Research Papers
GPS Collar Sampling Frequency: Effects on Measures of Resource Use

https://doi.org/10.2111/07-044.1Get rights and content

Abstract

A challenge in animal behavior studies using Global Positioning System (GPS) collars is selecting a sampling frequency to accomplish desired goals. High data resolution (i.e., frequent sampling) is appealing, because it maximizes behavioral information garnered. Extended sampling might be needed, however, to describe long term behaviors or seasonal dynamics. Because tradeoffs exist between high data resolution and sampling duration, we evaluated the effects of variable GPS sampling intervals on proportions of pastures used by cattle and distance traveled per day. This was accomplished with GPS collars configured to record cattle positions every 5 min for 15 d in three 829–864-ha pastures. Data were iteratively reduced to simulate increasingly longer GPS recording intervals from once every 10 min up to once daily. Two techniques were used to measure the percentage of pastures accessed by cattle. The first counted only pixels containing GPS coordinates. The second counted pixels containing coordinates and/or traversed by lines between vertices. Expansion of GPS recording intervals decreased (P < 0.01) estimates of the proportion of pastures visited by cattle with rates of decline best fit by exponential decay functions for both line and point techniques (R2  =  0.93 and 0.97, respectively). Spatial errors accompanying less frequent sampling intervals, however, were extremely large with the line technique and misrepresented areas visited by cattle. Expansion of GPS sampling intervals decreased (P < 0.001) distance traveled per day by cattle about 10% with each iteration. If travel corridors or accurate assessments of resources accessed are of critical concern, then longer GPS integration intervals should be avoided because they propagate flawed spatial interpretations. Similarly, if accurate measures of travel distances are critical, we suggest using a relatively frequent GPS recording interval.

Resumen

La selección adecuada de la frecuencia de muestreo para lograr las metas deseadas es un reto en los estudios de comportamiento animal cuando se usan collares del sistema global de posición (GPS). Generalmente se busca obtener un número extenso de datos (es decir muestreo frecuente) porque proporciona información más acertada sobre comportamiento. Un muestreo prolongado puede ser necesario dado que el GPS describe el comportamiento a largo plazo o las dinámicas estaciónales. Debido a que existe una relación entre la alta resolución de los datos y la duración del muestreo, se evaluó el efecto del intervalo de muestreos variables del GPS en proporciones de potreros usados por el ganado y la distancia caminada por día. Con este fin, se utilizaron los collares del GPS configurados para registrar la posición del ganado cada 5 minutos durante 15 días en tres potreros de 829–864 ha. La información fue interactivamente reducida para simular intervalos cada vez más largos en la grabación del GPS usando intervalos desde 10 minutos hasta una vez al día. Dos técnicas fueron utilizadas para medir el porcentaje del potrero al cual el ganado tuvo acceso. El primer conteo incluyó únicamente pixeles que contienen coordenadas del GPS. El segundo contó los pixeles que contenían coordenadas y/o líneas a través de los vértices. Al aumentar la extensión de los intervalos de la grabación del GPS se disminuyó (P < 0.01) la estimación de la proporción del potrero visitado por el ganado con tasas de reducción que se explicaron mejor usando tasas de reducción de la función exponencial por las técnicas de línea y de punto (R2  =  0.93 y 0.97, respectivamente). Los errores espaciales producidos por los intervalos de muestreo menos frecuentes, fueron extremadamente grandes con la técnica de línea y no representativas de las áreas visitadas por el ganado. La extensión de los intervalos de muestreo del GPS disminuyó (P < 0.001) cerca de 10% la distancia caminada por el ganado al día. Si el recorrido caminado o la acertada evaluación de los recursos evaluados son de preocupación crítica, entonces los intervalos más largos de la integración del GPS deben evitarse porque conducen a interpretaciones espaciales equivocadas. En contraste, si las medidas exactas de los recorridos son críticas, entonces sugerimos un intervalo relativamente frecuente de la grabación del GPS.

Section snippets

INTRODUCTION

Global Positioning System (GPS) technology is a relatively recent development (e.g., Rodgers and Anson 1994; Rodgers et al. 1996; Agouridis et al. 2004) for monitoring travel (Brosh et al. 2006), activity (Ungar et al. 2005), and resource use by medium- to large-sized animals (Mourão and Medri 2002; Bailey et al. 2006). GPS receivers in a lightweight collar or harness can be deployed for extended periods with little effect on behavior. Units derive coordinates from an internal receiver tracking

Study Site

Research occurred simultaneously in three pastures (829–864 ha) on the Northern Great Basin Experimental Range (lat 43°29′N, long 119°43′W [WGS-1984 datum]; elevation 1 400–1 674 m), 52 km west of Burns, Oregon. Mean annual temperature is 7.6°C with recorded extremes of -29°C and 42°C. Mean annual precipitation is 289 mm with about 60% being snow.

