• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
Where premium quality meets exceptional value. ZEISS Conquest HDX.

Analysing bat recordings (1 Viewer)

Capercaillie71

Well-known member
I've recently started experimenting with recording from a frequency division bat detector.

Last night I got a nice clear recording, shown below with a peak frequency at 55kHz. I presume this is a Soprano Pipistrelle.

Shortly afterwards I recorded what I thought at the time was the same bat, but much closer. The calls were very loud but the recording iss less clear, with a big smear of frequencies centred on about 50kHz. Can I read much into this peak frequency or is it an artefact of the signal strength overloading and distorting the recording? I had the manual gain on my minidisc player set to maximum. Any advice on whether this second recording is any use would be welcome.
 

Attachments

  • Sop Pip 27-09-09.jpg
    Sop Pip 27-09-09.jpg
    121 KB · Views: 284
  • Pip 27-09-09.jpg
    Pip 27-09-09.jpg
    195.1 KB · Views: 287
Capercaillie,
if your signal is saturated you loose all the amplitude information from your signal which means that you cannot recover the frequency information - this is mathematically impossible. Overloading of signal can take place at several spots: the initial gain of the detector, the mic, or the "low frequency" recorder. This depends on your setup.

Your first figure is quite nice indeed. Your second figure shows a few notes that are not overloaded and in these cases you can readily read the frequency information. In the saturated regions you can recover timing information, e.g. what is the pulse interval. This can be important for IDing. A saturated signal shows an "signal" that is very broad in frequency just as seen in the center parts of your second figure.

regards
Harry J
 
Many thanks for this information - it is very helpful for analysing the sonograms.

I was out again last night. I had some problems again with signal saturation on the louder calls, but got a couple of clearer recordings. These have peak frequencies of 46.8kHz and 53 kHz respectively. I am interpreting these as Common and Soprano Pipistrelles - is this right?
 

Attachments

  • Pip 29-09-10.jpg
    Pip 29-09-10.jpg
    111.5 KB · Views: 246
  • Pip2 29-09-10.jpg
    Pip2 29-09-10.jpg
    117.2 KB · Views: 231
Many thanks for this information - it is very helpful for analysing the sonograms.

I was out again last night. I had some problems again with signal saturation on the louder calls, but got a couple of clearer recordings. These have peak frequencies of 46.8kHz and 53 kHz respectively. I am interpreting these as Common and Soprano Pipistrelles - is this right?

Hello I am new to the forum and although not a technical expert on the equipment I believe that as well as the pipistrellus species you also have a myotis (the second of the first recordings you pasted). The pipistrelle recordings will typically have a hook at the bottom and yes P.pipistrellus around 45KHz and 55 for Soprano.
 
Hello I am new to the forum and although not a technical expert on the equipment I believe that as well as the pipistrellus species you also have a myotis (the second of the first recordings you pasted). The pipistrelle recordings will typically have a hook at the bottom and yes P.pipistrellus around 45KHz and 55 for Soprano.

Welcome to the forum, and thanks for your comment. Can I ask what features of the second recording makes you think it is a Myotis?

In this part of the world we only have two regularly occurring Myotis species - Daubenton's and Natterer's. Natterer's is very scarce and hasn't been recorded closer than 40km to the site where I made the recording.
 
Hello I am new to the forum and although not a technical expert on the equipment I believe that as well as the pipistrellus species you also have a myotis (the second of the first recordings you pasted). The pipistrelle recordings will typically have a hook at the bottom and yes P.pipistrellus around 45KHz and 55 for Soprano.

Welcome to the forum.

I'm finding this post fascinating. There is so much to learn on this subject.
 
Many thanks for this information - it is very helpful for analysing the sonograms.

I was out again last night. I had some problems again with signal saturation on the louder calls, but got a couple of clearer recordings. These have peak frequencies of 46.8kHz and 53 kHz respectively. I am interpreting these as Common and Soprano Pipistrelles - is this right?

This seems about right. Common pipistrelle (Pippip) has a max frequency around 45KHz and Soprano (Pippyg) around 55KHz. They have about 8 and 11 sounds/second respectively, so your id seems reasonable. These are sort of average values and have some variations. The sonograms do not look like myotis echolocation calls, which are short and usually extend a wide range in frequency.

