article / 9 March 2016

Comparing Geolocator and High-Precision GPS Data

Few recent technologies have been embraced by the ornithological community as rapidly as solar geolocation tracking devices. Although the first and rather large ‘geolocators’ became available more than two decades ago, with their recent miniaturization the number of tracking studies exploded.

Small passerines can now be followed throughout at least part of their annual cycle (Ouwehand et al. 2015, J. Avian Biol.) and larger birds tracked for multiple years (Senner et al. 2014, PLoS One). Literally hundreds of studies have now been published, leading to many novel insights not only into bird migration and distribution, but also foraging behaviour (Navarro et al. 2007, Oecologia), the division of parental labour among breeding pairs (Pinet et al. 2012, Anim. Behav.), and the accrual and dissipation of carry-over effects across an individual’s annual cycle (Conklin et al. 2013, PLoS One).

A challenge of using geolocators is that, unlike GPS, they do not provide the biologist with locations of the tagged animal; instead, researchers have to estimate these locations with a software tool. Although these tools were extensively used in bird studies, the underlying those methods were never ground-truthed on a migratory animal throughout its annual cycle. Our article ground-truths a geolocator by comparing the routine of one black-tailed godwit followed over a year with both a GPS tracker and geolocator.

Figure 1. Willem Bouten with one of the first godwits tagged with UvA-Bits trackers. Photo: T. Piersma

The genesis of the article was twofold: First, in the end of 2012 there was a meeting at the National Center for Ecological Analysis and Synthesis at the University of Santa Barbara in California focused specifically on developing new open-source analytical techniques for geolocators. This meeting brought together the developers of many of the programmes currently available for analysing geolocation data, as well as a number of geolocator users. After two weeks of intense discussion we realized that without a ground-truthed dataset, we could only guess at how well different methods performed. Second, in the spring of 2013, we began deploying UvA-Bits trackers (only then recently developed at the University of Amsterdam by Willem Bouten’s team, Fig. 1) on Black-tailed Godwits to study their pre-breeding movements in southwest Friesland. About halfway through deployment, Eldar convinced the godwit team to ‘proof’ a geolocator’s data by outfitting a bird with both a GPS tracker and a geolocator, thereby generating the data to fine-tune FLightR – an R package he has developed for solar geolocation analysis (Rakhimberdiev et al. 2015 Mov. Ecol). A year later we happily recaptured one of the double-tagged godwits on the nest. (Fig. 2).

Figure 2. Nathan Senner after the successful catch. Photo: Mo Verhoeven

Our first goal in publishing this paper was, of course, to show what FLightR can do as it is a new programme that we think represents a significant improvement in the analysis of geolocation data. However, our second goal was to also make public the underlying datasets. We recognize that FLightR is only a stepping-stone toward even more improved analyses of geolocation data. To aid that improvement, we want to ensure that everyone has access to the type of dataset necessary for creating and perfecting geolocator analytical techniques, as not everyone will have an opportunity to simultaneously deploy geolocators and GPS tracking devices. For that reason we have made the GPS and geolocator data  and the R scripts used for the analysis publically available as supplementary material. If you are interested in how geolocators perform on migratory birds, what they show and what they do not – we hope you enjoy the paper.

 

About the Authors

Nathan Senner is a Post-Doctoral Fellow in the Division of Biological Sciences at the University of Mantana, USA. He asks both basic and applied questions in an effort to identify the mechanisms that limit the flexibility of individuals to respond to current environmental changes and determine how adaptive management can buffer populations against future environmental changes.

Eldar Rakhimberdiev is a Post-Doctoral Fellow in the Department of Marine Ecology (MEED) at the NIOZ Royal Netherlands Institute for Sea Research. He is interested in fast evolution, the interplay between population ecology and evolution, bird migration and applied programming. 

This article first appeared in the blog of the Journal of Avian Biology and was republished with permission. 

The header image is published under a Creative Commons Attribution-ShareAlike 2.0 Generic (CC BY-SA 2.0) License. The image was not alterned in any way. The original image can be accessed here


Add the first post in this thread.

Want to share your own conservation tech experiences and expertise with our growing global community? Login or register to start posting!