sentinel tags & housing

Skip to first unread message

Yolanda Morbey

Mar 18, 2016, 11:19:50 AM3/18/16

Hi everyone,


Does anyone have a recommendation or best practices for placing sentinel nanotags in the environment for about 1 month? One option is to place tags uncovered on the ground (or hanging from a tree trunk or shrub), and without placing the tag in housing. Alternatively, we could cover the tag with a little roof.






Yolanda E. Morbey, Ph.D.

Associate Professor, Department of Biology

Chair, Undergraduate Program in Environmental Science

Western University, London, Ontario, N5A 6B7

519.661.2111 ext. 80116

Office: BGS 2074



Stuart Mackenzie

Mar 18, 2016, 11:52:35 AM3/18/16
to Yolanda Morbey,
Hi Yolanda,

One of the largest problems with 'sentinel', or test tags at fixed locations, is emulating tag function on living animals as they amplify the signal. This can be remedied somewhat by affixing the tag to a water balloon or water bottle roughly equal in mass to the species you're trying to emulate.

Weather-proofing the tags shouldn't be an issue as any kind of case or cover could influence the tag signal.

Tag Antenna orientation is another consideration as the antenna will never be facing a constant direction in birds at least, so some schedule of rotation, or have tags placed simultaneously with different orientations.

Happy to discuss further.



Stuart A. Mackenzie

Migration Programs Manager
Motus Wildlife Tracking System | Long Point Bird Observatory
Bird Studies Canada | Long Point Bird Observatory
PO Box 160, 115 Front St.
Port Rowan, ON. N0E 1M0
Office: 519-586-3531 x 162 | Mobile: 820-6040<><> |<>

You received this message because you are subscribed to the Google Groups "Motus Wildlife Tracking System" group.
To unsubscribe from this group and stop receiving emails from it, send an email to<>.
To post to this group, send email to<>.
Visit this group at
To view this discussion on the web visit<>.
For more options, visit

Jim Moore

Mar 18, 2016, 4:02:00 PM3/18/16
to Stuart Mackenzie, Yolanda Morbey,
Hi Stu-
I thought I would jump in here since this is an area of interest to me.  As you may recall from my email of July 14, 2015 I had expressed an interest in working with Willistown Conservation Trust on their Motus project.  I am planning on getting some field test signal strength data as a function of distance (LOS and not), altitude, and antenna orientation.  I have a drone (DJI P3) that I plan to use in these studies.  You had mentioned the amplification factor of living animals at that time but I would submit that a nanotag attached to an animal will attenuate the signal if anything. However, the attenuation due to water is minimal at 0.16 GHz where as at cell phone and UW oven frequencies of 2.5 GHz the attention is significant thus much of the RF energy is converted to heat.  This is probably why the nanotag does not use the cell phone frequencies even though the ~100 cm whip antenna would be more efficient. 
The signal strength could theoretically be increased by nearby conductive (metal) but that is not a likely scenario. Polarization (antenna orientation) and multi-path certainly do affect received signal strength and typically can be as much as +/- 15 dB
Here is a relevant discussion of water and microwaves.  Note that the dielectric loss at 2.45 GHz is significant compared to 0.3 GHz.  However, addition of salt to the water greatly increases the loss and inverts the relationship to frequency.  I am not sure how applicable this models a bird's body but the bird will absorb some fraction of the radiated energy from the nanotag but would not amplify it.
I am awaiting warmer weather to get started on the above mentioned field tests!
Jim Moore
34 Moore's Rd.
Elverson, PA
, ----- Original Message -----
To unsubscribe from this group and stop receiving emails from it, send an email to
To post to this group, send email to

David Bell

Mar 18, 2016, 4:27:05 PM3/18/16
to Jim Moore, Stuart Mackenzie,
Hi Jim,

I look forward to the results of the test! From experience in the field, however, I can say that being attached to a bird does seem to amplify the signal from a tag. We have had several birds lost to predators, or birds that somehow slipped out of their harnesses in the last few years, and have found that the detection distance is generally 100-200m on these tags when trying to refind them. If they are on the ground it can be as low as 50m but that is likely due to vegetation blocking the signal. When on the birds, detection distance increases to 1-1.5km using the same system at the same gain in the same habitat (5-el Yagi with a Lotek receiver in semi-open spruce/fir forest). Some of these tags have been at the top of 8m spruce trees, in a forest where the trees generally cap out at 8-9m (stunted coastal forest), so vegetation blocking the signal is not an issue in these cases as the birds are typically in more sheltered areas than this.

