A summary of our methods
Targeted catching will take place on the northern beaches of Roebuck Bay (to ensure that birds marked occur at places accessible to human observers) using cannon-nets. Catching will take place throughout the year, and will include the different age classes.
Each bird receives four colour-bands and one leg-flag as the combination. Four colours are used for colour-banding: lime green (L), yellow (Y), red (R) and blue (B). However 34 birds (17 Bar-tailed Godwits, 17 Great Knot) have been marked with white (W) as part of the combination. After staining problems with the white bands used in New Zealand we changed to Lime Green in NWA.
For the majority of marked birds the metal band occupies a position not used in the combination, and is not part of the combination. However the first 18 Bar-tailed Godwits banded for the project were banded with the metal as the upper band on the left tarsus.
Apart from those 52 birds mentioned above the scheme has two colour-bands on each tarsus. The flag can be on either tibia, or in any of three positions on the tarsus (above, below or between the colour-bands). There are therefore eight possible positions for the leg-flag (see figure 1). The leg-flag is yellow, retaining a NW Australian identity in the event of an incomplete combination being recorded while a bird is on migration outside the study area.
Blood sampling, storage and processing
Of each individually banded bird a 10-40 microlitre blood sample will be collected by a small puncture of the brachial vein and the collection of 1-5 drops of blood in a hematocrit capillary tube.
The blood samples will be preserved in 96% ethanol in small cryogenic vials and stored at -20°C in a normal deep-freeze before transport on dry ice to Royal Netherlands Institute for Sea Research (NIOZ), The Netherlands.
Genetic laboratory work
At NIOZ, DNA will be extracted from each of the samples, which will be used to sex each banded individual to enable assays of sex-specific mortality patterns (using the verified protocol published by Baker, A.J., Piersma, T. & Greenslade, A.D. 1999. Molecular versus phenotypic sexing in Red Knots. Condor 101: 887-893).
Blood and DNA will be stored for later, more advanced DNA assays related to population structure, the incidence of blood parasites, and specific markers of functionally relevant genetic traits.
Demographic data analysis
Efforts will be made to resight individuals at the high tide roosts and during the incoming and outgoing tides, and during low water at neap tides throughout the year with emphasis on the late August-October (returning) and late February-April (pre-departure) periods.
Apart from efforts by visitors of the Broome Bird Observatory (BBO), dedicated efforts will be made by experienced observers during three periods of the year:
- after arrival from the breeding grounds (late August-October)
- mid non-breeding season (December-January)
- before northward departure (late February-April).
This would enable us to determine season-specific mortality rates and will inform us about local movements. Re-sighting scans will also be conducted at Bush Point, Coconut Wells and 80 Mile Beach, areas known to be good habitat for varying numbers of shorebirds, 100′s at Coconut Wells, 1000′s at Bush Point and 10′s of 1000′s at 80 Mile Beach. Observations elsewhere along the flyway would help to build the bigger picture on the migration routes used and the timing of migration, but season-specific survival rate estimates would not depend on it.
The resighting of a colour-marked bird is equivalent to recapture in analysis packages, but has a recovery rate 20-50 times greater. Survival and population size estimates will be made using mark-recapture methods, which have undergone rapid developments over the past ten years.
Analysis will predominantly use the mark-recapture program MARK. This programme is flexible, versatile, user-friendly and incorporates the most recent theoretical and computational developments in survival analyses. Additional data from past metal-ringing of the study species in the Broome region and at 80 Mile Beach (by the Australasian Wader Studies Group) will be built into models that combine the two sources of information of recaptures and resightings.
Reporting and publication
It is proposed that progress reports on the banding and resighting effort and results are published at ca. yearly intervals in the Stilt, the Tattler and/or the Newsletter of the Friends of BBO. Three years after the start of the project the first more formal analyses will be made on species, age and sex specific survival and resighting rates for publication in a more formal ornithological outlet. Hereafter, it is envisaged that the data will be used in more global analyses to be published in the most competitive refereed scientific journals. When the need arises, e.g. when drastic changes in annual survival are found, feedback will obviously be given to the relevant conservation and funding bodies.