장음표시 사용
141쪽
where a particular Snake Specimen is staying, plenty of other ecological and he-haviourat observations can be gained. BehaViour, Pre' utiliSation Especialty with direct tracking, radiotelemetry gives an opportunity to Observe the bellaviour of the snake. Detatis os laeding, suci, as sit and wait or active foraging Strategies, Sheltering Shedding, couning, copulation, etc. can be collected. The mod of the snakes can be identi Ged by excrement analysis during the active intervention stransmitter implantation, battery change) or in the fiet d. based on insect chitin cuticle. mammal hair or bone remnants REINERT et al. l 984). There are a number of peculiar en viron mental conditions, when the bella violirat aspects of the gi ven species can hardly be stud ted in any other way than radiotelemetry. For example the investigation of the bellaviour of the aquaticarasura filesnake Acrochordus ara furae) was almost impossibie hecati se of themuddy water fult with crocodiles. Iis dat ly activity, including temperature regula
l98J) suggest two different statisticat methods for calculation proportionalsimilarity index PSI) and the well-known Chi-square test.
Hibernatiori There are severat ex amples silowing that snakes frequently use a common
hibernation site. somelimes years after years ALEKSIUM l9J6. BURGER et til. l 988. JACOB Ω PAINTER l 980). Is tracking a specimen supplied with radio transmitter is successsul untii it finds the hibernation site, actualty the wintering den of the whole population can be mund . CIOFI and CHELAZZI l99 l) mund that specimens of Coluber viridi flavus are faithmi to their hibernation sites and return to itnot only every autumia, but also for hiding in their active period.
142쪽
Biotele metric studies are at med at studying various phySiological, ecological and bellavioural parameters. In the first period os using biotelemetry, implanted radio transmitters provided data on the temperature relations hip of the digestion of snakes. NAULLEAU l9J9) gave a good summary on the beginning and devel opinent of the investigation os temperature regulation. Besore the advent of radi Otelemetry, qui Ck-recording mercury thermometers were put in to the cloaca of snakes and the hody temperature of snakes was measu red during different acti- vities sKITCHELL l 969). In this way continuous temperature regulation is difficult to follow, since meastirements are interrupted and always go together withthe disturbance of the bellavio ur. Later the thermosonda stelethermometer) was
Later the position of the implantation was influenced by the study purpose whether to meastire temperature dynamics of digestion, hibernation or en Viron mental effectS. Picinning a Diotelemetric Stui Measuring can be done in severat different ways. In laboratory a shortrange recei Ver is enough, and mensure menis cata be taken automaticalty or hyhand at regular intervat s. In the Geld usu atly long distance recei Vers are needed. and automatic data collection is more costly. Is finances or Salaly conditions donot allow automatic data processing an Omnidi rectionat whip antenna or a liand-held Yagi antenna can do a good service for manual signat recording. In both cases the main purpose is to not disturb the animais' natural activi ty. The principie of most temperature sensitive transmitters is to give signat s at different pulse rates according to the change of temperature. A calibration diagram is usualty provided by the manufacturer to identify the proper temperature frona the regular time tength sin milliseconds) between the individual signat s. The degree of deviation is generalty not more than 2'C. A simple receiver cannot directly show the intersignat period in this case counting the number of signat s during a certain time intervat se. g. l0 seconds) can hel p to calculate it. Modern transmitters display the intersignat period in milli seconds. To understand thebody temperature changes of the snake, it is recommended to monitor continu-
143쪽
otis ly the ambient en viro iamental factors like soli. air and vegetation temperature light or radiation intens ity, wind Speed, etc.). Frequency of recording is usualty determined by the en viron mental factorsand the study conditions. NAULI EAU and MARQUES sl 9J3) e. g. made notes in every 5 minutes when the temperature change was faster and in every tb minutes is it flowed down. LILLYwHITE l 980) adjusted the measti ring intervals to anaverage of 20 minutes. hut somelimes faster or flower, according to the dat ly activities of sive Australi an sinake species. We have mund that a measurement in every hour during a 24 hours period is quite appropriate to record the activity pat-tem of a temperate Species. Automatic systems can be ad justed to any desii edrecording intervat. Data processors bulli into the receiver can record temperature
MOORE l9J8) determined the temperature categories on the basis of thesnakes' dat ly activi ty palteria. Preferred temperature PT) was defined as thenaean os ali hody temperatures recorded with in the normat activi ty period. it represenis the mean Voluntary temperature of an active individual. Voluntary minimum Vmin) was defined as the lowest hody temperature recorded in an active individual: whereas voluntary maximum Vmax) was the sanae at the higher encl. Heiace the relationship between the three is obvious: normat activity range NAR) is the range of body temperatures of active individuals frona voluntaryminimum to voluntary maximum temperatureS.
