2010 SOUTHEASTERN NATURALIST 9(3):595–604
Avian Radio-transmitter Harness Wear and Failure
Garth Herring1,2,* and Dale E. Gawlik1
Abstract - Although investigators have compared radio-transmitter attachment
devices and their likelihood of failure before the end of a study, few have directly
evaluated the harness materials and fastening methods that are to be shed by a bird
after the research period is over. We compared the likelihood of effective detachment
after transmitter life of four harness materials (7-mm- and 9-mm-wide polyester ribbon
tape and polyester-coated rubber elastic) and three fastening methods (polyester
thread, cotton thread, and Gorilla Super Glue™) using dummy transmitters exposed
to the elements for >1.5 years. Both polyester ribbon and polyester-coated rubber
elastic materials resulted in similar physical wear and remained intact for longer
than a typical field seasonal, but fastening harnesses using Gorilla Super Glue™
resulted in the earliest and most consistent harness failure. Polyester ribbon material
and glue fastening resulted in the earliest failure; mean failure time for 7-mm- and
9-mm-wide polyester ribbon tape with glue fastening was 408 days ± 30 SE, and 249
days ± 29 SE, respectively. Failure times for both 7-mm- and 9-mm-wide polyestercoated
rubber elastic and Gorilla Super Glue™ fastening treatments were in excess
of one year (438 days ± 14 SE and 438 days ± 13 SE, respectively). All harnesses
with sewn thread fastening treatments lasted a minimum mean of 456 days, and in
the case of both 7-mm-wide polyester ribbon and polyester-coated rubber elastic,
neither treatment ever failed over the period of study. Results suggest that using
Gorilla Super Glue™ as a fastener maximized the likelihood of eventual harness
failure, whereas transmitters fastened via sewing showed minimal signs of wear and
were unlikely to be shed by a bird during a period of time less than two years. Additional
experimental studies are warranted to examine alternative harness material
types, fastening methods, and harness styles to maximize the potential of successful
radio transmitter shedding.
Introduction
Radio and satellite telemetry studies are commonly used to better understand
behavior, space use, migration routes, and survival of birds. Important
assumptions for studies using radio or satellite transmitters (hereafter
transmitters) are that the transmitters do not adversely affect the behavior
(Mong and Sandercock 2007), reproductive success (Whidden et al. 2007),
or survival (Palmer and Wellendorf 2007) of the study species. To test these
assumptions, investigators have examined the use of different attachment
methods (see Hill et al. 1999, Johnson et al. 1991, Woolnough et al. 2004).
However, the intent of these studies was to simply ensure that transmitters
1Florida Atlantic University, Department of Biological Sciences, 777 Glades Road,
Boca Raton, Florida 33431. 2Current address - US Geological Survey, Western Ecological
Research Center, Davis Field Station, University of California - Davis, One
Shields Avenue, Davis, CA 95616. *Corresponding author - gherring@ucdavis.edu.
596 Southeastern Naturalist Vol. 9, No. 3
were not prematurely lost and that transmitters and harnesses meet assumptions
of behavior, reproductive success, and survival.
The Ornithological Council, a collective of eleven Western Hemisphere
ornithological societies, developed guidelines for use of wild birds in
research, including the ethical use of radio transmitters (Ornithological
Council 1999). The intent of these guidelines is to minimize distress and
pain to birds, consistent with sound research designs. The guidelines recommend
that investigators consider alternatives to allow self-removal or
failure at the end of the useful life of transmitters (Ornithological Council
1999). As a result, many investigators have included a weak point in their
harnesses, usually cotton thread, assuming that transmitters can be shed after
studies are completed (see Bedrosian and Craighead 2007, Chaves-Campos
et al. 2003, Karl and Clout 1987, McIntyre et al. 2006). Dissolvable surgical
thread has also been used as a harness material that will eventually fail, allowing
transmitters to be shed. Doerr and Doerr (2002) examined transmitter
retention time using dissolvable (surgical thread) harnesses and found that
19% of radio-tagged treecreepers harness loops prematurely broke and shed
transmitters before harnesses failed. Of the remaining harnesses that were
not shed prematurely, 85% remained intact for 149 ± 89 (SD) days (n = 13)
before shedding (Doerr and Doerr 2002). Thus, the attachment devices were
effective in that most transmitters remained attached for the duration of their
study but not much longer.
