|
|
C. Krembs, R. Gradinger, M. Spindler, A. Goehl (1996): New instruments for current and diffusion measurements at the sea ice-water interface: instrumental design and first results from field measurements in the Arctic Ocean, Oceanology, International 96, conference proceeding vol. 3, p. 95-116.
|
|
|
Introduction
The Arctic Mediterranean sea is covered by pack ice all year round (Maykut 1985). Sea ice in polar regions comprises a unique habitat for a variety of organisms. The sea ice-water interface is recognised as a biologically active interface (Alexander and Chapman 1981) which serves as substrate for a population of ice algae. Pennate diatoms are important primary producers in the ice (Poulin 1990). Diatoms serve as food for higher trophic levels. The ice is colonised by a specially adapted community which can be clearly separated from the community living in the water below. Sea ice is also inhabited by diverse populations of bacteria, fungi, proto- and metazoa (e.g. Horner 1985, 1990) which live in a channel system between the ice crystals. The habitat within sea ice is a branched network of brine channels which develops during freezing. Brine channels have a mean diameter of approximately 200 to 400 µm and can be highly interconnected in certain types of ice. Resin casts of sea ice (Weissenberger et al. 1992) illustrated their complex three dimensional structure. The contribution of brine channels to total ice volume ranges between 1-30 % (Frankenstein and Garner 1967) depending on the initial salinity in the water prior to freezing and the temperature of the ice. Sea ice poses various requirements onto the organisms by its complicated 3-D structure and its constantly changing character.
The distribution of organisms within the ice is not homogeneous. Physico-chemical parameters produce steep temporal and spatial gradients that structure the habitat (Eicken 1992). High concentrations of organisms were found in various horizons in the ice with highest abundances in the lowermost cm in Arctic sea ice (Horner 1985). Spatial variability in the concentration of organisms are important in the understanding of the sea ice community. High horizontal differences in concentration of one order of magnitude were reported on scales of 30 cm (Spindler and Dieckmann 1986) for Antarctic sea ice. It is speculated that light, difference in texture and modi of ice formation play a major role for such variabilities (Cota 1989, Eicken 1992).
Transport of water and matter has an important impact on the growth of algae since both provide inorganic nutrients to the plants (Cota et al. 1987). Physical processes permit flushing of sea ice at the interface. About the mechanisms of matter exchange several opinions exist. Meguro et al. (1967) described three sources for nutrients for ice algae. He suggested in situ regeneration of organic material, nutrient import via desalination processes from the upper parts of the ice and a directed nutrient transport from the water column into the ice. Cota et al. (1987) investigated the relative significance of these mechanisms in a region with strong tidal currents. They found that the nutrient requirements of ice algae could be covered by the exchange of water with the underlying water column. Cota (1985) measured a 65% increase in the concentration of chlorophyll in sea ice that was exposed to higher current velocities.
Propagating waves under the ice also can contribute significantly to the water exchange between ice and water (Ackley et al. 1987, Ackermann et al. 1990, Shen and Ackermann 1990).
For natural sea ice the range of fluid motion inside the brine channels is not well documented. Theoretical estimates showed that brine transport should be laminar (Gradinger et al. 1992). Brine can oscillate (Eide and Martin 1975) or move bi- directional (Niedrauer and Martin 1979). Reeburgh (1984) calculated a very short residence time for the liquid phase in the skeletal layer at the ice-water interface. He compared measurements in the literature with estimated brine channel fluxes using rates of nutrient uptake by ice algae and adjacent sea water nutrient concentrations. He concluded that fluid motion in growing ice may be important in maintaining the ice algae community.
Measurements that determine the spatial variability of water exchange between ice and the water below are still lacking. To what degree a heterogeneous distribution of algae can be explained by such a variable exchange of water across the ice- water interface is unknown.