Biotele metry is osten connected to the Observations On movementS, homeranges, territories or hibernation sites of the snakes. However with biotelemetrytemperature sensing is always in focus, whicli can give hinis on the mobility as
144쪽
well. Night movemenis, for example, are so metimes indicated by temperature
change, and acti Vlty at night can be determined by the varying strength of the
transmitter signat heing received as the snakes crawled Dom one place to another
TO determine the home range or range of activity of a snake many factorsshould he taken into consideration at ready at the plania ing Stage of a Study. The firsi and perhaps the most important is the sampling frequency. It is extreme lyimportant to gain even information frona the different activity periods of the animal. Selecting an appropriate Sampling intervat can be cruciat to the success of the investigation. There is a conflici belween the number of individual observations. which should be as dense and as many as possibie. and the disturbance of the snake's bellaviour, which should he, on the other hand as low aS possibie. The tength of a sampling procedure should be relatively shori, but enough to collect the maximum information abolit the gi ven locality. With the increasing num-her of observations the fige of the home range and the accuracy of iis estimationis generalty increas ing too. Sampling frequency should differ in the breed ingperiod. hefore, after and during hibernation and in other relatively passive peri-ods. Range of activi ty is a function of the fige, age and Sex of the specimen asweli as of iis laedin g, shedding and other bellavi oural or sociat habiis. In dividualmoVements of males are usualty larger than that of the se males, and in copulationtinae it can even exceed their energetic requirements. Juvenile snakes have usu atly Smaller home ranges than the adulis, whicli is the result of the sexuat an dintraspecific interactions of the lalter. It was observed in the case of ii Zards ROSE l 982) that males have extended their home ranges after sexuat maturi ty tomeet more se males, whicli, on the Other hand, Occupy oVerlapping areaS. The Si Ze of the home range may also depend on the densi ty Of the population, IS an antagonistic relation. Home range siZe should be regarded also dynamical ly, a gi ven individual can change iis range of activi ty according to the presence Or abSence of conspecific competitors. lack of 1 ood, changing en Viron mental conditions, etc. It is important to make a difference belween the error of measti rement and
145쪽
This is the simplest and most recommended method in the case of relatively law data collected at constant frequency during a stiori period. Estimation of exact home range si Ze is difficult hiat dat ly move ments and activi ty patieria is easy to follow. The technique is made up os pulting the different locations ora the map and simply connecting them in time sequence FUISZ l 995. SZEMETHY l 995).
