Tracking birds around the globe is not easy. Colour-banded birds can be resighted at points along the migration route but these resightings, while very informative, are generally sparse and irregular. When detailed information is needed about when birds migrate, or how long they travel for, some direct measurement is required.

Bar-tailed Godwits in Alaska have been suggested to have a wind-sensitive migration strategy, leaving on migration when low-pressure systems in the North Pacific provide winds favourable for southward migration. Knowing the departure date and, hopefully, the arrival date, means that we can evaluate the winds encountered along the entire migratory route. This is an important step towards understanding how the trans-Pacific migration of godwits evolved, and assessing whether climate changes due to global warming might affect birds' ability to make long flights where stopping is not a possibility.

Three options used by biologists to look at movements are radio tracking, data loggers and satellite tracking. Radio-tracking generally gives the least information for large-scale movements, as a signal must be detected by a nearby receiver. If birds occur in limited habitats, however, where flocks can be scanned along the migration route, this approach can work very well. Data loggers are larger than radio tags, and collect and store data on activity or location (via GPS or light-levels). The unit must be retrieved or approached closely to download the data, so this technique works best for medium-large breeding birds. Satellite transmitters are much larger than radio-tags, but allow the location of an animal anywhere in the world to be determined. Two constraints on satellite tags are the mass of the unit and the cost for the tags and the download time, and until recently the only shorebird this technique has been attempted on was the Eastern Curlew.

In 2005, two attempts were made to track and time the migration of Bar-tailed Godwits from Alaska to New Zealand. As part of the Beringia Swedish-US research expedition, automatic radio-receivers provided by Martin Wikelski of Princeton University were set up on the staging grounds on the Yukon-Kuskokwim Delta and at six location around New Zealand. A previous reconnaissance expedition in September 2004 had shown for the first time that it was possible to catch fuelling godwits on the delta. The aim for 2005 was to radio-tag up to 60 godwits in Alaska and try to detect departure dates and maximum travel times to New Zealand, where more than 50% of the national godwit population would be covered by the receivers. Unfortunately, despite the best efforts of a highly skilled team of international ornithologists, fate conspired to foil us. A lack of birds, smoke clouds that restricted flying, gales, more gales and three floods prevented the field teams from succeeding in their quest. Consequently, this attempt failed dismally to gather the first direct evidence on the timing of trans-Pacific godwit migration.

The same year, Bob Gill, Brian McCaffery and team had surgically implanted subcutaneous satellite transmitters in a small number of breeding female godwits on the YK Delta. This method of deployment should avoid the serious drag problems that were probably associated with reduced migration completion in Eastern Curlews, though it had not been tried on gdowits before. All was looking good for ground-breaking results - after the breeding season the birds were moving around the YK Delta, one bird even visiting the Alaska Peninsula before returning. Then, as if in a bad movie, one by one the transmitters stopped working due to a technical fault. We know that it was failure of the transmitters rather than the birds, because several of the tagged birds were subsequently sighted on the non-breeding grounds after migration. 2005 was not destined to be the break-through year in godwit tracking.

2006 saw that work followed up with more PTT tagging of breeding birds, again with mixed results. Battery life issues meant that some transmitters failed at critical stages, but the data show clearly that godwits do cross the Pacific Ocean. Birds with failed transmitters were subsequently seen in New Zealand, confirming that they had made successful journeys.

With this history of trial, failure and improvement, it is a relief that tagging in 2007 is working a treat! In February 2007 a team led by Bob Gill, Nils Warnock from PRBO Conservation Science and me tagged 16 godwits in New Zealand with implanted or backpack satellite tags. Females (with the implants) did extremely well, flying direct to eastern Asia with flights of up to and just over 10,000 km. By the end of May, four females had settled onto breeding grounds in SW Alaska, one was in the air en route, and a male had ended up in eastern Russia. See the USGS website for their migration maps.