The methods adopted by BaTML for recording the echolocation calls of Pipistrellus spp. using Time Expansion Detectors and the analysis thereafter

Pages 6
Views 6

Please download to get full document.

View again

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
The methods dopted by BTML for recording the echoloction clls of Pipistrellus spp. using Time Expnsion Detectors nd the nlysis therefter Authors: Neil E Middleton*, Kirsty Gourly nd Crig R Mcdm Dted: 27th
The methods dopted by BTML for recording the echoloction clls of Pipistrellus spp. using Time Expnsion Detectors nd the nlysis therefter Authors: Neil E Middleton*, Kirsty Gourly nd Crig R Mcdm Dted: 27th October 2005 *Correspondence Address: 25 Killin Drive, Polmont, Flkirk, Scotlnd, United Kingdom, FK2 0QQ emil: Abstrct The BATS & The Millennium Link (BTML) project undertook Time Expnsion bt detector surveys during the period 2001 to 2005 in order to record the distribution of Pipistrellus pipistrellus nd Pipistrellus pygmeus long the cnl corridors within the Centrl Belt of Scotlnd. This pper reports on our methods for dt collection nd subsequent nlysis. Key Words: Bts, Millennium Link, surveys, pygmeus, cnl, Centrl Belt, Scotlnd Introduction BATS & The Millennium Link (BTML) ws lunched in 2000 to study the bt popultions long nd djcent to the cnl network in the Centrl Belt of Scotlnd (Middleton et l., 2004). During our ctivities we conducted survey progrmme, using Time Expnsion bt detectors (TED), to mp the distribution of Pipistrellus pipistrellus (Common or Bndit pipistrelle) nd Pipistrellus pygmeus (Soprno pipistrelle). Due to the recent seprtion of these species (Brrtt et l., 1997) most existing locl bt records fell short, in tht they could, since the split, only now be described s Pipistrellus spp. We therefore undertook to contribute more precise dt in order to ssist in estblishing the locl geogrphicl distribution of ech species. In order to llocte the bts we encountered to species level, our nlysis methods used the recorded frequencies of mximum energy for these species (Russ, 1999). An overview of the species being studied Pipistrellus pipistrellus (Common pipistrelle) ws thought to be single species, however reserch during the 1990s (s summrised in the following prgrphs) begn to suggest otherwise nd culminted with the support for its reclssifiction, fter DNA nlysis, s two different species (Brrtt et l., 1997). These two biologiclly seprte species re now referred to s P. pipistrellus (Schreber, 1774) nd P. pygmeus (Lech, 1825) (Jones & Brrtt, 1999). Ech of these species occupy their own seprte mternity nd mting roosts (Prk et l., 1996) nd occur throughout the UK nd mny prts of Europe, lthough in some prts of Europe there re res where either of the two species pper to be the more dominnt (Jones & vn Prijs, 1993). Subtle differences in their physicl chrcteristics, which cn be difficult to scertin (Dietz & von Helversen, 2004), hve lso been documented. P. pipistrellus is, on verge, slightly lrger nd lso ppers to hve drk fce giving the impression of bndits msk, hence one of its commonly referred to nmes, Bndit pipistrelle. A subtle difference occurs in the pttern of the vention in the wing membrnes of ech species. A difference is lso often pprent when compring the rtios of the lengths relting to the 2 nd phlnx to the 3 rd phlnx, on the third finger. Hbitt preferences hve lso been described. P. pygmeus ppers to show stronger ssocition with hbitt feturing wter (Vughn et l., 1997; Brtonick & Rehk, 2004) nd nlysis of its diet (Brlow, 1997) hs supported this. P. pipistrellus ppers to be more of generlist occurring in fr wider rnge of hbitts (Vughn et l., 1997; Russo & Jones, 2003; Dvidson Wtts & Jones, 2005), lbeit including hbitt ssocited with wter fetures. Pipistrellus pipistrellus typiclly produces frequency modulted echoloction clls with frequency of mximum energy (FmxE) verging t circ 45/46 khz, whilst P. pygmeus produces similr looking cll but t higher frequencies with FmxE usully between 52 khz nd 56 khz (Jones & vn Prijs, 1993; Vughn et l., 1997b; Russ, 1999; Prsons Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 9 & Jones, 2000). Figure 1 shows comprison of echoloction clls from ech species fter recording from TED nd displyed s spectrogrm (i.e. frequency compred ginst durtion/time) using sound nlysis softwre (Pettersson Elektronik AB, BtSound, V3.0). Figure 4 shows Power Spectrum (the