Fish tagging
Detecting biotelemetry signals
Localizing transmitters, positioning fish in their habitat
13. The topics covered in theory included the operation of transmitters, antennae and active physical tags (biotelemetry transmitters and data storage tags), fish anaesthesia, tag attachment, effects of tagging on behaviour and physiology, transmitter retention, basic principles of radio and acoustic signal propagation, positioning of transmitters, as well as cartography and related mathematical principles. Only the broad outline of each of these aspects is presented in this report. A more exhaustive description is available in a manual of some 160 pages developed during the preparatory phase of the workshop (E. Baras, V. Bénech and G. Marmulla) and distributed to all participants at the start. The syllabus contents can be found in Annex 5 of this report.
14. The first seminar focused on procedures for attaching telemetry transmitters or data storage tags to fish. Telemetry transmitters contain an energy source and transmit, on a precise wavelength, a signal which can inform the observer of the position of the animal. When a sensor is coupled to the transmitter, information on the measured variable (depth, temperature, activity) can be deduced from variations in the signal pulse rate. Data storage tags, on the other hand, do not transmit a signal, but record data in an internal memory. These data are only available for reading and decoding after the animal has been recaptured. Photo 2 shows telemetry tags.
15. Telemetry transmitters and data storage tags can be fixed externally, inserted inside the stomach or inserted into the peritoneal cavity either by surgical operation or via the gonoduct. Each method has its intrinsic, environmental or specific advantages and disadvantages, and these must be taken into account when a study is established. Intragastric insertion is the least invasive method but can affect feeding behaviour. External attachment is usually adequate in the short term but causes problems in the longer term especially in running water. Intraperitoneal implantation is more invasive, as it generally requires surgery, but once the abdominal wall is healed, it is the safest method in terms of both the study success and fish health. Insertion into the peritoneal cavity through the gonoduct is only possible in some species. Because of the increased lifespan of telemetry transmitters and data storage tags, intraperitoneal insertion has become the most popular attachment method. The success or failure of this method often depends directly on the adoption of appropriate prophylactic and therapeutic measures, the choice of the incision site determined on the basis of histopathologic criteria and the method used to close the wound (suture, adhesives, cyanocrylates, clips) taking into account healing dynamics in the particular species. These different aspects have been tested on a wide range of fish species from temperate and tropical regions. Photos 4 and 5 show a tagged tilapia.
16. The second seminar reviewed biases associated with tagging, including tag shedding. In the case of external fixing, an early and non-intentional loss may be the result of loose knots, skin erosion or cuts in the dorsal muscles made by the attachment threads, especially in fast-flowing rivers. In fish with intragastric tags, peristalsis or (mainly) regurgitation can cause rejection. The frequency of regurgitation, and the delay between tagging and tag loss, both vary considerably depending on the species and the relative tag size. Tag expulsion can also occur for tags implanted into the peritoneal cavity, either through the incision, through an intact area in the abdominal wall, or, in some species, through the intestine. The different techniques and procedures which allow rejection probability to be minimized were analysed.
17. In discussing the impact of tags on fish physiology and behaviour, stress was laid on the fact that it was indispensable to take into account the characteristics of the species being studied, especially the characteristics of the swim-bladder (physostomatous and physoclistic species). Biases specific to external, intragastric and intraperitoneal attachment methods were also discussed. The risk of transposing conclusions from one species to another was stressed together with the need for feasibility studies in the case of species for which no information is available.
18. Two seminars focused on the propagation of radio and acoustic waves in the aquatic environment. Their main objective was for participants to be able to undertake the calculations required to define study feasibility and the best strategy to adopt.
19. Radio signals (30-170 MHz) have omnidirectional propagation in the aquatic environment, but only wave vectors at an angle less than 6° to the perpendicular at the water-air interface can cross this interface and disperse in the air where they can be detected by an aerial antenna connected to a receiving station. The signal can only be detected if the accumulated damping during signal propagation has not reduced the signal/noise ratio below the receiver’s threshold sensitivity level. The various damping factors (related to propagation in the water, at the water-air interface, in the air, in the receiving antenna and in the cabling) were reviewed. This review included equations which can be used to model damping as a function of environmental characteristics (such as depth, conductivity, transmission frequency, etc.). The main restrictions on the propagation of radio waves are the depth of transmission and increasing water conductivity. Thanks to these mathematical foundations, it became possible for participants to calculate the minimum energy required by transmitters in order to set up a study in different environments or else to adapt their tracking strategy to particular transmitters and to environmental conditions. Participants were able, in a practical session, to apply their theoretical knowledge of radio wavelength propagation in the definition of research projects.