However, dissolvable surgical thread is not used regularly in attachment
of radio transmitters, perhaps because it is too small in diameter for use in
harnesses for larger birds or because of concern of premature wear of the
thread and loss of radio transmitters. Further, use of dissolvable harness
material (e.g., surgical thread) may be inappropriate for transmitters that
function longer than five months (Doerr and Doerr 2002) and for use with
waterbirds that occupy aquatic habitats, where harnesses might dissolve
rapidly. As a result, the assumption that radio transmitters are not permanently
attached, but instead are either lost or weaken sufficiently to allow
birds to remove transmitters remains untested for the most common attachment
materials.
In this study, we tested durability and likelihood of eventual failure of
four types of harness materials and three fastening treatments using the
common figure-eight harness design (Rappole and Tipton 1991). Over a
period of 547 days, we recorded physical wear and tested the likelihood of
harness failure using a tensile resistance test. We tested the hypothesis that
narrower harness materials and those fastened with natural cotton thread
would fail sooner than other harness material and fastening treatments.
This study provides recommendations for harness materials and fastening
methods to maximize the likelihood that transmitters will be shed after
studies are completed.
2010 G. Herring and D.E. Gawlik 597
Methods
Testing transmitters on wild birds provides the most realistic assessment
of wear (Powell et al. 1998). However, such studies require that birds be
repeatedly recaptured to assess harness fatigue, which is unattainable for
many species. While studies of captive birds (Small et al. 2004, Woolnough
et al. 2004) are more feasible, it is usually difficult to obtain adequate sample
sizes. Thus, in this study we used dummy transmitters attached to dummy
bird bodies exposed to the elements for 547 days.
We made radio-transmitter harnesses using a two-part epoxy resin
and hardener kit to create a replica of a radio transmitter of standard size
and shape (Model A1260; Advanced Telemetry Systems, Isanti, MN). We
drilled holes in either end of the dummy transmitter through which the harness
material fastened (Fig. 1). Dummy transmitters weighed 10.1 ± 0.2
(SE) grams.
We tested four harness materials: 1) 7-mm-wide (hereafter narrow)
polyester-coated rubber elastic (73% polyester, 27% rubber), 2) 9-mm-wide
(hereafter wide) polyester-coated rubber elastic (73% polyester, 27% rubber),
3) narrow 100% polyester ribbon tape, and 4) wide 100% polyester
ribbon tape. These materials have either been used in previous studies (Bedrosian
and Craighead 2007, Buehler et al. 1995, Chaves-Campos et al. 2003,
Hylton et al. 2006, Small et al. 2004, Weick et al. 2005) or in other concurrent
research associated with this project (Beerens 2008, Herring et al. 2010)
and are readily available at most fabric or craft stores.
We cut material to a standard length of 40 cm, sufficient for a harness to
fit on a Eudocimus albus (L.) (White Ibis; Herring 2008). We attached harness
materials to the dummy radio transmitter by threading them through
the drilled holes (Fig. 1) and then tying a knot on either side of the dummy
transmitter to prevent movement on the harness (Fig. 1). We mounted harnesses
on a closed-cell foam mold in the approximate shape of a bird’s body,
with the fastening point of the harness on the dorsal side of the dummy. After
tightening the harness, the loose end of the harness material was then fastened
using one of three treatments: 1) Gorilla Super Glue™ (hereafter super
glue) (Gorilla Glue, Cincinnati, OH), 2) 100% cotton mercerized thread, or
3) 100% polyester thread. For all harnesses, we left 1 cm of loose material at
each end of the harness for the fastening treatment (Fig. 1). When applying
the super glue treatment, we spread the glue liberally over one end of the harness
material and then clamped the adjacent end of the harness to the glued
section, allowing it to dry for approximately one minute before removing
the clamp (Fig. 1). For sewn-fastening treatments, we began sewing beside
the transmitter and moved away from it with each loop until eight tight
loops were completed, and then fastened in a knot at the end. During the experiment,
harness tension remained constant, with no apparent change in the
foam mold shape or size. We used a total of 120 dummy radio-transmitters
598 Southeastern Naturalist Vol. 9, No. 3
in the experiment, with 30 dummies per harness material and 10 dummies
per fastening treatment.
We placed all dummy radio transmitters with their respective harness
treatments on a frame in an outdoor area at Florida Atlantic University, Boca
Raton, fl; all dummy transmitters were equally exposed to rain and sun
during the experiment. Mean annual weather conditions at Boca Raton were:
annual precipitation = 145 cm, temperature = 24 ºC (range = 14–33 ºC),
and humidity = 72%. In this experiment, the degree of exposure was likely
greater than would be the case with wild or captive birds because birds often
preen feathers over harnesses (G. Herring, pers. observ.). Regardless of how
these conditions mimicked natural conditions, any biases were equal across
all treatments.