tool used to estblish the FmxE) of typicl P. pygmeus cll. The Power Spectrum nlysis llows the power of signl to be compred cross the rnge of its frequency components. It should be noted tht the possible extremities of rnge of the FmxE emitted by ech species is wide enough to produce n overlp in the frequencies between 48.8 khz nd 51.8 khz (Russ, 1999). This mens tht bts emitting FmxE within these prmeters cnnot be sfely llocted to species level on the grounds of FmxE lone. However, when considered in ssocition with cll length nd end frequency it my be possible to be conclusive (Myer & von Helversen, 2001). In ddition to their echoloction differences, ech species hs been shown to hve their own chrcteristic socil clls (Brlow & Jones, 1997). 100 khz 50 khz Spectrogrm, FFT size 512, Hnning window sec. Figure 1: Spectrogrm showing typicl echoloction clls of Pipistrellus pipistrellus () nd Pipistrellus pygmeus (b) Methods Time Expnsion A Summry This system llows sequences of echoloction to be temporrily cptured by bt detector in rel time before being plyed bck for recording purposes t slower speed (e.g. 10 times). This mens tht the ultrsonic echoloction (i.e. occurring bove the frequencies of norml humn hering) is presented to listener t lower frequency nd over longer period of time. For exmple, if bt is echolocting t frequencies of 50 khz over period of 0.6 sec (beyond humn comprehension), this cn be slowed down by 10 times to become n udible sequence t 5 khz nd lsting 6 sec. The recorded sound, plyed bck by the detector, is exctly the sme s the originl, only 10 times slower. This mens tht it cn be b 90 db 70 db 50 db 30 db 10 db cptured nd retined on recording device for review using one of number of computer softwre tools vilble for sound nlysis. The softwre llows sound to be nlysed through spectrogrm (frequency ginst time), oscillogrm (mplitude ginst time) nd power spectrum (power ginst frequency) fetures. Figures 2, 3 nd 4 give exmples of these formts (Pettersson Elektronik AB, BtSound, V3.0) using Pipistrellus pygmeus recording tken during one of our surveys (Fwnsprk, Union Cnl, ). A TED system, used in conjunction with sound nlysis softwre, is in mny respects idel for interpreting bt echoloction. The qulity of the results obtined is superior to nything else tht currently exists. However, unlike other systems where sound nlysis is possible (i.e. Frequency Division), wekness when using TED system is the fct tht whilst the detector is plying bck the results of cptured sequence, it is no longer ctive s surveying tool (Prsons et l., 2000; Ahlen, 2004). Therefore it must be borne in mind tht bt psses will be missed s result. If quntity, s opposed to qulity, is more crucil to proposed survey method then other systems, or combintions of systems, should be considered. We wished to focus on identifiction nd distribution, therefore we ccepted tht we would potentilly miss bt psses nd when describing ny results relting to ctivity nd bundnce we would need to mke cler the limittions of the methods we dopted. Dt Collection Methods Figure 5 shows the typicl set up of the TED system used by BTML during our surveys. The TED (Courtpn, EcoTrnquility) records ll echoloction, regrdless of frequency, within its rnge (15 khz to 120 khz). It is set to time expnsion fctor of ten times nd signl storge time of sec. This signl storge time llows more thn dequte length of recorded time to enble identifiction relting to Pipistrellus spp. The TED is mounted onto tripod which is extended to 150 cm in height. A stereo led is connected from the Record Line Out of the TED to the Microphone Line In on the recorder (Sony Professionl Wlkmn, WM D6C) which is ttched to the tripod beneth the detector. The TED is positioned fcing directly cross the cnl t 45 degree ngle towrds the sky. This is in order to mximise the chnces of picking up Pipistrellus spp. tht pss by over the cnl corridor. The TED is switched on nd the Record/Ply buttons on the tpe recorder re engged, quickly followed by the Puse button. A test signl is then generted using the TED in order Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 10 Spectrogrm, FFT size 512, Hnning window. 