This archive contains all currently available satellite GPS telemetry data for Egyptian Geese in southern Africa over the period from 2008 to 2016. The data were collected with two primary aims: (1) to understand the movement ecology of this species; (2) to better evaluate the potential role of Egyptian Geese in spreading avian influenza in southern Africa. Data colelction was undertaken in several phases. The first phase focused on just three sites (Manyame, Barberspan, Strandfontein) and occurred at the same time as a series of extensive bird counts and captures. Birds captured during this period were tested for avian influenza. We also undertook an extensive colour-ringing exercise on Egyptian Geese during the first phase. The second phase of the program involved extending our activities to some new locations (Voelvlei, Jozini) to test specific hypotheses about movement and the timing of moult, and to improve the generality of our findings. The third phase involved a translocation experiment in which six birds were moved from Barberspan to Strandfontein. The data have been analysed and published in a number of different venues and publications, as listed in the associated metadata.

We tagged birds with satellite GPS PTTs (Microwave Telemetry) at three different latitudes in areas located along a north-south gradient running from Lake Manyame in Zimbabwe (-17.846S, 30.601E; near Harare) through Barberspan (-26.590S, 25.589E; NW Province of South Africa) to Strandfontein (-34.086S,18.516E; on the edge of False Bay, near Cape Town). These data were later complemented with smaller numbers of tagged birds at Voelvlei (-33.337S, 19.036; Western Cape, north of Cape Town) and Jozini Dam (-27.417S, 31.967E; Kwazulu-Natal, West of St. Lucia). The tagging locations span 17 degrees of latitude (approximately -17 to -34 degrees S) and the ends of the transect are situated approximately 2,200 kilometres apart. Lake Manyame and Jozini Dam fall in a typical summer rainfall region; Barberspan receives variable but dominantly summer rainfall; and Strandfontein and Voelvlei are in South Africa’s winter rainfall region. 

Birds were captured at each study site using walk-in traps, mist nets, and a cannon net. We tagged individual birds from different traps and typically spread captures over several days to ensure that two birds from the same flock were not tagged. We deliberately selected known wing feather moulting sites for the study and tried where possible to place transmitters on adult birds that had recently moulted but had not completely regrown their wing feathers, for three main reasons: (1) so that we would know definitively that the bird was tagged at a moulting site; (2) birds that moult in a given location are more likely to be part of a local population, and hence to behave typically for that region, rather then being transient, lost, or vagrant individuals; and (3) since these birds do not moult their back feathers until the following year, tagging recently moulted birds extends the time until potential regrowth of trimmed back feathers over the solar panel of the PTT. 

Transmitters were attached to birds using a backpack harness made out of teflon ribbon, as described by Cumming & Ndlovu. We used 30g transmitters for Egyptian Geese (mean adult weight of tracked birds 2299g, standard deviation 432.46g). Transmitters were set to capture GPS location data every two hours (30g PTTs) and transmit data to the Argos satellite every 3 days. Six of the birds in the dataset were captured at Barberspan and translocated to Strandfontein before release, as described in the Diversity & Distributions paper to which this archive is linked. This experiment was designed to shed light on the navigation capacity of Egyptian Geese.

The data given here are unprocessed apart from conversion from the Argos format to a more readable format, using Microwave Telemetry's conversion software; and splitting any files from the same transmitter where they were collected from different birds. The associated R code on Zenodo repeats the analyses given in the Diversity & Distributions paper.

The data are high-quality GPS fixes with a typical accuracy of 10-15m. In several cases there are gaps in transmissions as a result of transmitters not recording for periods of up to several months, then resuming (we assume, due to feather covering up solar panels and/or birds moving into conditions where solar panels were more exposed). Note that the data set also includes several cases in which the PTT was removed from one bird and placed on another. We have divided these into separate data sets as explained in the relevant papers.