We surveyed avian telemetry literature published during 2007–2008
in the Journal of Field Ornithology, Journal of Wildlife Management,
and Waterbirds; the mean length was 136 ± 36 days (n = 36) for radiotransmitter
studies, and 286 ± 100 days (n = 7) for satellite-transmitter
studies. In this study, we monitored harness treatments for 547 days, a period
of time longer than most field studies where transmitters are used. We
sampled dummy transmitters approximately every three months, assessing
visual appearance of the transmitter and using a tensile test to mimic a
Figure 1. Dorsal view of dummy transmitters with figure-eight harness. Note the
fastening point on the left side of the dummy transmitter, and the knots tied on the
right side to prevent transmitter movement.
2010 G. Herring and D.E. Gawlik 599
bird grabbing and pulling the harness with either its bill or foot. To assess
visual wear, we used four scores: 0 = no wear (no apparent visual signs of
wear), 1 = little wear (sewn harnesses: only a few threads were loose and
or minimum thread fraying; glued harnesses: <33% of the glued portion
of the harness was disconnected; sewn and glued harnesses: only a few
threads were loose and/or there was minimum thread fraying and < 3%
of the glued portion of the harness was disconnected), 2 = moderate wear
(sewn harnesses: multiple threads were loose and/or there was intermediate
thread fraying; glued harnesses: >33% and <66% of the glued portion of
the harness was disconnected; sewn and glued harnesses: multiple threads
were loose and/or there was intermediate thread fraying and >33% and
<66% of the glued portion of the harness was disconnected), and 3 = heavy
wear (sewn harnesses: many threads were loose and/or there was extensive
thread fraying; glued harnesses: ≥66% of the glued portion of the harness
was disconnected; sewn and glued harnesses: many threads were loose and/
or there was extensive thread fraying and ≥ 66% of the glued portion of the
harness was disconnected).
To assess physical wear, we attached a 1-kg Pesola spring scale to one
side of the posterior point where the two loose ends of the harness material
were fastened, applied 400 g of tension, and noted if the harness fastening
point failed. During this test, we pulled the harnesses in both lateral directions
once. We considered this tension to be a reasonable estimate of the
potential pulling strength of a typical 800-g bird that the harness material
and size was developed for use on. Although no estimates of the pulling
strength of birds could be located to verify the tension used, the same weight
was used across all treatments and thus provided the same opportunity for all
harnesses to fail across the entire 547-day period.
We used a Kaplan-Meier survival model to estimate harness failure
(1 - φ) for all combinations of harness material and fastening treatments
(Pollock et al. 1989). To determine if failure curves differed among harness
materials and fastening treatments across the 547 days of the study,
we used a log-rank test (Pollock et al. 1989). We used a repeated-measures
two-way analysis of variance (ANOVA) to test for differences in
physical wear on harnesses by material and fastening treatments. All data
in the ANOVA model met assumptions of homoscedascity (Levene’s test),
and residuals were normally distributed. We conducted all data analysis
using program JMP (Sall et al. 2001). Values are presented as means ±
standard error.
Results
Harness failure only occurred at the fastening point during the experiment.
We found that harness failure rates differed among treatments (χ2
11 =
149.2, P < 0.001), with wide ribbon material fastened with super glue fail600
Southeastern Naturalist Vol. 9, No. 3
ing the soonest (mean failure time = 248 days ± 29 SE), followed by narrow
ribbon fastened with super glue (mean failure time = 407 days ± 30 SE), and
both narrow and wide polyester-coated rubber elastic material fastened with
super glue (mean failure time = 437 days ± 14 SE and 437 ± 13 SE, respectively;
Fig. 2, Table 1). All other failure rates were similar among treatments
(Table 1, Fig. 2). Physical wear on radio-transmitter harnesses was similar
among material types (F3, 8 = 2.1, P = 0.17), but differed among fastening
treatments (F2, 8 = 7.5, P < 0.0001). Physical wear of harnesses was highest
for wide ribbon and super glue treatment, with moderate to low wear on all
other treatment combinations (Fig. 3).
Discussion
Harnesses fastened with either polyester or cotton thread did not fail
during the period of this study (>1.5 years), meaning that short-lived
Figure 2. Mean physical-wear response of radio-transmitter harness material and
fastening treatments to exposure to sun and rain during 547 days of experiment. Wear
response scores: 0 = no wear, 1 = little wear, 2 = moderate wear, and 3 = heavy wear.