Just anecdotal evidence, I realize, but I have heard this from many people, for both birds and mammals.


For more options, visit

David Bell

Jim Moore

Mar 20, 2016, 11:34:53 AM3/20/16
to David Bell, Stuart Mackenzie,
Hi David-
Thanks for your response and field experience info.  What you observe does make sense and I think the issues are really semantic.  As a retired EE I associate amplification, in the case of rf energy, with an increase in power level by means of transistors (or even vacuum tubes) whereby DC power is converted to rf energy.  I would suggest that the effects you observe are due to reduction of attenuation of the rf signal.  If a tag is lying on the ground the rf energy is largely absorbed by the ground.  However, once it is elevated more than a wavelength the ground absorption largely disappears and the radiation pattern of the whip antenna on the tag approaches free space performance.  Thus when the tag is on a bird flying or perched in a tree the signal is much less attenuated than when the tag is on the ground.   
As a long time amateur radio enthusiast (W3ASA) I would expect my short-wave yagi antenna to be virtually useless if lying on the ground however elevate it 1/2 wavelength, or better yet a full wavelength (20m on the 14 MHz amateur band), and I can work the world!  If that doesn't work I can always amplify my 100 Watt transciever signal to over 1000 Watts with my vacuum tube amplifier.  Yes, hams still use vacuum tubes and in my case my 1000 watt amplifier uses a pair of Eimac 3-500Z tubes which gives my signal another 10 db of signal strength.
These same principles apply to nanotags (163 MHz, ~ 2m) just a difference in scale.

Phil Taylor

Mar 21, 2016, 10:28:38 AM3/21/16
There is a paper on this subject (showing a ~9 dB gain when on a bird)
so I believe it is a combination of what Jim (Ham Radio Jim) said
(attenuation of the signal) and the fact that there is some interplay
between the tag and the bird such that the bird acts as a meta-antenna.
I will drag out the reference and post to the list (at some point).

Cheers, Phil.
Phil Taylor
Bird Studies Canada Chair in Ornithology
Department of Biology, Acadia University
Wolfville, NS, Canada

(NOTE: All of my various email addresses end up in the same place)

Phil Taylor

Mar 21, 2016, 10:31:59 AM3/21/16
The other question is: what is this for? The answer to that question may
also help decide what exactly you should do.

Cheers, Phil.

Smith, Adam

Mar 21, 2016, 10:34:32 AM3/21/16
to Jim Moore, David Bell, Stuart Mackenzie, Motus Wildlife Tracking System
Hello all.

This excerpt from a 2009 IEEE Antennas and Propagation article by Martin et al. (attached) is relevant to the electrically small antennas characteristic of nanotags (and other VHF telemetry):

"While it is generally straightforward to measure the efficiency
of a generic antenna, measurement of the present radiating
system with an electrically small monopole antenna is challenging.
The transmitter itself is extremely small, on the order of 1cm3 or
less. Therefore, without the bird present, the monopole has a
very small counterpoise, at best; with the transmitter mounted onto
the bird, the radiating system becomes more complex. In fact, the 
lack of a large or well-defined ground plane in this case means that
both the body of the transmitter, as well as the bird, may serve as 
parts of the counterpoise for the monopole."

This manuscript also describes a bird model for evaluating antennas.

Other antenna and telemetry experts have provided additional suggestions for bird models.  

Beat Naef-Daenzer (Swiss Ornithological Institute) has used very simple models filled with Ringer-solution for medical use:

"For small "birds" the finger of a lab glove does very well, for larger 
models, I use plastic bottles. Qualitatively, the shape of these bodies
does not matter much, while the mass is really important. Transmitters 
can be fitted using the harness intended for attaching the radios to the 

I've also seen a suggestion from Kent Britain (, although it's not quite clear what volume of water to mix the sugar/salt into (I've sent a request for clarification):

"I suggest a small plastic bottle or plastic bag about 2/3rd the size of
the bird and fill it with water.  Add about 10 spoons of sugar and 1 spoon of
salt.  This will have the same Er and loss tangent as flesh.
Yes, it takes a while to get all that sugar to dissolve.
With a thick cloth to simulate the feathers, you can tune the antenna on
a simulated bird."