nil Polygon - This method is based on the work by IENNRICH and TURNER sl 969) whod rew the smallest convex polygon containing ali of the capture potnis. Locations are described in atwo-dimensional co-ordinate system. in Whicli the convex polygon also appears wheia connectingati the outermosi potnis ROS E l 982). The di sad vantage of the method is that it takes into consideration ali locations regat illess their possibie biological meanings, and that it cloes not differentiatebetween poliat s with variotis capture probabilities. The area os the polygon is an estimate sor thelio me range si Ze, and can he calculated with co-ordinate geometrical formula see WHITE & GAR
drawbacks. on the other hand. hi ought up in the literature. In addition to the two mentioned aliove the main problem is that the estimated si Ze of the polygon depends large ly on the number os observations. Increas ing sampling frequency extends the home range si Ze. whicli is not in accordance with the biological meaning. Estimation cloes not only consider regular movemenis, hut the wholeare a covered by locations, so the polygon includes areas that were actualty not visited by the animal. Reasons why certain potnis are far si om the Vcore ' area cannot he revealed with this method. To decrease hias resul ling 1 rom the number of superfluous observations the 95ri convexpolygon method was introduce l. whicli simply means that locations contributing mostly to the un- destred increase os the area are test out si ona the calculation. The limit is drawn at 5 , os ali the capture potnis. Knowledge on the ecology of the gi ven species can also suppori certain type os selections when the estimating the homo range os the aquatic Acrochordus arasurae by the convexpolygon method. terrestriat habitais inside the polygon were test out sor the snake never goes onto
MΛDsEN sl 984) calculated homo ranges by two disserent ways using con Vex polygonS. Totalhomo range was desined by summari sing ali locations during the tracking period. while combinedhome range was got by adding the areas observed during the activi ty periods in each monili. Minimum Polygon - Minimum polygon or minimum area method differs si Om convex polygon by restricting the connection os capture potnis to the more frequently repealed observations. Rare ly sonce) used potnts of occurre iace are excluded si ona the polygon and heiace the estimated area is much smaller than in the case os convex polygon. The ad vantage of this method that it istes s dependent on the number os locations. Graphical evaluation is also easter, since it hecomes geometri catly dissiculi as the number os capture potnis increases. Algebraic evaluation is simple JENNRICH 9 TURNER l 969). A dra back of the technique that the area drawn is tess accurate an l
ameters os an assumed distribution os the actuat observations ROS E l 982). Home range si Ze estimation depends on the statisticat methods applied.
146쪽
Recapture radius - The distributiori of the locations is approximated by the normal Gauss)distribution. First the geometric centre is de ned by averaging ali potnt co-ordinates, then the distance of every potnt Dom the centre is calculated. The average of these distances is taken as a radius of the estimated home range. The recapture radius method gives usualty slightly bigger areathan the convex polygon method. Estimations can disperse very much is the shape of the actualhome range is sar hom a circle. Individual Variations may also destroy the accuracy of the radius method. Severat aut hors recommend the determinant method instead sJENNRICH 9 TURNER l 969
Bivariate normal modela - It is true for ali of the ho me range estimation methods that therando inness of observations is assumed. It is also true in the majori ty of cases, that the capture
probabili ty of the specimen in the geographicat naiddie of iis activi ty range is suppo sed to be higherthan in the peripheri CS. In Order to avoid the bias of the previoris methods JENNRICH and TURNER l 969) invented the determinant estimator whicli is hased on the radius method but using covariant matrix. The method is also called JENNRICH-TURNER estimator WHITE & GARROTT l 990), or confideiace ellipse method REINERT l 992). It operates with a bivariate normal distribution and allows an elliptic homerange sit ape, too. The si Ze of the activi ty range is determined by a certain confide iace intervat. with in whicli the occurrence of the animal has a hi gli probabili ty value. For the ultimate calculations see JENNRICH and TURNER l 969) or WHlTE and GARROTT sl 990). The main ad vantage of the method that it does not depend on the sample si Ze. so different resulis can eas ity be compared. Iis drawback. however. that for a reli able estimation relatively high number of Observations is necded. At 20R covari ance levet the minimum is lGO samples sWHITE & GARROTT l 990). The weighted hi variate normal estimator is an improved JENNRICH-TURNER estimator. The differendo lies in the weighting procedure os ali data by the distance stom the average centre.