Treatments: 1 = narrow ribbon and cotton thread, 2 = narrow ribbon and super glue,
3 = narrow ribbon and polyester thread, 4 = narrow polyester-coated rubber elastic and
cotton thread, 5 = narrow polyester-coated rubber elastic and super glue, 6 = narrow
polyester-coated rubber elastic and polyester thread, 7 = wide ribbon and cotton thread,
8 = wide ribbon and super glue, 9 = wide ribbon and polyester thread, 10 = wide polyester-
coated rubber elastic and cotton thread, 11 = wide polyester-coated rubber elastic
and super glue, and 12 = wide polyester-coated rubber elastic and polyester thread.
2010 G. Herring and D.E. Gawlik 601
birds with radios attached using thread could carry the transmitters for the
remainder of their lives. Thus, for smaller birds, using harnesses fastened
with dissolvable surgical thread may be more appropriate (Doerr and
Table 1. Mean time (± SE) to failure of radio-transmitter harness treatments as determined during
547 days of exposure to natural weather conditions. Treatments with no reported data did
not fail during this study. Treatments are listed as material width, material type, and fastening
type, respectively. All treatments started with 10 experimental harnesses.
# of harnesses Mean days
Treatment failed to failure SE
7-mm - ribbon- cotton thread 0
7-mm - ribbon - super glue 10 408 30
7-mm - ribbon - polyester thread 0
7-mm - polyester-coated rubber elastic - cotton thread 3 456 14
7-mm - polyester-coated rubber elastic - super glue 3 438 14
7-mm - polyester-coated rubber elastic - polyester thread 4 520 15
9-mm - ribbon - cotton thread 3 538 11
9-mm - ribbon - super glue 10 249 29
9-mm - ribbon - polyester thread 5 547 0
9-mm - polyester-coated rubber elastic - cotton thread 3 456 0
9-mm - polyester-coated rubber elastic - super glue 6 438 13
9-mm - polyester-coated rubber elastic - polyester thread 4 520 15
Figure 3. Probability of harness material and fastening treatments failing during 547
days of exposure. Treatments are listed as harness material first: P = polyester ribbon,
R = polyester-coated rubber elastic, then fastening treatment, C = cotton thread, G =
super glue, and P = polyester thread. All are labeled at the survival end point after
547 days of exposure.
602 Southeastern Naturalist Vol. 9, No. 3
Doerr 2002). However, high premature shedding rates (Doerr and Doerr
2002) and limited life span for long-term studies make surgical thread an
inferior option for large birds.
During concurrent radio-telemetry field research using narrow polyestercoated
rubber elastic harnesses and cotton sewn fastening (Beerens 2008,
Herring 2008), we recaptured or recovered carcasses from several harnessed
birds, observing no obvious signs of harness deterioration after one full year
of natural exposure and wear. Also, contrary to our hypothesis, cotton thread
did not fail sooner than polyester thread. If investigators want transmitters to
be shed soon after a typical telemetry study, they should consider polyester
ribbon fastened with super glue. We suggest caution in interpreting our estimates
of error associated with wear and failure because we only measured
transmitters approximately every three months. More frequent measurements
(e.g., every month) in future experimental radio-transmitter harness
studies may provide improved estimates of when harnesses fail and of the
precision associated with those failure times.
Collectively, these results suggest that harnesses sewn with cotton or
polyester thread do not provide a reliable option for telemetry studies if
the intent is to guarantee that transmitters are shed at the end of a study.
Only several of the most common transmitter attachment mechanisms
were evaluated here, so more attachment options should be tested to
better match attachment period of radio-transmitters with the duration of
a particular study and the specific environmental conditions under which
particular bird species live. Future studies should compare harness life
when harnesses are covered in feathers, how preening oil from the uropygial
gland affects harnesses and fastening treatments, and how weather
conditions (e.g., sun exposure, rainfall) impact wear and failure of harnesses.
We also suggest testing alternative fastening methods (e.g., water
soluble sewing thread) and harness types that increase the likelihood of
radio-transmitter shedding.
Acknowledgments
Funding was provided by the US Fish and Wildlife Service. T. Dean provided
support in implementing this study. E. Doerr, H. Herring, S. Lantz, D. Tsao, and two
anonymous reviewers provided valuable comments on previous drafts of this manuscript.
M. Royer helped construct the dummy transmitter mold, and T. Beck and N.
Hill assisted with data collection.
Literature Cited
Bedrosian, B., and D. Craighead. 2007. Evaluation of techniques for attaching transmitters
to Common Raven nestlings. Northwestern Naturalist 88:1–6
Beerens, J.M. 2008. Hierarchical resource selection and movements of two wading
bird species with divergent foraging strategies in the Everglades. M.Sc. Thesis.
Florida Atlantic University, Boca Raton, fl.
