![]() |
||||
![]() |
||||
| Products > Bat detection > AnaBat system | ||||
|
AnaBat system
Typical monitoring approaches using Anabat
The AnaBat bat detector system is used to identify species of bats by detecting, recording and displaying their ultrasonic echolocation calls. AnaBat detectors can be used in two principal ways - to monitor bat activity in the absence of a human operator (Passive Monitoring) or to produce realtime sonograms of calls to aid identification of bats in the field (Active Monitoring). For passive monitoring, calls can be stored to Compact Flash cards, with a single card capable of storing data for several weeks or even months. For active monitoring, the detector needs to be interfaced to a PDA (palmtop) or laptop computer by a device called a ZCAIM. [link to details]
The basic role of any bat detector is to make it easier to observe wild bats through detection of their ultrasonic echolocation calls. Because bats are nocturnal and fly fast, they are often difficult to identify, even in the hand, and because they often fly tens of kilometres from their roosts to their feeding areas, they are amongst the most difficult of animals to observe and to study. Fortunately, all bats other than the megachiroptera (Flying Foxes and other closely related species) use echolocation to find their way around in the dark, to locate food and to avoid collisions with each other and with structures in their habitats. These echolocation calls reveal a bat's presence whenever it is flying or hunting, thereby providing a window into where bats are and what they are doing, in a way which doesn't interfere in any way with the bat's natural behaviour. Echolocation calls are highly optimised for the ecological requirements of each species of bat, and therefore, different species of bats tend to produce different types of signals. This gives observers the opportunity to identify which species of bats are being detected without having to capture them. In practice, this is much harder than it might seem, because bat calls vary greatly depending on where the bat is flying in relation to structures, and also according to what the bat is trying to achieve. This variation far exceeds the differences between species, so in order to identify bats acoustically, it is necessary to have an adequate understanding of each species' echolocation behaviour under a wide range of conditions. In practice, species which are very different often have surprisingly similar calls, while closely related species, which might be difficult to distinguish even in the hand, are sometimes quite easily identified by call. Therefore, a combination of visual and acoustic observations is usually the best way to identify species. Bat detectors can be used in two fundamentally different ways. Firstly, they can be used to watch bats in the field in much the same way bird-watchers watch birds. In this case, called Active Monitoring, a bat detector, and often a spotlight, are used instead of binoculars. Visual observations give the observer information about the bat's behaviour, where it is flying, its flight style and wing shape, and other morphological features, such as ear length, tail length and fur colour. Acoustic observations give the observer information about the types of calls produced by a bat, with the most basic parameters being the frequency of the call (its pitch), the shape of the call (how the frequency changes in time) and the rate at which calls are produced. The combination of visual and acoustic observations also enhance the identification process by providing a context in which to interpret the calls. For example, when the observer knows where the bat is flying, it becomes easier to understand how loud the call is, and how that call fits into the species' repertoire. Active monitoring is the best way to identify free-flying bats, but to be effective, it requires a great deal of skill on the part of the observer. The other major way in which bat detectors are used is called Passive Monitoring. In this approach, there is no observer present, and bats which pass near enough to the monitoring unit are recorded and their calls stored for later analysis. The bat detector effectively becomes a bat activity data logger. It is much more difficult to identify many bats through passive monitoring, because of the lack of visual and contextual observations. However, Passive Monitoring has many advantages which result from the far greater sampling effort it enables for a given resource commitment. A Passive Monitoring station can operate all night, every night for weeks or months or years with very little human interaction, so it is extremely well suited to determining temporal patterns of activity, which in turn can provide basic information about what a place means to bats. Other benefits accrue from the sampling effort achievable. For example, rare bats are much more likely to be detected than by other methods, because exposure to the possibility of detection is so much greater. Passive Monitoring opens up whole new frontiers of bat research, by enabling long term observations on how bats use their environment without interfering with the bats. Data collected by Passive Monitoring stations can be manually scanned by people with the expertise to identify the bats recorded, but there is also great potential for the use of software designed to automate the identification process. This approach has its limitations, because of the lack of contextual information and the inherent problems in bat acoustic identification, but it is already a useful tool which can, for many purposes, be used to deal with the large amount of data which can be collected. Such automated identification provides other benefits, by freeing the identification process from the biases of human subjectivity, and by allowing bats to be usefully surveyed by interested parties without the expertise to make the identifications themselves. Automated identification techniques are