The ATOLL Laboratory and other Instruments Developed at Kiel

by Uwe Kils

The ATOLL Laboratory consists of three banana-shaped fiberglass hulls (Fig. 1). These 25 meter long hulls can be connected in series for transportation. At the measuring site they are switched into a horseshoe-shaped arrangement. In this operational mode the structure surrounds 150 m3 of the sea. Although the 5 meter wide hulls offer 75 m3 of laboratory space, 220 m3 of supply- and storage-facilities and 350 m3 deck area, they project only 38 cm below the surface and have a 2.8 m wide flat curved cross-section. The small submergence minimizes disturbances of the natural turbulence regime. The lagoon opening is oriented towards the sun for natural light regime. The hulls are not coated with antifouling paints and are constructed only from fiberglass, stainless steel, aluminum and wood to minimize chemical interference. The main instrumentation room (25 m3) is air-conditioned to protect the electronics. All displays and controls are centralized in a glasshouse on deck (30 m3) to give a good overview for the operators. The lab can accommodate three scientists easily. A small lecture room is available for teaching.

An underwater observation room allows for a direct inspection of the investigated scenery and control of the instrumentation via two 40 x 40 cm windows located 20 cm below the water surface. The water around these windows is accessible with scientific equipment via four portholes. Air pressure in the observation room can be increased to allow opening of the underwater portholes to change the glasses or mount equipment onto the outside of the windows without docking the lab. Five balconies give access 10 cm above sea level; four of these are sheltered. Underwater bubble curtains softly guide the fish schools into the scanned areas, or prevent their escape during disturbances. After the in situ measurements are complete, the organisms can be captured by raising a pop-up-net from below. This provides animals for taxonomic identification and for estimation of condition indices, enzymatic activity, and RNA/DNA relations.

A crane-deployed boat is available to assist in mounting instrumentation near the lab and for monitoring the nearby environment. The boats systems are connected to the laboratory computer via a radio link. Positioning and tracking of the boat is done by RADAR or SONAR from the lab. The laboratory is also equipped to support diving. Air is supplied from a light umbilical for extended and relatively quiet underwater inspections.

The central processing unit (CPU) is a CMOS industry microcontroller, communicating on a private and exclusive frequency with the mainframe of the institute. Several functions can be remotely controlled and evaluated by telephone. The CPU system performs all alarm functions as well as a "call on event" system triggering an EUROCALL beeper. The lab has been working autonomously for months with only occasional checks for retrieving tapes and disks. Processed data and selected images are transferred by radio communication. On-line data processing of the acoustics, optics and physics is done on a UNIX workstation (NeXT cube, RAM capacity 40 Mbyte) and several MOTOROLA 680XX subsystems. Data compression (Delta- and JPEG) is conducted on board and processed data are stored on rewritable optical disks. Because data inspection and image analysis can be completed shortly after data acquisition, small adjustments of the scanning setups can readily be performed. SP highband -matik and HI8 machines are used for mass storage of the raw optical and acoustical data.

Some instruments are mounted directly on the hulls of the lab; others are carried by remotely operated vehicles (ROVs). For some types of experiments, e.g., behavioral studies, the propulsion systems of the ROVs cause disruption of the natural hydrodynamic environment. We minimize the hydrodynamic disturbances by positioning the ROV using three negatively buoyant thin rubber bands and a variable buoyancy system (see Fig. 1). This operational mode allows for a quiet approach from below towards the highly evasive organisms with minimum disturbance to the natural turbulence- and light-regime.

A scanning SONAR is used to locate the juvenile herring schools, guide the ROV, and for quantifying positions, distances and speeds. Salinity, temperature and oxygen are measured with probes, water velocity with acoustics and microturbulences with optics. Plankton-, particle- and bubble-concentrations and their size distributions are measured with an optical plankton recorder (OPR) (KILS 1981, 1989), with high resolution acoustics (KILS et al. 1991), or with net- and pump-samples. Low-light cameras and high speed video cameras with shuttered LASER-sheet or infrared LEDs are used for quantification of animal behavior and for control of the experimental setup (STRICKLER et al. 1992). The ecoSCOPE (KILS 1992), an endoscope-system for non-invasive optical measurements, is used to record the microscale dynamics and behavior of the highly evasive herring. The disturbance of the microturbulences by this sensor is relatively low, and its data make possible evaluation of microstructures of the flow field. The ecoSCOPE can be mounted directly to the floating platform or can be deployed using an ROV.

For the evaluation and visualization of ocean- and bio-dynamics, dynIMAGE software has been developed (KILS 1992). First, dynIMAGE compensates for the swaying and rolling of the optics due to low-frequency microturbulences and prepares the raw data for evaluations of animal-motion and high-frequency microturbulences. Then, video-clips are reconstructed from the processed images for visualization of the fast oceanographic processes.

