After reminding the group of its Terms of Reference, David Farmer chaired a discussion about the technological needs of the CoML Pilot Projects. This was followed by a discussion of possible synergies between the various technologies.

ChEss
The concept of this project (Biogeography of Chemosynthetic Ecosystems) was to explore the fauna and flora of hydrothermal vents in the North Atlantic using chemical sensors fitted to an autonomous underwater vehicle. This was cutting edge technology and there was some discussion about the availability, reliability and operational capability of both sensors and vehicles. Although AUVs were being used in survey work and costs had been calculated, application to oceanography was lagging and quite a lot more work was required before they became standard tools. Autosub was operational, but most other AUVs were not and the autonomous capability needed further development; steering and endurance were both issues. The operating costs of REMUS, which carried a sensor for bioluminescence, were reasonable, but it required support from a research vessel. Because they were designed for minimal weight and long life, gliding AUVs carried few sensors and there would be problems in trying to add more. A hovering AUV was desirable but was not available. The recent development by Al Hanson at the Graduate School of Oceanography, URI (akhanson@gso.uri.edu; http://www.subchem.com) of a sensor that could detect very low concentrations in real time and follow a chemical gradient was, however, close to satisfying the needs of a 'sniffing' AUV. In general, however, sensor development was lagging.

top of page

NaGISA
This project (Natural Geography in Shore Areas) was described as the Coastal Survey of the Western Pacific (part of DIWPA) at the previous meeting of WG 118 in Mar del Plata in 2001. Nagisa is now funded (by the Sloan Foundation and Japan) and is already underway, although further participants are sought, for example in South America. The plan involves developing both skills and infrastructure. A system for taking bottom samples and recording videos is in place, but the identification of meiobenthos still presents a major challenge. Several possible solutions were identified. One option would be for roaming molecular biology laboratories to visit archives of samples and undertake data gathering and training exercises. Another would be for roaming taxonomists to visit the various observing sites to train local staff, who would subsequently be able to get supplementary help by exchanging electronic images with experts back in their home laboratories. This approach would tie in well with the NSF PEET programme (http://web.nhm.ku.edu/peet/) in the USA, which requires experts to train 5 students.

NaGISA is likely to have both sorting and archiving problems, given the wide range of material (macrophytes to meiobenthos) that it proposes to collect and the lack of basic facilities (electricity and microscopes) in some areas. A basic data collection protocol based on digital images and sequences from stereoscopy and mosaics (underwater archaeology) offers the prospect of cheap storage in minimal space. Physical samples might perhaps be archived at regional nodes. But, regardless of the location of the archive, sub-samples must be preserved for subsequent genetic analysis. This is a simple process, which entails fixing material quickly in formalin and then transferring it to ethanol. Although it is possible to travel with amplified DNA, local archives will probably increase in importance as it becomes progressively harder to ship samples, because of CITES regulations or security precautions.

top of page

GoM
At the previous meeting of the WG in Mar del Plata, Ken Foote had identified four challenges for the development and application of acoustics for use during CoML. The first two were to make good acoustic measurements and quantitative biological measurements; the third was to classify and identify acoustic targets. The fourth challenge was to extend the range of observation of optical instruments by integrating optical and acoustic technology. It was quickly apparent in discussion that, although no new technology could be recommended to GoM (Census of Marine Life in the Gulf of Maine) in the short term, there was considerable scope for acoustic species identification over the next 5-10 years. At high latitudes with a simple ecosystem, it was already feasible to distinguish between three dominant species of fish. In the tropics, however, with 2-300 species the problem was rather different. Here it was necessary to widen the scope of diagnostic features to include both the schooling behaviour of the fish and their behaviour in response to environmental factors, such as tides. Multi-beam sonars offered exciting prospects, although the instruments currently made by Reson and Simrad, which could be used in a towed body as well as from a ship, were expensive and limited in range by their operating frequency. Decreasing this from 200 kHz to 80 kHz would give significant improvements.

Progress with acoustic identification in tropical regions, which would inevitably be slow, if tackled by individual research institutes, would be much faster and better co-ordinated if tackled by regional centres of excellence, which should be favourably regarded by international funding agencies. Molecular biology, optical technology and taxonomy would be other obvious candidates for centres of excellence, each with its own range of expertise and matching technology. Although it would still be necessary to provide technology appropriate to local problems, the South American participants agreed that there was strong support for international collaboration between their countries. They welcomed the proposal for a system of complementary centres of excellence, which they saw as a means of obtaining advanced technology and improving the infrastructure for marine research. At present, whilst some countries (e.g. Mexico) had good research vessels but poor technology, others had neither the vessels nor the technology. Mariano Gutiérrez Torero agreed to set up a sub-group to discuss possibilities for centres of excellence and prepare an agreed statement of needs. In addition to the World Bank, support might be forthcoming from the EU, which had apparently provided resources to single laboratories when arrangements were made to ensure collaboration with institutes in neighbouring countries. Another way of stimulating development would be for SCOR/CoML to bring international meetings to the centres of excellence. For example, Bill Karp suggested that FA,ST would welcome an invitation to hold its 2004 meeting in Peru now that country was an observer at ICES. Sponsorship from FAO might also be possible. Although Venema's retirement had removed FAO's internal driver for acoustics, FAO had recently joined the Fishing Technology and Fish Behaviour (FTFB) working group as a co-sponsor. FTFB and FA,ST both reported to the Fishing Technology Committee (FTC) at ICES.

