One step closer to automatic ID of bat echolocation

News from iBats is the development of iBatsID a tool for classifying bat calls using ensembles of artificial neural networks (eANN’s) to classify time-expanded recordings of bat echolocation calls from 34 European bat species.

The tool has been trained to identify echolocation calls of Barbastella barbastellus, Eptesicus bottae, E. nilssonii, E. serotinus, Hypsugo savii, Miniopterus schreibersii, Myotis alcathoe, M. bechsteinii, M. blythii, M. brandtii, M. capaccinii, M. dasycneme, M. daubentonii, M. emarginatus, M. myotis, M. mystacinus, M. punicus, Nyctalus lasiopterus, N. leisleri, N. noctula, Pipistrellus kuhlii, P. nathusii, P. pipistrellus, P. pygmaeus, Plecotus auritus, P. austriacus, Rhinolophus blasii, R. euryale, R. ferrumequinum, R. hipposideros, R. mehelyi, Tadarida teniotis, Vespertilio murinus. A fairly comprehensive list.

At present  ID of M. bechsteinii/M. brandtii/ M. daubentonii/ M. mystacinus is only to sub group rather than to species level ( M. myotis/M. blythii/M. punicus;  M. bechsteinii/M. brandtii/ M. daubentonii/ M. mystacinus;  M. emarginatus/M. alcathoe for European readers)

But it’s a great start. The only downside is the current need to process calls with Sonobat first to extract a text file of call parameters.

We can’t be too far now from integrating capturing, geolocating, analysing and identifying echolocation in a single app on a smartphone or tablet such as the Nexus 7.

Link to the paper: http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2664.2012.02182.x/pdf

Link to the iBatsID site: https://sites.google.com/site/ibatsresources/iBatsID

 

Preparing for swarming season

One month to go before swarming season is on us again. I’m looking forward to using the UltraMic250k this year to record swarming myotis species at some of my study sites.

Here’s some light readng for anyone interested in the use of underground sites by bats for swarming.

The Ecology and Conservation of Cave Roosting Bats in the Yorkshire Dales

Autumn swarming behaviour of Natterer’s bats in the UK

Cave selection and use by swarming bat species (you’ll need Athens for that one)

Swarming of bats at underground sites in Britain—implications for conservation

The role of swarming sites for maintaining gene flow in the brown long-eared bat (Plecotus auritus)

The effect of gates on cave entry by swarming bats

Swarming of Myotis mystacinus and other bat species at high elevation in the Tatra Mountains, southern Poland

High gene diversity at swarming sites suggest hot spots for gene flow in the endangered Bechstein’s bat

Host–parasite determinants of parasite population structure: lessons from bats and mites on the importance of time Swarming and mite dispersal

Late summer and autumn swarming of bats at Sikspārņu caves in Gauja National Park

Identification and characterization of swarming sites used by bats in Nova Scotia

 

 

 

 

Video and IR LED Illuminator set up

I was so impressed when I saw Pat Waring’s video presentation at the BCT northern bat workers conference I thought I’d build something similar. While Pat’s system from www.batmanagement.com was great it was also quite pricey and would involve importation from the states.

Pat Waring's set up

Pat Waring’s set up

So I headed off the eBay.

I quick search of the term “IR LED Illuminator” brought up a fair array of choices, most of which were located in China or Hong Kong but given a max price of £10 for a 48 led 850nm unit I guessed it was worth a punt. Sit back for a few weeks until the units arrive and then a quick trip to Maplins for the necessary leads and fittings.

eBay IR LED Illuminator set up

eBay IR LED Illuminator set up

 

Video IR setup from the back

Video IR setup from the back

For just under £50 I managed to get a 4 x 48 IR LED Illuminator set up which runs off the same 7ah 12v battery I use through autumn to power remote Anabats monitoring swarming sites.

I’m using a Sony HDR SR5 video camera which has the all essential 0 lux nightshot setting and records direct to a 40gb hard drive (again an eBay purchase – £100 brand new unused).

Note the 850nm illuminators have a very dull red glow when on, but emerging Brown Long Eared bats haven’t been affected by them so I figure in a barn setting they’re unlikely to be a disturbance factor. Last night they managed to pick up non echolocating BLE’s light sampling in the apex of a barn roof it would have been easy to miss if you were relying just on detectors (I was using both Anabat and Pettersson 240x, Anabat missed them and the 240x only recorded very faint echolocations which couldn’t be seen in Batsound v4).

If you’re surveying sites such as barns likely to have Brown Long Eareds or Myotis sp can you afford not to use video to strengthen your survey methodology?

Here’s an example of the footage and detail you can capture (actual footage is fullHD out of the camera)

[tube]2v-_8MGfhnE[/tube]

 

 

Two new publications for bat workers from BCT

Two new publications for bat workers from the Bat Conservation Trust have become available.