With the non-parametric methods the normal distribution of data is not a condition and theydo not have a demand sor minimal data set. Apari frona the following two. there are severat Other parametric and non-parametric methods such as multiple ellipses, Dunn estimator. Fourier series sinoothing, etc. HITE & GARROTT l 990), but hecause of their rare applications we do not go intodelails with these procodUreS. Grid celi - The si Ze of the activi ty range is determined with the hel p of a gi id covering the observed area. Grid celi si Ze is calculated with the average distance belween consecutive locations, somelimes adjusted by the si Ze of the study area. Homo range estimation is made by including celis that have at least one occurrence. The area obtained may be a concave polygon, but it may at Soconsist os severat disjunci fornas. Is relative frequencies of the occurrences are also considere l. theu se of the different paris of the ho me range can also he estimated 'core areas ). The overtas' intime and space belween different individuals of the fame population is easy to sollow and depict. Using background information. such as vegetation maps habitat prefere iaces can also he investigated. Harmonic mean - The method was invented by DIXON and CHA PM AN l 980). In herpeto
the study area. Distances of every localities Dona ali grid corners are measu red. then the localitiesare categori sed according to the harmonic mean values of these distances. Potnis in the fame category are connected and forini an isopleth. The isopleths de limit areas used with equat probabilities
147쪽
Summary Osdata evaluationSummarising the different home range sire estimation methods describedaboVe, the con Vex polygon can be recommended in most cases where sample si Zeis reason ab ly high ROS E l 982). Iis main disad vantage is that the si Ze of the esti- maled area increases with the sample sire. Proin the statisticat methods grid celland harmonic mean are the most widespread REI NERT et ah l 984). The ad vantage of the grid celi method is that it gives information on the locations that arethe result os accidental movemenis and occupy peripherat places. It is also fuit- able for analysing overi apping patierias and habitat preferences. Iis drawback lies in the dependence on the reali sation. Harmonic mean calculates only with the areas reatly occupied by the animal and can represent home ranges of any kind of shape. The areas limited by the isopleths are determined by the intens ity of the activi ty. Selecting the different isopleths different home range paris can hedefined: 95N isopleths to represent total home range or total activity range, f0Gisopleths to represent a Core area, or area of most intense use, J5N intermediaterange can be calculated lar comparative purposes. The core areaS repreSent Ona Verage only about P. -9R of the total home range sire REI NERT & CUNDALLl 982. REINERT & TAPPA LORTI l 988. SECOR l 994. TIEBOUT L CARY l98 ). The harmonic mean is also well-suited to time- series analysis used to detect shises in activity centres TIEBOUT & CARY l98 ). SECOR sl 994) compared convex polygon and harmonic mean methods and has mund severat hectares differetice to the ad vantage of the former. With the development of computer programs a PC Sostware package was made for estimating home range si Zes with minimum convex polygon and harmonic mean methods Micro-computer Progranas for the Analysis of Animal Locations - McPAAL Ver. l. 2l, Michael Stuwe, National Zoological Park of the Smith sonian institution, Front Royal. Virginia).