top of page

POST
The concept underlying the Pacific Ocean Salmon Tracking project is a series of acoustic listening stations on the seabed off the west coast of Canada and the USA. Stations laid in lines across the narrow continental shelf will detect the passage of fish marked with a simple acoustic 'pinger'. Tagged salmon smolts will be detected as they pass the listening stations on their way to Aleutian Islands and the open ocean and the date and time of passage recorded. Whilst this system works well in confined bodies of water, it may not be so practical in the open sea. The effective range of the listening stations is determined by the size and power of the tag, which is inevitably limited when tags are used with small fish; there is also an inherent weakness with the blind transmission technique. There are various solutions, one of which is to use a more intelligent system, for example a transponding tag, as widely used in Europe for many years, and as discussed at the previous meeting in Mar del Plata. Another is the passive 'fish chip' recently conceptualised by Tom Rossby at the University of Rhode Island, which will record the reception time of signals from a series of fixed transmitters and track fish in the same way that oceanographers track RAFOS floats. The new chip, which is very cheap to make, has been designed and tested in the laboratory. Funds are available for further development, including a miniature hydrophone, and it is planned to test the system in the sea in 2003, using a research vessel. This system has tremendous potential for the animal tracking community and it was agreed to recommend that the PIs of the POST project should evaluate it as soon as it was possible to do so.

In response to questions from Carlos Robinson and David Mellinger about possible effects of acoustic signals on other animals, Van Holliday and David Farmer pointed out that small fish tags produce no more noise than snapping shrimps and this will merge with the background noise within a few hundred metres. RAFOS signals are different, being transmitted in code at low power and at a much lower frequency, which changes with time. Neither is likely to have much effect on other animals, however, despite the problems of public perception that there have been with the ATOC experiments.

During further discussion, direct questions from David Farmer established that, although electronic tags are not currently used in South America, there was considerable potential to use them with both fish and marine mammals. Mariano Gutiérrez Torero explained that he wanted to use pop-up satellite-detected tags with tuna and hake and David Farmer drew attention to the opportunity to record oceanographic data from diving mammals at the same time as recording their behaviour. David Mellinger commented that Bruce Mate was keen to train people in South America to use electronic tags with marine mammals. Electronic tags were also highly suitable for benthic organisms as well as fish and mammals. As had been shown in Europe, where electronic tags had been applied to fisheries investigations for several decades, it was necessary to acquire descriptive data with individuals before trying to construct testable models of population behaviour. An automated tagging system allowed large numbers of pelagic fish to be tagged with PIT (passive integrating transponders) tags, which could detected relatively cheaply by scanning the catch on board the purse seine fleet. Data could be sent ashore by radio and the technique could provide a quantitative estimate of the stock, as well as information on its distribution. After discussion, it was the agreed to recommend the uptake of electronic tags in South America.

Landers can be used to estimate the number of species and individuals in an area, using bait, a flash camera and a simple current meter, all relatively cheap and simple technology. More sophisticated systems (e.g. the AUDOS system developed by the University of Aberdeen in Scotland) are available with scanning sonar and other advanced technology. The are many opportunities to develop innovative methods of attraction and repulsion, using light, sound and other factors. For exploited species, landers can be combined with long-lines to obtain much better estimates of fish density than can be provided by the lander itself. Conceptually, too, there is no reason why landers should not be used in midwater as well as on the bottom. Because they are likely to provide new insights and support new research projects at low cost, development and application of landers was recommended as a technical area deserving support.

top of page

MAR-ECO
This project (Patterns & Processes of Ecosystems in the Northern Mid Atlantic) planned to use acoustic surveys and a variety of other technologies, including landers and longlines, to investigate the ecology of the mid-Atlantic Ridge. Olav Rune Godø explained that the first major problem was to extend the operating depth of 38 kHz echosounders to 2000 m; at present the TVG (time varied gain) only worked to 800 m. There were also problems in deciding how to get an idea of seasonal variation, how to analyse results and get the maximum benefit from photo transects of the seabed recorded by ROVs and AUVs, and how to increase range by, for example, acoustic imaging. In the subsequent discussion, Van Holliday suggested that using a sweeping system on the timer might increase the range of the TVG and David Farmer commented that seasonal changes might be revealed by looking for differences between assembled images, using military signal processing technology developed for mine hunting. Sidescan sonar with accurate positioning and good data processing was also an appropriate and proven technology. Questar Tangent and Roxann, two commercially available systems used by the fishing industry to identify the nature of the seabed, were empirical and of uncertain scientific value. A physics-based investigation of the causes of backscatter, which was currently underway and which had been reported to IEEE Oceans in recent years, was likely to produce reliable tools for investigating bottom sediments within about 10 years (see IEEE Journal of Ocean Engineering 27(3): 341-601, July 2002, Special Issue on High Frequency Sediment Acoustics; E.I. Thorsos et al., An overview of SAX99: Acoustic Measurements, IEEE J. Oceanic Engineering 26(1): 4 - 25, 2001; M.D. Richardson et al., An overview of SAX99: Environmental Considerations, IEEE J. Oceanic Engineering 26(1): 26 - 53, 2001).