First up is the second edition of the Bat Surveys: Good Practice Guidelines which sees new chapters on; Pre planning considerations, Equipment and techniques, Assessing survey reports, Long term surveys for larger infrastructure projects, Surveying for wind farms and Interpreting results.

Secondly BCT have now published Professional Training Standards For Ecological Consultants, a document “designed to raise standards in professional bat work and outline
the knowledge and skills” required to be a responsible consultant.”

Want to research bats underground?

Want to research bats underground? Here’s some sources of information on where to look in your area; a perfect spring and summer activity to get prepped for swarming and hibernation seasons.

A word of caution: You need to be licenced to survey hibernation sites. You need to be competent and aware of dangers working underground. Cavers and Mine Explorers are sensitive to bat ‘issues’ so don’t go wading onto forums with requests for information of bats in caves and mines, a bit of diplomacy is called for.

A combination of the sites below should reveal sites you never realised were in your area. Why not approach the local caving group for a spot of joint working?

Yorkshire Dales Cave Maps: http://www.cavemaps.org/

UK Caving: http://ukcaving.com/board/index.php

AditNow – Mine exploration: http://www.aditnow.co.uk/

Mine Explorer: http://www.mine-explorer.co.uk/

28DaysLater Urban Exploration: http://www.28dayslater.co.uk/forums/

Approach your local Sites and Ancient Monuments records team for any records of Ice Houses, Lime Kilns, etc.

 

Wildlife Acoustics EM3

Wildlife Acoustics EM3 Active Ultrasonic Bat Detector

From: http://www.wildlifeacoustics.com/

Price: £979.99 plus £199 for GPS

Pro’s: Compact system, full spectrum and zca recording, simplicity.

Cons: ZCA file production needs tweaking, gps noise interference

EM3

EM3 packed with everything you need to start recording (gps optional extra)

The Wildlife Acoustics EM3 is a full spectrum bat detector / recorder with the capability to georeference recordings via an optional plug in gps unit. One of the bonuses from a workflow point of view is the ability to record in wav/zca setting which produces both a realtime full spectrum recording and a zero crossing data file suitable for analysis in AnalookW. This in theory allows you to process zca files to identify easier species such as Pipistrelles but have the use of full spectrum calls for more in depth analysis of harder to identify species.

EM3 first look

EM3 up close

First impressions are here.

I’ve been using the EM3 for a full survey season now and have to agree with Dave Dodd’s that whilst it’s the detector I reach for automatically for most work it is also capable of slowly irritating me over the span of a night.

Ergonomically I’m happy with the package of EM3 and gps puck, it’s easier to hold and lighter in the field than an Anabat, gps and pda. I’m still working on eliminating the inevitable sound files created as you shift it in your grip, something a silicone sleeve might resolve. The detector is a perfect set up for transects and with the option of a seperate wired input for an additional microphone should work perfectly for car use too. Although capable of static programmed surveys I think I’d rather leave that to the SM2Bat+ which is designed for the job.

Recording wise the detector took a fair bit of tweaking in threshold settings before I was happy with the production of zca files, this isn’t such a great issue now that Wildlife Acoustics have now released Kaleidoscope a very handy utility for converting wav files to zca that seems to produce very reliable results. Recordings are crisp and clear and the live black and white sonogram display is great for in the field ID of easier species.

The main advantage over the Anabat SD1 and pda is the lack of connections and potential for issues arising from them. Previously with the Anabat SD1 there were weak points between anabat-pda, pda-bluetooth gps that had the potential to decide out of the blue to not talk to each other; a frustrating scenario when you’re out on site and in the dark. I often carried a separate gps and serial lead “just in case”. With only the gps puck to plug in to the EM3 that’s all disappeared, I no longer have that worry at the back of my mind.

I’ve found I tend to monitor calls in the EM3’s RTE mode, effectively a realtime time expansion and keep the four option buttons programmed to flick between 20,45,55, and 110khz hetrodyne modes. A nice and simple system to flick between even in dark conditions.

All good so far. So that annoyance?

It’s a frustrating one; without the gps plugged in headphone sound is superb but once the gps connects it’s a different matter with a low level variable shussing feedback that inevitably becomes a distraction. This seems to happen regardless of whether you’re listening in hetrodyne or rte mode and is the only downside to what could be a very capable detector. Hopefully it’s something that can be cured with a frimware update, support is one thing Wildlife Acoustics have certainly got dialled with user forums and firmware updates.

*Update: After speaking to Sherwood from Wildlife Acoustics at the BCT National Conference it appears the gps noise issue is to do with hardware so unable to be solved by software. It’s a niggle we’re going to have to live with.

 

 

Dodotronic UltraMic200k digital USB microphone

 

Dodotronic UltraMic200k digital USB microphone

From: http://www.dodotronic.com/index.php?center=11&left=1

Pro’s: Real time, full spectrum recording. Futureproof.

