Bats!

Bats

The bat project had its genesis a few years ago when the kids started getting interested in bats... The Introduction below gives a bit of background, and you can click on Project Status for a fairly up-to-date report.

Introduction

Our house is about 150 years old - quite new for the village we live in, where many buildings originated 400 years or more ago. The countryside around us has rivers and woodland, and is fairly open. Before the Second World War, most fields in the area were hop gardens (a field of hops is called a "garden"), but changes in farming practice, economics and drinkers switching from bitter beer to lager, has meant that very few traditional hop gardens remain in production.

Hops (or more correctly, the hop flowers) are picked in September and taken to oast houses for drying. When they are at the right level of dryness, they are packed into long sacks called "pockets" and taken to the brewery.

Now that there are far fewer hop gardens, many oast houses were converted into domestic dwellings. Ours was originally converted back in 1949, and in the years since we have owned it we have been slowly renovating and rebuilding it. Remember, these were farm buildings; they tend to have poor foundations, no damp-course, poor insulation etc. All-in-all, a long-term project!

We've always known about the bats and have been careful not to disturb them, but have always wondered what flavour they were. Eventually, I built a heterodyne bat detector. This is a simple electronic device that shifts the bats squeaks into the human audible domain. Most bats call between 125kHz and 20kHz, with peak energy around the 50kHz mark. Humans, on the other hand, have an absolute upper limit of 20kHz, and typically (in adults) far less.

The bottom line is that humans can rarely if ever hear bats without assistance. The heterodyne receiver simply subtracts a known frequency from the bat's call. So if the bat is squeaking at 55kHz, and you adjust the receiver to 50kHz, you will hear the bat at 5kHz which is easily audible by humans. As a comparison, human speech tops out at about 3kHz.

Its not commonly known, but the "clicks" that you hear from such simple receivers conceal what the bat is really doing. The clicks are actually very fast downwards frequency sweeps; a call may start at 100kHz and descend to 30kHz in as little as 5mS (5 one-thousandths of a second). Not only that, the call tends to have a peak energy frequency as well so the amplitude, and thus energy distribution, is not linear over the spectrum.

Different species of bat have calls that may differ in start, peak and end frequency. Further, even for a given species, the characteristics may change by location, environment or individual bat. You can see that tuning a heterodyne receiver to the peak energy frequency will mean that you will never hear the upper and lower limits of the call. However, in the UK it is often possible to differentiate species by the peak frequency and the pattern of clicks. Experienced bat people do this all the time.

Other simple detectors can also be made, including the so called "frequency division" type. In this, the bat signal is simply divided by a constant number, typically 16. Therefore a 100kHz signal will be heard at about 6kHz, and a 30kHz signal at about 2kHz. using this method you never have to tune the receiver as it can listen to the whole bat ultrasonic spectrum and convert it at once to be human-audible. Unfortunately, you tend to lose all amplitude information (loudness) as cheap detectors use readily available logic dividers that only work in terms of on or off, and hence have no level information in the conversion.

Professionals (biologists and the like), use different, and much much more expensive technology. This includes "time-expansion", where the call is "stretched" electronically so that instead of lasting maybe 5mS, it would last perhaps 20 times longer, and thus the frequencies in the call would be 20 times lower than the original. The huge advantage of this method is that it loses no information from the original signal, and the resultant data can be accurately recorded on an audio tape for later analysis and reconstruction in the lab.

Whilst there are a number of cheap and functional amateur detector kits around, they really are not a lot of good for the more adventurous. They lack accuracy, functionality and robustness. Professional equipment is very expensive, quite functional and robust. However, the technology is essentially crude, and if you want analytical functionality in the field, you need lots of power and very strong arms (typically a laptop computer is required), so you lose flexibility and robustness (try lugging a laptop into the Gunung Mulu caves in Sarawak!).

So that was the germ of the idea. I'm an electronic engineer and mathematician who can do real-time programming, with an interest (non-biological!) in bats. There must be scope for a good project there...

Click on Project Status for more information.