148쪽
their externat attaching method, whicli is good when one has to replace the bat-tery or repair the transmitter, but we have reservations with the drastic way offixing the device to the vertebrat column with a plastic thread. Even is it is on lyon the tali, it can cause serious troubles to the snake when moVing hetween thedense iussocks in the grass and stays open to infection due to the permanent
In our opinion the best method to attach a transmitter is the subculaneous implantation. For immobili sation during surgery we recommend hul Othane anaeSthesia using a professional respiration machine. With the machine respiration canhe regulated and an aesthesia maintained at the edge of cessation os longue flicks. With the proper close of the narcotic and the oxygen the risk can be minimi sedand awakening shortened only to severat minutes. Despite this, indoor controi of the vipers in the following one or two Jay S is alwnyS neceSSary. Besore selecting a transmitter for implantation it is wise to ask informationabout the si Ze lifespan, and effective range froin the manufacturers On the mar-ket. Although companies can come and go, a u Sesul Selection Os present manufacturers can he found in Table t of this paper. Is it is possibie. a triat os a sample transmitter could also he extremely beneficiat in the choice of the different trade
In Our Study we selected the transmitters produced by Holohil Systems Lld. Ontario. Canada stype BD-2GT) to study the move ments and temperature rela
is 2 MHZ. Their life span was an nounced to be around J-8 monilis. depending onthe enViron mental temperature. Average hody mass of the vipers is abolit 40 g tength 35 40 cm. A transmitter which was implanted Subculaneous ly made up todia of the body mass of the Snnke. About one monili aster the implantation, the antenna wires of the transmitters pierced the skin of the snakes. In one specimen a loop came out where thewire was originalty threaded. in the other the end of the wire pierced the cuticle. The antennas were put back into the animais and the openings were fixed withsurgical tapes. Another one monili later the animal S were in proper conditions re-garding the implanted transmitters. Later one transmitter was mund on iis own Onthe ground sit was probably pushed out of the animal). The other gave normal signat s 4.β montiis after the implantation, but then suddenly indicated uia belle v- able high body temperatures and stopped to function in the next week. We haVelost both the animais and the transmitter inside. An other female vi per received aswitchable transmitter frona the sanae set ordered frona Holohil in l995. huistopped to function in only l. 5 monili. Our inquiries both abolit the wire rigidi tyand the expected life span remalia uiaresponded by the manufacturer.
149쪽
The signat s of the transmitters in the field were received by a two- or three- element Yagi antenna, a liandy loop antenna and a Simple radio recei Ver apparatus. For biotele metric purposes more advanced receivers with bulli- in data processor) and an omnidi rectionat whip antenna can also he usefui. S mali grassiandvipers are sit and wait predators FARREI L et ah l 995) and have a relativelysmali activity range . They usualty spen d a tot of time in or near a rodent hole or a
It is very much recommended to compile a good map on the Study area Oreven combine it with a vegetation map with preci se horders of the different plantassociations s grassiands may have extremely important vegetation microstructure
whicli is hard to identi sy and describe without the hel p of a botan isti). With theaid os a good map we can immediately determine our position sand that of the
etc.), and relati Vely Scarce sonce every week) in the more passive periods summer 'aestivation , shedding digestion etc.). The habitat of the Hungariannaeadow viper is altogether only severat hectares, so locations were taken eVeryweek not to disturb the animal s. Successsul locations were marked with a poleand nexi time search could be started frona that poliat. To record their dat ly activ-lty and temperature regulation we conducted severat 24-hour long continuous monitoring, when locations were Omitted, only body temperatures by remote sensing were collected in every hour. During fast changes fearly moria ing and late asteria oon) half an hour or even more frequent recordingS are recommended. During the Jay or especialty at night on the other hand the temperature of the vipers is relative ly stabie or only oscillates around a certain potnt so two h Oursrecording intervals may prove to be adequate. Sudden hut in their extent si alltomperature changes, howeVer, can indicate move ments like sheltering in a hole so 'stabilityV should always he handi ed carem l ly. Evaluation of the data collected depends on the sample si Ze, obserVationfrequency and the aim of the study. In a poorly known species even few data and the compilation of a simple trackogram can provide many useful and new information see more delatis: VIVARI & MORSOS l99 l 999. VIVARI et til. l 999). Por home range analysis with more locations we recommend the harmonic mean method. Biotelemetric data si . e. temperature regulation) should be evaluated inconnection with the environ mental factors.
150쪽
Summari sing ali the information avat labie for iis frona the literature and DOm OUr Own experiences, we can conclude that radiotele metry is an extreme lyuseful method in the ecological study of snakes. In the case of large species it is very uses ut to follow their move ments. to describe the seasonat and dat ly activi typaltern S, estimating home range dynamics, and many other ecological and bella-viourat aspecis. In s mali snakes iis applicability is limited to a certain degree by the si Ze of the transmitter, hut the method can stili provide extreme ly useful and missing information on the biology of the animal s. which otherwise would almost be impossibi e to Study.