected (on mechanical grounds) to be capable of supporting. The Brent Geese stopped often during the ice crossing, suggesting that they could not sustain sufficient aerobic power for continuous flight. The fat fractions of the White-fronted Geese were lower across the ice cap, and they climbed faster and stopped less often. Energy height was used as a species-independent measure of energy reserves. All seven northbound geese arrived in their nesting areas at estimated energy heights exceeding 200 km (around 350 km for the two Brent Geese). All of the geese achieved average energy gradients of at least 11, including short stops, meaning that their arrival energy heights were sufficient to fly a further 2,200-4,000 km. We propose that these reserves represent the energy height needed to initiate successful breeding in these arctic habitats, with an element of insurance against contrary winds.

Mots clés: Bird migration; Computer simulation; Flight mechanics; Energy height

C. J. Pennycuick, L. R. Griffin, Kendrew Colhoun, Richard Angwin. A trial of a non-statistical computer program for monitoring fuel reserves, response to wind and other details from GPS tracks of migrating geese. Journal für Ornithologie = Journal of Ornithology, 2010, 152 (S1), pp.87-99. ⟨10.1007/s10336-010-0633-6⟩. ⟨hal-00652645⟩