These new tools would permit accurate descriptions of habitats, of the sort already required by EU governments ahead of the relevant technological developments. Changes in populations of benthos could already be recorded by sequential surveys using a sidescan sonar and good quality GPS, a point illustrated by Van Holliday, who showed tracks of dispersing animals recorded with a 100 kHz sidescan sonar in a patch of the burrowing urchin Brisaster. Emmanuel Boss pointed out that LIDAR and laser line-scan could also be used for habitat mapping and Bill Karp mentioned that scientists at the NMFS laboratory in Seattle are evaluating technologies for characterizing demersal habitat including video, sidescan sonar, multibeam sonar, and laser line-scan. Laser line-scan also holds promise for assessment of crab abundance. It was pointed out that there was a seabed group in Seattle pursuing this subject. Jorge Castillo drew the group's attention to two requirements in South America, the first to survey fish populations around sea mounts using a combination of acoustics and optics, and the second to measure the size of fish with a stereo camera. As Van Holliday pointed out, the second problem could be solved by using the camera in conjunction with two parallel laser beams, separated by a known distance.

top of page

TOPP
This aim of this project was to understand how marine animals from several trophic levels use the distinct oceanic regions in the North Pacific. Advanced electronic tags would be used to identify migration routes and critical habitats and link behaviour and distribution to oceanographic processes. Available technology included archival tags that could record high quality data for several years and pop-up tags that could transmit data via the Argos satellite system. Whilst only a proportion of archival tags were ever recovered, pop-up tags could, at present, only transmit a limited amount of the data they recorded because of technical limitations of the Argos system. This problem might be solved in future, now that the IRIDIUM system, which could transfer 100 times more data than Argos, was live again. The float and glider community was converting to IRIDIUM and the US government had purchased a block of time, which was free to all PIs in the USA. At present, however, IRIDIUM could not be recommended to TOPP, or any other tracking project with marine organisms, because the transmitters were too large.

Although not of specific relevance to TOPP, there followed a general discussion of ways of transmitting oceanographic data via the IRIDIUM system using Argo floats and gliders, whose characteristics were summarised by Emmanuel Boss. Both platforms offered exciting possibilities for biological oceanography, if the physicists could be persuaded to add the extra sensors. There were many Argo floats in use in the open ocean, costs were coming down and programmes were underway to add optical sensors (e.g. beam transmissometer). Biofouling problems could be solved with copper shutters and floats could remain at 1000 m for 10 days. Gliders contained a bladder and were able to change their buoyancy and centre of gravity with moving internal parts. They had small wings and were designed to make double-oblique dives at speeds of 20-25 cm s-1. On surfacing, they recorded their position by GPS and transferred data by IRIDIUM, or cellular phone link. Gliders could currently carry a CTD sensor and developments were underway to add sensors for measuring oxygen, fluorescence and backscatter (at two frequencies). They could maintain station for up to two weeks in a tidal regime with currents of 2 knots. Gliders were being developed in the USA by three groups, one of which was a commercial company. At present, costs were about $30-40 k.

In discussion, it was concluded that gliders had considerable potential for investigating smaller, coastal ecosystems at a much lower cost than a research vessel. They were non-intrusive, could be used for adaptive sampling and could be launched from a zodiac in shallow water. At present they were power-limited and could probably not carry a sonar. However, as Van Holliday pointed out, the power requirements of electronic devices decreased by an order of magnitude every few years. With appropriate sensor development, there was therefore considerable scope for the use of gliders over the next decade. More information is available from www.webbresearch.com using the link 'The SLOCUM glider' listed under Products and Projects.

top of page

Synergies between technologies
In the subsequent discussion, a number of synergies between the various technologies were identified. The interpretation of acoustic observations of secondary production can be assisted by rapidly profiling the water column with optical and other physical instruments on a winch-driven cable. The range-gating properties of LIDAR match those of sonar and offer excellent prospects for synergy. It can be used to map the seabed and is also an excellent search tool for identifying aggregations or other nodes of interest. This property suggests that LIDAR’s primary role in CoML is likely to be in directing sampling to best effect and especially to sub-metre scale features that now appear to be of very great importance. Greatly under-sampled, these features can contain as much as 80% of the local biomass, extend over tens of kilometres and persist for several weeks. Neither LIDAR nor acoustics will, however, take samples of the organisms making up these thin layers and this remains a challenge; a similar challenge exists in the benthos.

top of page