Vegetation is characterized by a sparse western juniper (Juniperus occidentalis subsp. occidentalis Hook.) overstory and a shrub layer dominated by

GPS Sampling Interval Effects on Percent of Pasture Visited

With a 5-min GPS schedule over a 15-d interval, cattle were found in 286.6 ±  pixels · pasture−1 with the point technique and 307.0 ±  pixels · pasture−1 with the line method. Pastures averaged 846.6 ±  pixels, so cattle visited approximately 33.8 ± 0.5% and 36.3 ± 0.3% of their pastures, respectively, with the point and line methods (Fig. 2).

As GPS sampling intervals were progressively expanded, estimates of the percent of pasture visited by cattle declined with both methods (Fig. 2). The rate of

Discussion

Decisions regarding the resolution and analyses of GPS collar data have profound consequences on the quantity and accuracy of information garnered in animal behavior studies. With expanded GPS sampling intervals, we progressively underestimated pasture use and travel by cattle. In all instances, rate of decline was best fit with exponential decay functions (Figs. 2 and 4).

When estimating the proportion of pastures visited by cattle as GPS recording intervals were expanded, measures declined

MANAGEMENT IMPLICATIONS

Although GPS collars are well proven tools for quantifying resource use, activity patterns, and travels of free-ranging animals, researchers must still address the quandary of memory and battery constraints specific to their instruments. An expansion of our most frequent recording interval of 5 min to 10 min could extend a study duration from 15 to 30 d without exhausting batteries. Misrepresentations of areas occupied would simultaneously increase from about 7% to 10%, respectively, an error

Literature Cited (13)

There are more references available in the full text version of this article.

Cited by (41)

  • Two dimensional searching paths exhibit fractal distribution that change with food availability (Normalized Difference Infrared Index, NDII)

    2022, Ecological Indicators
    Citation Excerpt :

    Long GPS sampling intervals may lead erroneously interpret the complexity of movements patterns (Codling and Plank, 2011; Plank and Codling, 2011). Optimal GPS sampling intervals for studying herbivore movement patterns range between 5 and 10 min (Johnson and Ganskopp, 2008). We chose 10 min intervals to be the minimal interval that allows recording a length sequence that satisfice our fractal analyses in 15 days intervals.

  • Low-Cost Livestock Global Positioning System Collar from Commercial Off-the-Shelf Parts

    2019, Rangeland Ecology and Management
    Citation Excerpt :

    Both Clark et al. (2006) and McGranahan et al. (2018) reported 95% CEP of GPS fixes taken at 1-s intervals. However, because precision of GPS measurements varies with fix interval length (Johnson and Ganskopp, 2008), we first recorded GPS readings at 1-s intervals and then performed a 24-h accuracy test at 5-min fix intervals. We calculated mean displacement and 95% CEP for both intervals.

  • Human vs robot: Comparing the viability and utility of autonomous underwater vehicles for the acoustic telemetry tracking of marine organisms

    2016, Journal of Experimental Marine Biology and Ecology
    Citation Excerpt :

    The difference between the UDs highlights one of the benefits of this technology, where the larger 95% UD of the AUV compared to the human tracker demonstrates that the AUV is likely better able to capture quick forays and movements that individuals make expanding the area an individual uses. Thus, the AUV more accurately reflects episodic animal movement and habitats that may be seldom used, while this information is lost due to aliasing with a human tracker (positions every 5–10 min) (Frair et al., 2004; Johnson and Ganskopp, 2008). The AUV also had a substantially smaller 50% core UD than a human tracker, suggesting that the increased temporal resolution of the AUV better reflect the areas and microhabitats where individuals spend the majority of their time (Andrews et al., 2011; Fieberg and Kochanny, 2005).

View all citing articles on Scopus

Eastern Oregon Agricultural Research Center, including the Burns and Union Stations, is jointly operated by the Oregon Agricultural Experiment Station of Oregon State University, Corvallis, Oregon and the United States Dept of Agriculture–Agricultural Research Service.

Mention of trade or proprietary names is for information only and does not convey endorsement of one product over another.

View full text