Investigations to date have concentrated on one of the most important food chain transitions: the linkages between the early life stages of herring (Clupea harengus) and their principal prey (copepods). A major hypotheses of fisheries ecologists is that the microdistribution of prey, the microturbulence of the ocean, or the retention conditions are normally not suited to allow strong year classes of fish to develop. In most years more than 99% of herring larvae do not survive. Occasionally however, physical and biotic conditions are favorable, larval survival is high, and large year-classes result. The aim of our investigations using the ATOLL laboratory and the instrumentation described above is to improve our understanding of the effects of small-scale dynamics on fish feeding, predator avoidance, and year-class strength.

Scientific Questions

What are the effects of the natural light gradient on predator-prey interactions? How can the predator best see the prey without being seen? How does the focussing of small waves oscillating light regime influence camouflage and attack strategy? What are the influences of the different frequencies of microturbulences? How do such effects change at the moment when herring larvae join into schools? What role does the phenomenon of aggregation play? Does ocean physics create or alter organism-aggregations? Can the dynamic of aggregations effect ocean physics at the microscales? Are there effects of the surface waves? What are the distribution and dynamics of microbubbles caused by turbulences and gas-oversaturations? How can the organisms orientate in respect to micro-gradients of the ocean physics? How do they survive in the direct vicinity of undulating anoxia and hypoxia? Why are eelpouts, sticklebacks and herrings so extremely successful in the Baltic while cod is not? What are the effects and functions of schooling for feeding and microscale-orientation? All this can best be investigated in situ.

The areas of investigation are the estuaries of the western Baltic. The drastic ecological shifts during the last decades qualify this area as an excellent experimental site for the examination of global change effects on marine ecosystems. Plankton concentrations of up to 800,000 cells Prorocentrum minimum per milliliter are a challenge for herrings searching prey under the drastically deteriorated visibility -- and a challenge for the scientists to quantify their strategies.

The laboratory has been in operation since 1982. It was a donation from the ATOLL Swimming Structure Development Company, Munchen, Kaiserplatz 8. The company is based on the ATOLL trademark and the ATOLL international patents. The BMFT Ministry of Science and Technology funded the first scientific experiments on behavioral studies in marine aquaculture. It has been run for the last four years under the VOLKSWAGEN Bio-Science-Award and by inputs from SONY, NeXT, ATARI, BP, ARD, ZDF, SAT1, RTLplus, GREENPEACE, the Ministry for Nature and Environment Schleswig-Holstein, the Kiel-Canal Administration, and some private sponsors.

Summary of Optics Developments at the IfM, Kiel:

Optical Ichthyoplankton Recorder (Schnack, Welsch, Wieland, Kils): Towed GULF III type net similar to Ortner et al. (1981) but with a video camera mounted at the net end, image and data transmission via single conductor cable, prototype employed since 1990 on herring and cod surveys in the North Sea and Baltic. Distribution, large scale long time series monitoring.

Optical Zooplankton Profiler (Lenz): Vertical towed net-system, image and data transmission (two cameras) via fiber optics, prototype planned for October 1992: Distribution, large scale long time series monitoring

ecoSCOPE (Kils): Remotely operated or free floating vehicle, image transmission via fiber optics or onboard storage, several prototypes employed in herring schools for predator prey interaction studies. Microscale behavior, microdistribution. See Figure 2.

Optical Plankton Recorder (Kils): General purpose compact instrument with optional preconcentration-nets, image and data storage onboard (1-3 cameras), employed since 1979 in anarctic krill studies, hand operated from small working boats in school studies, anchored for orientation- and ecotoxicology-studies, in aquaculture for particle-flow-studies. Mesoscale monitoring, environmental impact on behavior. (Until recently, Uwe Kils was a research scientist at the Institute fur Meereskunde, Universitat Kiel. Dr. Kils is now at Rutgers University Institute of Marine and Coastal Sciences)

More details of the cited instrumentation are described in:

Kils U (1981) Swimming Behaviour, Swimming Performance and Energy Balance of Antarctic Krill, Euphausia superba. BIOMASS Sci Ser, 3, 1-122

Kils U (1989) On the Micro-Structure of Micro-Layers -- Results of an in situ Zooplankton-Counter. Coun Meet Int Coun Explor Sea 1989/L15:1-4

Kils U, Ruohonen K., Makinen T (1991) Daily feed intake estimates for rainbow trout (Oncorhynchus mykiss Wahlbaum) evaluated with SONAR and X-ray techniques at commercial net cage farms. Coun Meet Int Coun Explor Sea 1991/F3:1-8

Kils U (1992) The ecoSCOPE and dynIMAGE: microscale tools for in situ studies of predator prey interactions. Int Rev gesamten Hydrobiol.

Strickler R, Schulz P, Bergstroem B, Berman M, Donoghay P, Gallager S, Haney J, Hargreaves B, Kils U, Paffenhofer G, Richman S, Vanderploeg H, Welsch W, Wethey D, Yen J (1992) Video based instruments for in situ studies of zooplankton abundance, distribution and behavior. Arch Hydrobiol.


homepage contents previous article next article