Adam Smith
Quantitative Ecologist
USFWS - National Wildlife Refuge System
Southeast (R4) Inventory & Monitoring Branch
135 Phoenix Rd.
Athens, GA 30605
Office: (706) 425-2250 (no voicemail)
Cell: (706) 201-1846

2009_Martin-Methodology for efficiency measurements for antennas on birds.pdf

Yolanda Morbey

Mar 21, 2016, 1:03:36 PM3/21/16
Hi Phil,

Our goal is to geoposition birds to the scale of several hectares or better. The sentinel tags will be there to assess and possibly control for environmental effects on signal quality, detectability, and position estimation. We had two stationary tags at Old Cut (lost) and these showed strange patterns with time of day (partly as a consequence of temperature effects). We didn't find out until the season was long over that these were lost tags. We didn't retrieve them and so don't know where they were and how the antennas were oriented. I don't know if environment effects will influence estimates of position (I hope they don't), but the more realistic our sentinel tags and the more we understand environment effects, the better our chances of accurate geopositioning.


Yolanda E. Morbey, Ph.D.
Associate Professor, Department of Biology
Chair, Undergraduate Program in Environmental Science
Western University, London, Ontario, N5A 6B7
519.661.2111 ext. 80116
Office: BGS 2074

-----Original Message-----
From: [] On Behalf Of Phil Taylor
Sent: Monday, March 21, 2016 10:32 AM
Subject: Re: [motus-wts] sentinel tags & housing

To unsubscribe from this group and stop receiving emails from it, send an email to
To post to this group, send email to
To view this discussion on the web visit

Phil Taylor

Mar 21, 2016, 2:06:04 PM3/21/16
Sounds like a good plan. Environment does influence signal strength, so
it will matter.

I think the best thing to do is to purchase 'gull' tags (that last 2-3
years) with coded ids of 1 or 2 (we never use those numbers on real tags
because they aren't reliable). Then I think you could just put them up
on a fixed platform, with the antenna pointing in a specific (known)
direction (either horizontally or pointed at a 45 degree angle downwards
to mimic the bird). As noted by Jim, you don't want the antenna touching
anything, and you don't want the tag affixed to or near any kind of
metal (both will influence the signal in some unknown way.

The ideal would be to have a bunch of these within the area you are
trying to 'georeference' and to pair at least some of them so that the
antennas were angled 90 degrees from one another (e.g. one pointing N
the other pointing E).

Cheers, Phil.

Kristin Jonasson

Mar 21, 2016, 2:52:34 PM3/21/16
Hi All,

I have one more thought to consider regarding the sentinel tags. The power of small batteries is often weaker in the cold. In my hibernating bats (~5C) signals often disappeared and became undetectable until the animal re-warmed to euthermic body temperatures a few weeks later. I'm not sure how warm radio-tags are when attached to a bird, but suspect they gain some heat once they are nestled in the body feathers. If a sentinel tag was around ~5C, I would expect it's signal to be reduced.

I'm sure it would be a pain to warm these tags up, but perhaps it's something you can take into account or measure.

Best of luck,

Jim Moore

Mar 21, 2016, 3:19:53 PM3/21/16
to Yolanda Morbey,,
Hi Yolanda-
I am unfamiliar with your project so I have a few basic questions:  Are you using the SensorGnome system with multiple towers or some other type of receiver?  There are many systems for DFing if your are using triangulation to locate the tags including a handheld DF loop (produces a sharp null when pointed towards the signal).  Also when you say several hectares is that the precision you are looking for or is that the total area of study?.  David Winkler's Ponds project at Cornell comes to mind.  This is about 10 acres in extent with several hundred tree swallow nest boxes.  I do recall some work being done there by Flying Fox Technologies to geoposition rf tags.  However that project seems to have fizzled.
I picked up the following on your website::" project to relate protandry to individual-level movement of migratory birds"  Is this the project you are talking about?
----- Original Message -----
Reply all
Reply to author
0 new messages