90 db 70 db 50 db 30 db 10 db 100 khz 50 khz sec. Figure 2: Spectrogrm showing typicl serch phse echoloction pulses emitted by Pipistrellus pygmeus (Fwnsprk, ) Amplitude 100% 50% 0% 50% 100% Figure 3: Oscillogrm for the sme echoloction sequence shown in Figure sec. 0 db 20 db 40 db 60 db 80 db 100 db 120 db 56.9 khz, 6.8 db Distnce: 0.0 khz, 0.5 db 50 khz khz 5326 ms Figure 4: Power Spectrum relting to pulse shown in Figures 1 nd 2. Frequency of mximum energy (FmxE) = 56.9 khz Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 11 to set the mnul recording volume level on the tpe recorder. A level tht llows strong signl to be received from the TED without potentil clipping/distortion occurring to the received recordings is selected (note tht during nlysis recordings cn be checked to ensure tht no clipping hs occurred using the oscillogrm feture). Hedphones re connected from the tpe recorder so tht surveyors cn check wht is being recorded during the evening. The Mster Volume on the Tpe Recorder llows the hedphone level to be set independently from the recording volume. Noise coming directly from the speker of the TED is reduced by turning the volume control of the TED to zero. At 30 minutes fter sunset the Puse button on the tpe recorder is disengged, surveying begins nd ctivity is recorded for 90 minutes therefter. All dt is timed nd recorded onto two ppropritely numbered C90 tpes (TDK FE, IEC I /TYPE I, Norml Position). Tpe 1 is inserted into the recorder t the beginning of the survey nd whilst testing the equipment. During the survey, fter 45 minutes, Tpe 1 is removed nd replced with Tpe 2 which upon its completion (nother 45 minutes lter) concludes the survey. The survey methodology lso llows for the collection of wether conditions (Middleton, 2004). Figure 5: Time Expnsion Recording System. Typicl set up flow chrt (items not to scle) Stereo Cble Microphone Time Volume Tpe Recorder D 640ms Time Expnsion Bt Detector Bttery Chmber Tpe Comprtment B C A Mster Volume Tpe Deck Controls Hedphones Side elevtion of TED nd Tpe Recorder mounted on tripod Key: A Record Line Out B Microphone Line In C Hedphone Line Out D Recording Controls The BTML project hs successfully used this method to record Pipistrellus spp. in Scotlnd between 2001 nd Over this period, 117 surveys were completed nd the equipment nd protocol proved to be robust nd consistent. Over 5,000 bt psses were recorded nd nlysed. Sound Anlysis Methods The tpes from ech survey re nlysed using sound nlysis softwre (Pettersson Elektronik AB, BtSound, V3.0). The tpe recorder is set up nd the two tpes relting to the survey re rewound to their strting positions. The tpes re numbered ccording to the order in which they were used during the survey evening. During the nlysis four 23 minute time slots re creted so tht ny chnges in behviour reltive to sunset cn be considered. All dt is recorded onto pper survey forms (Middleton, 2004) nd lter stored in digitl formt (MS Excel nd MS Access, 2003). Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 12 In order to crry out the ctul nlysis the tpe recorder is connected to PC vi stereo udio cble. This cble is connected to the Line Out of the tpe recorder nd Line In on the PC. Anlysis is then undertken using the spectrogrm, oscillogrm nd power spectrum fetures within BtSound V3.0. We used smpling rte of khz, with 16 bits/smple nd 512 pt FFT with Hnning window for nlysis (Pettersson Elektronik AB, 2000; Ctto & Agte, 2003; Russ, 1999). Every sequence of echoloction producing recording of dequte qulity is mesured using the Power Spectrum feture (Figure 4). To estblish n verge FmxE for ech sequence we select three clls within the pss nd clculte n verge mesurement, rounded off to one deciml plce. Individul cll frequencies re then logged onto spredsheet (Microsoft Excel) nd the results summrised on our pper survey forms (Middleton, 2004) nd MS Access dtbse files. All sound files re lbelled nd stored on disc in cse further nlysis is required t lter dte. The nlysis dopted ws consistent during the five yers, nd in ll but hndful of occsions the sme equipment ws used throughout. The equipment ws regulrly checked for under/over performnce s we were ble to udit ctul recording timescles ginst the 640 ms time frme selected during surveys. In order to further check the equipment/processes, collection nd nlysis relting to other non BTML sites ws lso introduced. On two occsions we used other TED mchines nd minidisk recorder side by side with the equipment used in our min study in order to compre results from the sme groups of bts. The nlysis compred fvourbly during these exercises nd provided ressurnce tht our equipment performnce nd nlysis therefter ws