To determine the degree and spatial variability of exchange processes across the ice-water interface methods are required that can combine elements of ice structure, brine motion, pressure oscillation under the ice, current velocities and current profiles, with the distribution and concentration of important groups of ice organisms.
Field experiments on, in and under the ice floe are hampered by climatic difficulties in polar regions. Partial equipment failure is common for expeditions. Handling and long term deployment of instruments is very difficult. Changes in salinity and freezing of water can influence the operation of instruments that are in continuous contact with water in the field. Special simple instrumental designs that are reliable and robust constructed and can be used in combination with other techniques are needed for scientists operating in remote polar areas.
Statement of theory and definitions
Polar research poses restrains on the equipment and instruments that can be used. They have to be robust to withstand cold temperatures and rough handling. Instruments have to be fairly light and should be easy to operate and easy to repair. Unsteady electricity supply (AC and DC) can be sporadic and under severe climatic conditions are often prone to fail. Prolonged contact with salt and water wears on the materials and instruments that are immersed in water often freeze into the ice. To conduct experiments below and in the ice instruments have to fit through a drilled ice hole (Ø 12 cm) commonly used by ice researches.
Current and direction sensor:
When instruments are deployed under the ice to measure the exchange of nutrients across the ice-water interface, the natural flow of water is destroyed. To re-establish the natural hydrological conditions with technical instruments the direction and character of the water flow below the ice have to be determined. We used a Savonius rotor to measure the current velocity under the ice.
In our construction all electrical parts were kept to a minimum and are installed above the waterline. The instrument was designed to be alternatively used even under complete power failure. The current meter was equipped with an inflatable tube that fixed the current meter in the hole of the ice floe and enabled the probe to float in case of pack ice deterioration.
The design of the current meter was extended to fulfil additional requirements. We wanted to combine water current velocity measurement with simultaneous tracer studies at different horizons in the ice. To fulfil these tasks an inflatable sleeve system was constructed that prevents sea water from percolating laterally through the hole of the ice floe. A sampling port for water and a temperature sensor at the level of the current sensor head were additionally installed.
Incubation tank:
Within the incubation tank the ambient hydrological conditions can be simulated and allowing to conduct experiments of dye transport across the ice-water interface under in situ conditions.
Exchange of dissolved nutrients across the ice-water interface and transport within the ice via brine motion are important for organisms living in the ice. To determine whether the variability of organism abundance was coupled to the variability of water advection, dye tracers were brought out at the ice-water interface. Waves propagate under the ice and currents constantly disperse and replace water from the interface. To maintain a defined parcel of water containing dye of a known concentration at a fixed place under the ice without destroying the flow field and pressure oscillations at that particular site, we used an incubation tank that could be placed at the underside of the ice. A flexible bottom allowed pressure waves to be transmitted into the interior of the tank. The water in the interior of the tank communicated freely with the ice above. A propeller that could be adjusted in revolution created an even, 1 m long flow field along the ice-water interface in the interior of the tank.
The underside of an ice floe is not always
flat. Small bumps and depressions are common (Jezek et al. 1990). In order
to seal off the tank against the sea ice, the walls of the tank had to be
flexible, and rigid enough to maintain a defined shape. As for the current
meter the core hole needed to be sealed to prevent penetration and circulation
of water into the ice. The tank required a minimum size of 100 x 30 x 14 cm
and was to be placed and retrieved from the position under the ice floe through
a Ø 12 cm ice hole. This demanded that the hole had to maintain its
initial diameter of 12 cm for several hours during the entire incubation.
Description of equipment and processes
Current sensor:
The current sensor enables the measurement of water temperature as well as velocity and direction of the water current below the ice-water interface. Three main parts can be defined: the probe, the readout unit and the deployment system.
Probe:
The probe is divided into 5 compartments (Fig. 1) by 4 disks (17), a headpiece (14) and a base (20), all having a diameter of 100 mm. The parts are assembled on three stainless steel tubes (19) and determine the total diameter of the instrument leaving a 10 mm clearance to the wall of the ice hole.