2010 G. Herring and D.E. Gawlik 603
Buehler, D.A., J.D. Fraser, M.R. Fuller, L.S. McAllister, and J.K.D. Seeger. 1995.
Captive and field-tested radio-transmitter attachments for Bald Eagles. Journal
of Field Ornithology 66:173–180.
Chavos-Campos, J., J.E. Arevalo and M. Araya. 2003. Altitudinal movements and
conservation of Bare-necked Umbrellabird, Cephalopterus glabricollis, of the
Tilaràn Mountains, Costa Rica. Bird Conservation 13:45–58.
Doerr, V.A.J., and E.D. Doerr. 2002. A dissolving leg harness for radio-transmitter
attachment in treecreepers. Corella 26:19–21.
Herring, G. 2008. Constraints of landscape-level prey availability on physiological
condition and productivity of White Ibises and Great Egrets in the Florida Everglades.
Ph.D. Dissertation, Florida Atlantic University, Boca Raton, fl.
Herring, G., D.E. Gawlick, M.I. Cook, and J.M. Beerens. 2010. Sensitivity of
Great Egret and White Ibis nesting to landscape prey availability. The Auk
127:660–670.
Hill, I.F., B.H. Creswell, and R.E. Kenward. 1999. Field-testing the suitability of a
new back-pack harness for radio-tagging passerines. Journal of Avian Biology
30:135–142.
Hylton, R.A., P.C. Frederick, T. De La Fuente, and M.G. Spalding. 2006. Consequences
of pre-fledgling health on the survival of juvenile Wood Storks, Mycteria
americana. Condor 108:97–106.
Johnson, G.D., J.L. Pebworth, and H.O. Krueger. 1991. Retention of radio-transmitters
attached to passerines using a glue-on technique. Journal of Field Ornithology
62:486–491.
Karl, B.J., and M.N. Clout. 1987. An improved radio-transmitter harness with a weak
link to prevent snagging. Journal of Field Ornithology 58:73–77.
McIntyre, C.L., M.W. Collopy, and M.S. Lindberg. 2006. Survival probability and
mortality of migratory juvenile Golden Eagles from interior Alaska. Journal of
Wildlife Management 70:717–722.
Mong, T.W., and B.R. Sandercock. 2007. Optimizing radio retention and minimizing
radio impacts in a field study of Upland Sandpipers. Journal of Wildlife Management
71:971–980.
Ornithological Council. 1999. Guidelines to the use of wild birds in research. Available
online at http://www.nmnh.si.edu/BIRDNET/GuideToUse/guidelines_use.
html. Accessed 10 August 2008.
Palmer, W.E., and S.D. Wellendorf. 2007. Effects of radio-transmitters on Northern
Bobwhite annual survival. Journal of Wildlife Management 71:1281–1287.
Pollock, K.H., S.R. Winterstein, C.M. Bunck, and P.D. Curtis. 1989. Survival analysis
in telemetry studies: The staggered-entry design. Journal of Wildlife Management
53:7–15.
Powell, L.A., D.G. Krementz, J.D. Long, and M.J. Conroy. 1998. Effects of
radio-transmitters on migrating Wood Thrushes. Journal of Field Ornithology
69:306–315.
Rappole, J.H., and A.R. Tipton. 1991. New harness design for attachment of radiotransmitters
to small passerines. Journal of Field Ornithology 62:335–337.
Sall, J., A. Lehman, and L. Creighton. 2001. JMP Start Statistics: A Guide to Statistics
and Data Analysis Using JMP and JMP IN Software, Second Edition. SAS
Institute, Cary, NC.
604 Southeastern Naturalist Vol. 9, No. 3
Small, M.F., R. Rosales, J.T. Baccus, F.W. Weckerly, D.N. Phalen, and J.A. Roberson.
2004. A comparison of effects of radio-transmitter attachment techniques on
captive White-winged Doves. Wildlife Society Bulletin 32:627–637.
Weick, S.E., M.G. Knutson, B.C. Knights, and B.C. Pember. 2005. A comparison
of internal and external radio-transmitters with Northern Leopard Frogs (Rana
pipiens). Herpetological Review 36:415–421.
Whidden, S.E., C.T. Williams, A.R. Breton, and C.L. Buck. 2007. Effects of transmitters
on the reproductive success of Tufted Puffins. Journal of Field Ornithology
78:206–211.
Woolnough, A.P., W.E. Kirkpatrick, T.J. Lowe, and K. Rose. 2004. Comparison of
three techniques for the attachment of radio-transmitters to European Starlings.
Journal of Field Ornithology 75:330–336.