20. When considering the choice of frequency bands in planning a project using radiotelemetry, it is strongly recommended to study carefully the existing possibilities. Of course, first it will be necessary to find out what bands are legally available in the country or countries of the project.
21. The second seminar on propagation dealt with acoustic wavelengths (30-80kHz). In acoustic telemetry, the transmission frequency is inversely proportional to the diameter of the piezo-electric transducer, so that tagging of small fish can only be done with high frequency transmitters and these suffer more damping in their propagation in an aquatic environment. The speed with which sound is propagated varies according to salinity and temperature. Any interface between environments with different propagation speeds gives rise to reflection and refraction of the signals so that thermoclines, haloclines and immersed macrophytes are major barriers to the propagation of acoustic waves. It is also for this reason that acoustic signals absolutely must be detected by an immersed hydrophone. As with radio waves, the damping factors of acoustic waves were reviewed and modeled as a function of environmental conditions. In a practical session, participants used these mathematical models to decide whether studies using acoustic biotelemetry were feasible in particular environments.
22. The next seminar concerned the problem of the positioning of the signal-transmitting source by an operator. To this end, reception diagrams of directional and omnidirectional antennae or hydrophones were reviewed and compared with associated receiving strengths. The intrinsic performance of directional systems, in terms of opening angle and reception gain, was analysed according to the context of the biotelemetric study, in the case of both automatic passive monitoring stations and active tracking by boat or vehicle. The participants were then able to determine objectively several key elements of the study strategy in a practical session, in particular: the type of antenna (dipole, loop, H-Adcock, Yagi) best adapted to the environment; the number of frequencies that could be scanned by a receiving station and the speed at which a motorized team could progress without running the risk of not detecting the signal. Should environmental conditions become too restrictive for one of the variables concerned, technical alternatives were considered, particularly receiving stations linked to antenna multiplexors (or switch boxes) and the use of coded transmitters operating on a similar carrier frequency.
23. The accuracy of positioning by triangulation is a matter of distance between the receiver and the transmitter, and of angle between the different bearings that serve to localize the transmitter by triangulation. Depending on the study context, positioning can be worked out from landmarks whose coordinates are known (e.g. area markers) or precise points whose coordinates are determined in situ (e.g. GPS). Some software utilities, specifically developed for radio-tracking applications, were presented and their mathematical principles set out so that they could be used in the practical sessions. Other alternatives were considered for the positioning of mixed (radio and acoustic) transmitters, and for the automatic positioning of acoustic transmitters by an array of omnidirectional hydrophones. In the latter case, positioning is achieved by an algorithm which deduces the position of the transmitter from the respective times of arrival of the signal at the hydrophones of the array (the inverse principle of hyperbolic navigation). With the combination of angular, metric and temporal information, the observer can determine the dimensions of the error polygon around the true position of the transmitter, this element being crucial to determine the collection strategy for cartographic and habitat data.
24. Fish locations are later positioned on habitat maps, using Cartesian coordinates. The home range or daily activity area can thus be calculated by one of the following methods: minimum convex polygon, bivariate statistics (i.e. ellipses with an N% confidence interval) or squares on gridded maps. The precision of the localization and the structure of the database (binormality constraint for ellipses) dictate the choice of method. It is also crucial, within the framework of this analysis, that each location of fish by telemetry is equally representative in time and therefore the data must be collected at regular intervals. The choice of this interval depends on the biology of the species being studied and the study logistics. In practice, it is recommended to use daily localization as a basis, complemented by 24-hour cycles wherein fish activity is described as often as possible, but still on a regular basis. Once this database is available, it is possible, via subsampling, to determine the importance of information loss through the increase in time intervals between successive observations. This loss can be compared with the budgets needed to undertake studies with different localization frequencies and hence a project rationale can be determined. This procedure can also be applied to data collection during a daily cycle. It is also very useful when it comes to programming data storage tags, programmable telemetry transmitters and automatic listening stations as it is a good way to optimize the tag lifespan with regard to the useful frequency of data collection. Photos 6 and 7 show trainee teams detecting signals.