Each compartment is used by a separate component: (top to bottom) a temperature sensor (digital thermometer accuracy +/- 0.5 °C, Roth) (6), a code disk (9), a vane (10), a Savonius rotor (11) and a second code disk (13). Additionally a PVC tube (7) with an internal diameter of 5 mm ends in the top compartment and enables the measurement of pressure changes introduced by propagating waves as well as sampling of water from the ice surface. The vane and the Savonius rotor are suspended on two stainless steel spindles (Ø 6 mm) that are held by Teflon bearings. An encoded acrylic code disk is mounted to each spindle in the compartment adjacent to the rotor/vane. The compartments are surrounded by bars (15) to protect the fragile disks from floating matter (e.g. blocks of ice). The Savonius rotor is manufactured from sections of (Ø 20 x 1 mm) PVC tubing that are fitted between two round PVC plates (Ø 64 x 2 mm). The design is based on existing instruments and reduced in size to allow for a clearance of 8 mm to the stainless steel tubing (11). The revolution speed of the rotor and the relative position of the vane are transmitted to the readout unit by 8 pairs of optical fibres made of acrylic (PMMA). Each fibre has a diameter of 1 mm and is coated with polythene for protection. With a clearance of 1 mm the code disks spin freely between pairs of optical fibres, their polished face being located opposite one another. The translucent disks are partially blanked and can interrupt the optical paths proportional to the revolution of the spindle. Two pairs of optical fibres detect the motion of the Savonius rotor. The disk (13) has two tracks creating 2 and 32 evenly long interruptions of the light beam per revolution. The remaining 6 pairs of fibres intersect 6 tracks on the second disk (9) to determine the relative position of the vane. The disk is blanked with a 6 bit binary code offering 64 values.
Readout unit
The readout unit consists of a waterproof case made of polythene (14 x 16 x 5 mm). It houses a 6 V halogen bulb and a battery pack to supply light for the optical system, a display field for 7 optical fibres, a frequency meter (Conrad Electronic GmbH, D-Hirschau) and a electronic temperature meter (digital thermometer accuracy +/-0.5 °C, Roth). Of each 8 pairs of fibres one fibre is fed by the halogen bulb or alternatively by sunlight. The returning fibre that reads the motion of the Savonius rotor with 32 strobes per revolution is fed to the frequency meter. The meter is powered by a 9 V block battery and counts the amount of strobes over a period of 0.4 sec. The amount of revolutions per minute is shown on an electronic display. The accuracy of the instrument is +/- 2%. The other fibre reading the rotor motion with 2 strobes per revolution has a polished face and terminates in the display field. Here it can be observed by eye in case the frequency meter should fail. With two blanks per revolution a current speed of up to 1 m s-1 can still be identified. Six fibres feed back from the vane thus delivering binary information on the relative vane position. The six bit binary value can be converted to degrees by calculation. After measuring the absolute orientation of the entire sensor (e.g. with compass or GPS navigation) the absolute vane position can be determined with an accuracy of +/-5.6°.