still in the early stages of development, and it is to be expected that this approach will become even more useful in the future. The AnaBat SD1 Compact Flash Bat Detector is designed to provide the functionality necessary for operation in both Active and Passive Monitoring modes. It has many features which have been specifically designed with both these roles in mind and which have been found to be important after many years of experience using earlier models for these purposes. Most bat echolocation calls are ultrasonic, which means they cannot be heard by unaided human hearing, and this is why a bat detector must be used before humans can make use of these signals. In general, bat detectors have two roles. The first is to make a signal audible to humans, thus revealing the presence of the bat and allowing the operator to get some information about the nature of its calls from their sound. Because humans are relatively poor at characterising brief, simple sounds like bat calls, it is usually important to take a second step, which is to make the signal visible through some kind of sound analysis, so that its important characteristics can be seen in more detail. Traditionally, two methods have been used to enhance human perceptions of bat sounds. Heterodyne bat detectors can be tuned so the call frequency can be determined from the frequency to which the detector is set. This is the most sensitive approach to bat detection, and more detail can be elicited by very skilled users by re-tuning the detector so some idea can be gained of how the signal varies across a range of frequencies. However, this process is slow, and requires repeated encounters with the same bat. Heterodyne techniques also have the disadvantage that a bat cannot be detected unless the tuning is roughly correct, which means the detector must be continually scanned across a range of frequencies in order to detect different species of bats, further reducing responsiveness of the system. Another disadvantage is that heterodyne output signals are of no use for further analysis using more sophisticated sound analysis techniques. The other commonly used technique is time-expansion, which has the same effect as recording the sound on a high speed recorder then playing it back at a much reduced speed. This reduces frequencies to make them audible and also slows down the sound so it is more easily perceived by humans. The trouble with time-expansion is that it takes much longer to listen to the calls than it does to record them, so it is not a realtime process. There is also no immediate feedback of the frequency of the call, which must be judged by human hearing or analysed in more detail later on by analysis of the recordings. For field use, both heterodyne and time-expansion techniques rely heavily on human hearing as the main way to interpret call details. AnaBat has taken a completely different direction, instead relying on techniques called frequency division, to make calls audible, and Zero-Crossings Analysis (ZCA) to make them visible. These techniques make it possible to see sonograms of bat calls in real time, so for field use, the emphasis is taken off human hearing as the primary agent of analysis, and instead is placed on the visual sense, which is far more effective for assimilating the fine detail required. Sonograms are graphs of call frequency against time, which show the frequencies and shapes of calls in ways which are immediately meaningful to human users. They provide access to details of bat calls at a much finer level than is possible by hearing alone, and they provide this access to users without any requirements for exceptional hearing abilities. There is no tuning requirement, no lag between recording and display, and no need for repeated encounters with the same bat. Frequency division has often been criticised for its incomplete representation of bat calls and its relatively lower sensitivity. In reality, any bat detector technology ends up being a tradeoff between many competing desirable traits and technological limitations. Lower sensitivity, for example, needs to be seen in the context of what the technology can achieve in relation to other technologies which provide inherently higher sensitivity. Even at the simplest level, lower sensitivity can't be equated to the detection of fewer bat calls, because other factors also affect how many calls can be detected and how those calls can be used. In practical terms, the frequency division technique used by AnaBat detectors has been found to be highly effective for almost all purposes for which bat detectors are used, and it has led the way in making it feasible to deploy long-term Passive Monitoring stations and in providing realtime sonograms of bat calls in the field. The AnaBat SD1 Compact Flash Bat Detector has pushed the limit further by making it possible to display realtime sonograms on a unit which can be held in one hand. This is only possible because this bat detector uses frequency division technology, which is exceptionally well suited to the task in its speed of operation, low power consumption and highly efficient storage capabilities, making it possible to build into the detector all the functionality required to interface directly to a PDA. Typical monitoring approaches using Anabat
Anabat is a complex system with many features built in to aid data management and analysis. In addition, some of the skills needed to most effectively use the system are more easily taught through hands on experience. Training courses and consultation are available either on a request basis or as part of courses offered in various locations around the world. Please check here for upcoming courses with open enrolment. For groups needing training or individuals needing specialized skills, please contact us for information about the availability and cost of special trainings or consultation.
|
||||
|
© Copyright 2006 Titley Electronics. All rights reserved. Design by PROdesign.
|
||||