robust. Summry The survey nd nlysis methods dopted pper to hve supported the ims for this study well nd we re now in the process of interpreting the results which will be published seprtely in the ner future. Acknowledgements We re gretly pprecitive of the funding obtined from Scottish Nturl Heritge, British Wterwys, Flkirk Environment Trust nd The Royl Bnk of Scotlnd Group. In ddition to our funders we re lso keen to cknowledge the support given to us by the following orgnistions: The Bt Conservtion Trust, BTCV, Centrl Scotlnd Bt Group, Clyde Bt Group nd Lothins Bt Group. Finlly we would like to thnk ll of the volunteers tht hve been ssocited with this project, unfortuntely too mny to nme, nd in prticulr the following people: Chris Gould (BTML), Dvid Dodds (BTML), John Hddow (Centrl Scotlnd Bt Group), Olivi Lssiere (British Wterwys), Lilin Lind, Shon Mckenzie (BTML), Kirsty Morrison (BTML), Sturt Smith (Lothins Bt Group) nd Anne Youngmn (The Bt Conservtion Trust). References Alhen, A. (2004). Heterodyne nd time expnsion methods for identifiction of bts in the field nd through sound nlysis. Bt Echoloction Reserch, Bt Conservtion Interntionl, Auston, Texs, USA. Brlow, K. E. nd Jones, G. (1997). Differences in songflight clls nd socil clls between two phonic types of the vespertilionid bt Pipistrellus pipistrellus. J. Zool., Lond. Vol 241, Brlow, K. E. (1997). The diets of two phonic types of the bt (Pipistrellus pipistrellus) in Britin. J. Zool., Lond. Vol 243, Brrtt, E. M., Deville, R., Bylnd, T. M., Bruford, M. W., Jones, G., Rcey, P. A. nd Wyne, R. K. (1997). DNA nswers the cll of pipistrelle bt species. Nture, Vol 387, Brtonick, T. nd Rehk, Z. (2004). Flight ctivity nd hbitt use of Pipistrellus pygmeus in floodplin forest. Mmmli, 68(4), Ctto, C. nd Agte, J. (2003). Using Btsound with frequency division recordings. The Bt Conservtion Trust. Dvidson Wtts, I. nd Jones, G. (2005). Differences in forging behviour between Pipistrellus pipistrellus (Schreber, 1774) nd Pipistrellus pygmeus (Lech, 1825). Journl of Zoology, 268, Dietz, C. nd von Helversen, O. (2004). Illustrted identifiction key to the bts of Europe. Version 1. Tuebingen & Erlngen (Germny) Jones, G. nd vn Prijs, S M. (1993). Bimodl echoloction in pipistrelle bts: re cryptic Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 13 species present? Proceedings of the Royl Society of London. Vol 251, Jones, G nd Brrtt, E. M. (1999). Cse Vespertilio pipistrellus Schreber 1774 nd V. pygmeus Lech 1825 (currently Pipistrellus pipistrellus nd P pygmeus; Mmmli, Chiropter): proposed designtion of neotypes. Bulletin of Zoologicl Nom. Vol 56 (3) mens of brod bnd coustic method. J. Appl. Ecol. Vol 34, Vughn, N., Jones, G. nd Hrris, S. (1997b). Identifiction of British bt species by multivrite nlysis of echoloction prmeters. Biocoustics. Vol 7, Myer, F. nd von Helversen, O. (2001). Symptric distribution of two cryptic bt species cross Europe. Biologicl Journl of The Linnen Society. Vol 74, Middleton, N. E., Gould, C., Mcdm, C. R., Mckenzie, S. nd Morrison, K. (2004). Introducing BATS & The Millennium Link. A study of bts nd their use of cnl corridor hbitt in the Centrl Belt of Scotlnd. BTML Publictions, Vol 1, 2 5. Middleton, N. E. (2004). Survey form templtes used by BTML in connection with the monitoring nd distribution of bts in centrl Scotlnd. BTML Publictions, Vol 1, Prk, K. J., Jones, G., Altringhm, J. D. (1996). Assortive roosting in the two phonic types of Pipistrellus pipistrellus in the mting seson. Proceedings of the Royl Society B. Vol 263, Prsons, S. nd Jones, G. (2000). Acoustic identifiction of twelve species of echolocting bt by discriminnt function nlysis nd rtificil neurl networks. Journl of Experimentl Biology. Vol 203, Prsons, S., Boonmn, A. M. nd Obrist, M. K. (2002). Advntges nd disdvntges of techniques for trnsforming nd nlyzing chiroptern echoloction clls. Journl of Mmmlogy, 81(4): Pettersson Elektronik AB. (2000). BtSound. Rel time spectrogrm sound nlysis softwre for Windows 95/98. Russ, J. (1999). The Bts of Britin nd Irelnd. Aln Ecology Ltd, ISBN X. Russo, D nd Jones, G. (2003). Use of forging hbitts by bts in Mediterrnen re determined by coustic surveys: conservtion implictions. Ecogrphy. Vol 26, Vughn, N., Jones, G. nd Hrris, S. (1997). Hbitt use by bts (Chiropter) ssessed by Copyright BTML Publictions 2005,, ISSN Volume 2 December 2005 Pge 14
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!