Deployment system
The deployment system consists of three major components: a rod (Fig. 1 (2)) to which the probe is attached, a pipe (4) serving as a guiding shaft for the rod and an inflatable sleeve (5). Both rod and pipe are made of glass reinforced phenol resin. A fabric coated with Neoprene inside and Hypalone outside is used as material for the sleeve. By compressing air into the volume between the pipe (4) and the sleeve (5) the sleeve can be inflated to a diameter of 125 mm and a pressure of 0.3 bar. This enables the instrument to seal itself against the entire wall of the ice hole. In this way the instrument can be located in the desired position with the lower end of the pipe to be flush with the ice-water interface. The inflated sleeve additionally prevents freshwater or sea water from percolating into the ice hole and performing any vertical exchange. This allows experiments with dye-tracers to be carried out (see diffusion rings). The sleeve is inflated utilising a hand pump. On an external barometer the pressure can be monitored. Due to it's smooth texture Hypalone does not have the tendency of freezing to the ice. Even after long exposure to in situ conditions the instrument can be retrieved as easily as it was inserted into the hole when the sleeve is deflated. The rod slides inside a 3.5 m long pipe and is guided by 12 nylon bushes (1). The inside of the pipe fills with sea water of high salinity. Therefore the bushes do not freeze to the pipe during arctic summer. As the rod is 5.5 m long, assuming a ice thickness of typically about 3 m, the probe can be slid to a depth of 2 m below the interface. Three threaded nylon pins at the top end of the pipe locate the rod in any desired horizontal position. The instrument is almost entirely made of plastics and weighs 6 kg. When inflated it entraps 20 litres of air which additionally allows it to float.
Diffusion rings:
Several diffusion rings can be clamped to the pipe of the current meter or the incubation tank to carry out further experiments using dye-tracers (Fig. 2(1)). The rings are made of nylon and are fastened outside the sleeve (Fig. 2, (2)) allowing air within the sleeve to pass. Two silicone washers (Fig. 2, (1)) with a diameter of 127 mm and a thickness of 3 mm seal off a defined volume between the sleeve and the ice. Through one of two PVC tubes (Ø 3 mm) dye tracers can be introduced into the ice (Fig. 2, (3)) while at the same time water is drawn out from the second tube (Fig. 2, (4)) allowing the dye to distribute itself evenly inside the ring.
Incubation tank:
The core structure of the incubation tank is a glass reinforced phenol resin pipe (Fig. 3, (12)) with a diameter of 76 mm and a length of 3.5 m. Attached to it are the following main parts: a head piece (Fig. 3), an inflatable sleeve (Fig. 3, (9)), a base (Fig. 4, (9)), a deployment mechanism (Fig. 5) and a incubation tank (Fig. 6).
Head piece:
A traction mechanism which installs the incubation tank in position under the ice can be driven by rotating an aluminium hand wheel turning two gears (Fig. 3, (3)). Two pins (Fig. 3, (4)) can be slid into the shafts in order to locate the deployment mechanism in the desired position. The gears drive two chains that are attached to stainless steel cables (Ø 5 mm) looping through the lower part of the instrument. An electric motor (220 V, 800 W, 2700-10,000 rpm.) can be attached to the head piece. Via a dog clutch (Fig. 3, (2)) it connects to the spindle (Fig. 3, (8)) driving a propeller pump in the incubation tank.
Inflatable Sleeve:
Like the current meter the instrument has an inflatable sleeve (Neoprene and Hypalone coated fabric) that can be pressurised to 0.3 bar and a diameter of 125 mm. This enables the instrument to seal itself against the entire wall of the ice hole locating it horizontally. Additionally the inflated sleeve prevents freshwater or sea water from percolating into the ice hole. The pressure can be monitored on an external barometer. The incubation tank can be retrieved easily from the ice hole as soon as the sleeve is deflated. After long exposure to in situ conditions the lower opening of the ice hole may freeze and cause problems in retrieving the instrument. During such conditions hot water can be poured down the main pipe (Fig. 3, (12)). Several openings at the lower end of the pipe (Fig. 4, (16)) allow to flush and melt off any build-up of ice.
Base:
The base houses a propeller pump that produces a continuous stream of water inside the incubation tank. A two bladed brass propeller (Fig. 4, (10)) is mounted to the motor driven spindle (Fig. 4, (2)). On the entire length (3.5 m) the spindle is suspended by 8 stainless steel ball bearings. A stainless steel pipe (Ø 25 mm) encases the spindle and the bearings to provide alignment and protection. The flow rate of the pump can be adjusted by altering the rotation speed of the motor. In order to reduce the initial momentum on the spindle and to permit a smooth commence of the current in the tank an electronic circuit accelerates the motor gradually (2 sec). The motor was not integrated into the base nor was the water led to an external pump to avoid the influence of temperature change of the enclosed water.