Cons: Needs netbook and software so not suitable for transects. Not suitable for all UK species

Dodotronic UltraMic200k

Dodotronic UltraMic200k

Although not strictly speaking a bat detector, combine the Dodotronic UltraMic200k with a netbook and you’ve got a  light and portable mechanism for capturing full spectrum echolocation in realtime (which is where bat detectors are heading) and all for circa £500.

Ultramic200k and Netbook

Ultramic200k and Netbook

With a sample rate of 200k you’ll only record sound up to 100khz with the UltraMic200k so for some species you’re better off spending the extra on the UltraMic250k which captures sound to 125khz. Sound can be captured using BatSound, SeaWave or other software.

Sure it’s more suited to static surveys (I’ve been testing this one at swarming sites), and identification during data sifting is going to take longer; to say nothing of data storage issues associated with large files.

But, and this is a big but, I believe we’re on the cusp of the next stage in bat identification with automatic identification software just over the horizon. Couple one of these mics with a smartphone or tablet running the right software and you’ve got a powerful, affordable and accurate method of surveying and identifying bats in flight.

Link to the User Manual

Expect to see a Ultramic250k review soon.

Anabat SD1 and SD2

Anabat SD1 (now superceded by SD2)

From: http://www.titley-scientific.com/uk/index.php/anabat-bat-detector

Price: £2062 for SD2 and pda kit

Pro’s: Versatile, small data files.

Cons: microphones sensitive to water damage, cost compared to rivals.

Anabat with pda

Anabat set up with gps and pda

The Anabat is a zero crossing bat detector from Titley Electronics and has long been the workhorse for bat workers needing static longterm monitoring. It’s a versatile bit of kit, capable of either static programmed use or coupled with a pda and gps or just a gps to record transect surveys.

I’ve been using Anabat SD1’s and SD2’s for quite a few years now and up until recently relied on them for most of my survey requirements. The live display of sonograms on the pda is a great tool for instant in the field ID and also invaluable as a teaching tool when running bat detector workshops.

The zero crossing analysis files produced by the Anabat are tiny (think kb’s not mb’s) allowing long term monitoring in busy active sites without the need of purchasing multiple expensive data cards. Coupled with a decent external 12v battery it’s possible to leave the detector running up to a month without filling a 2gb CF card. Workflow using AnalookW to analyse all those collected files is reasonably quick too. Note however this is a very visual system with little useful audio output either in the field or back at the computer.

Hi and Lo MIcs

Ergonomically  and weight wise that Anabat and pda are going to be noticeable if you’re carrying one all night although easy enough to handle in the field.

For european batworkers the two mics of interest are the standard mic and the hi mic, the hi mic is recommended where Horseshoe bats may be encountered or where you’re using an extension cable longer than 1 metre.

For me the weak point of the system is the microphone that needs careful thought to protect it from water contact when deploying it in the field. This is easy on open sites where a cover and 45degree sound deflector can be used to keep the mic sheltered. Underground in typical UK 99% humidity conditions I’ve found condensation eventually builds and takes the mic out. Some users compromise recordings and sensitivity by covering the mic in clingfilm to avoid this.

In use with a pda I’ve also had the system corrupt data files in areas of high multiple bat species feeding activity, a frustrating thing to find at the end of what you think has been a good nights survey. Having started using an EM3 this year, I’ve now gone back to connecting the Anabat to a gps only which when combined with BatNav software is a great reliable tool for simple and quick transect work.

In spite of the above drawbacks I still carry two Anabats in my survey bag because they are just so useful to have and use. The anabat is a great system for long term static surveying with low power use and low data storage requirements but with a few tweaks it could be a lot better. In an time of increased competition I’m looking forward to seeing the next iteration of this detector. If it follows the Anabat Roost Logger I’ll be more than happy

Anabat example sonogram

Anabat example sonogram

Anabat users are advised to keep an eye on Chris Corben’s site for the latest firmware and software.

Lascar USB2 Temperature and Humidity Datalogger

Lascar USB2 temperature and Humidity Datalogger – £49.95 (at time of press)

From: http://www.lascarelectronics.com/temperaturedatalogger.php?datalogger=102

Pro’s: Small, cheap and easily programmable.

Cons: Batteries need ordering (ebay is your friend here)

Lascar USB2 Datalogger

Lascar USB2 Datalogger

The Lascar USB2 datalogger is small, cheap and a very unobtrusive datalogger that’s perfect for recording conditions in bat roosts and hibernation sites. This datalogger is very suitable for monitoring crevices due to it’s size.

Lascar monitoring crevices

Lascar USB2 monitoring Temp/Rh in crevices used by Natterer's bats

Very easy to program with the supplied EasyLog USB software I tend to set them up to monitor at half hour intervals which means the battery lasts one year even in sub zero temperatures experienced during winter in exposed sites. Shorter intervals are possible but this obviously affects battery life.

With a netbook and a pocket of spare 1/2AA lithium batteries you can download and reset dataloggers in the field with ease.

Data downloads can be viewed either within EasyLog or imported into excel or similar spreadsheet programs.