Regards
Harry J
 
Hi

You may find the following of some use:

http://www.echoesecology.co.uk/documents/AnIntroductionToTheBatsofScotlandEdition1Dec2006_002.pdf

http://www.somersetbirder.co.uk/sonograms.htm


There is a booklet with sonograms of most UK bats:

RUSS, J. 1999. The Bats of Britain and Ireland. Echolocation calls, sound analysis, and species identification. Alana Books, Powys

It is out of print Im afraid, but to be published soon (same author):

http://www.alanaecology.com/acatalo...ntification_Using_Sound_Analysis.html#a030006

I guess the hold up is to add in Alcathoe's Bat! Recently John Altringham wrote a short note on Alcathoe's stating the sonogram is one of the more identifiable myotis.

If you are looking at doing at lot more sonogram analysis of bats Im sure the to be published book will be worth getting. When I first started sonogram analysis (with time expansion not frequency division) I found it very useful in providing a secondary opinion to the bats heard in real time in hetrodyne. For the last 5-6 years Ive not had access to a time expansion detector (and havent had a go with sonograms on my frequency division detector (Duet)), so im a bit rusty on identifications in sonograms, and they probably have come on for Myotis anyway.

Anyway, you are basically looking for the following in a sonogram (I assume there is no difference between time expansion and freq. devision):

- Frequency modulation (FM) - the sweep through frequency providing the vertical line. The more vertical line (i.e. steep FM) provides a tick on the detector, a line at a gradient (i.e. shallow FM) provides more of a tock on the detector.

- Constant Frequency (CF) - a sweep along time at the same frequency (eg Horseshoes)

- length of call (potentially better measured from the oscillogram - bit at the top)

- length of call interval (time between each call - based on bat size and wingbeat)

- highest frequency point

- lowest frequency point

- point of maximum energy of the call (best obtained from power spectrum if your analysis software gives this)

with all these you can look at narrowing your sonogram to speices

Pipistrelles usually provide a sonogram with a Steep FM followed by a short CF (the 'slap' noise on a detector when you hit the right frequency). This provides a characteristic inverted tick shape on the sonogram. Yours dont show this very well (perhaps because you are using frequency division?) but shows the maximum point of energy well at (more or less) the right point for each pip, and it is kind of a teardrop.

Dont worry if its a 2-3khz out, this has been explained by the doppler effect (sound change due to bat moving), but maybe also the bat is peaking slightly differently to standard.

This time of year bats are mating and on territory so also look out for social calls (c30khz range) as they are most often used at this time of year.

Finally, be aware that nothing is absolute, most bats can produce calls with no CF element, especially in very enclosed environments like woodland. The vertical FM sweep provides a much more detailed picture of the surrounds, handy if you need to navigate blindly in an enclosed space. And just the opposite, pipistrelles for example can produce just CF calls in very open environments (they dont need the detail in the environment, just a look at whats broadly going on). This is detectable on a hetrodyne too, ever noticed that the sound of a pip is much more rich in very open environments?

If all that is of interest, then Im sure the Jon Russ book will be a good investment (I havent seen what it looks like though, but do have the old booklet!).

Cheers

edit: whilst correcting the dodgy urls above I came across this, if you cant wait for the new book, the old one has been re-issued on cd

http://www.alanaecology.com/acatalog/The_Bats_of_Britain_and_Ireland__PDF_.html
 
Last edited:
"I'm surprised no-one else on birdforum seems to be lusting after a Bushnell Trophy Cam after watching this program! They cost around £200."

From your other post re Land of the Lost Tiger would these be able to pick up bats if in sufficient numbers .
Brian
 
It looks like three different bat species to me in the four recordings, the first recording is a soprano pip the slight tailing to the peak which is shown much better in the third sonogram indicated pip 55hz indicates sop pip. The fourth one is common pip same reasons but peak frequency at 45.
The second sonogram is as already indicated by a previous poster a myotis species they have a very fast repitition rate and a large range of frequencies covered indicate it is probably a Natterer's or whiskered, the frequency lines are very straight in most myotis species and particularly Natterer's.

Hope this helps

Mark
 
Warning! This thread is more than 14 years ago old.
It's likely that no further discussion is required, in which case we recommend starting a new thread. If however you feel your response is required you can still do so.

Users who are viewing this thread

Back
Top