Deployment mechanism:
The stages of the deployment of the incubation tank are illustrated in Fig. 5. The components of the mechanism are: a vertical pipe (Fig. 6, (12)), a supporting rod (Fig. 6, (14)) and a flap mechanism (Fig. 6, (13)) on which the incubation tank is mounted. The vertical pipe has a diameter of 25 mm, the supporting rod and the flap mechanism are constructed from rods with 10 mm diameters. The pipe and the rods are made from glass reinforced phenol resin. A nylon block slides along the slotted vertical pipe (Fig. 6, (12)) and is tracked by the steel cable looping down from the head piece (Fig. 4 (3)). With two pivoting points (Fig. 6, (14)) the supporting rod is connected to the nylon block on one end and the flap mechanism on the other. As the block is drawn up along the vertical pipe (Fig. 6, (12)) the supporting rod forces the flap mechanism (Fig. 6, (13)) into a horizontal position. Once this motion is completed the flap mechanism can be unfolded by the second steel cable. This spreads the incubation tank to a full width of 300 mm ready to be inflated. As long as the incubation tank is not unfolded the water drag acting on it is very little. Current velocities are lower at the ice-water interface due to boundary layer condition. Here the tank can be unfolded safely. When the instrument is retrieved the deployment system is operated in reversed sequence (Fig. 5). The flap mechanism folds itself reducing the deflated incubation tank in size. The supporting rod pulls the flap mechanism down to a near vertical position. At this point the instrument can be pulled out of the ice hole. As the flap mechanism folds downward it does not obstruct the retrieval procedure but folds itself thoroughly as the tank is by being drawn into the opening of the ice hole. This is an important feature as the instrument can be retrieved even if the traction mechanism should fail. The deployment mechanism simply collapses into folded position.
Incubation tank:
Two inflatable pontoons (Neoprene and Hypalone coated fabric) (Fig. 7, (3)) of 1 m length and a diameter of 140 mm are mounted on the flap mechanism (Fig. 7, (5)) and can be inflated by a hand pump from the ice surface. Each has a volume of 16 litres and can be pressurised to 0.3 bar. The pressure is monitored by an external barometer. Inflating the pontoons gives the incubation tank it's stiff shape and a size of 1000 x 300 x 140 mm (volume 42 litres). As soon as ambient water pressure is reached the buoyancy of the entire instrument becomes positive lifting the instrument up until the flexible pontoons seal themselves against the underside of the ice. At this stage the sleeve will be inflated. When the incubation tank expands only water from the ice-water interface enters the incubation tank maintaining water in the incubation tank of the interface. This is important since steep temperature and salinity gradients sometimes prevail under melting ice. The incubation tank itself is manufactured from a flexible silicone film of 0.5 mm thickness. The film encloses the volume between the pontoons being only open towards the ice. The two ends of the tank are sealed by two lips firmly pressing against the ice (Fig. 4, (15) and Fig. 7, (2) and (10)). An additional layer of film (Fig. 7, (11)) suspended between the pontoons forms a middle floor and divides the incubation tank in an upper and a lower compartment. The propeller pumps water through flexible openings (Fig. 4, (11) and (12), Fig. 7, (8) and (9)) that connect the base (Fig. 4, (9)) to the incubation tank. The flexible outlet permits a free water flow even if the incubation tank is inclined due to a sloped ice underside. Water leaving the opening passes through a silicone film with (Ø 5 mm) pores (Fig. 4, (13)) to stabilise the flow. Water then circulates to the opposing end of the tank in direct contact with the ice. Through a slot (Fig. 6, (8)) at the end of the upper compartment the water reaches into the lower compartment were it is sucked to the intake of the pump closing the circulation. The bottom of the construction is flexible so that propagating waves can be transmited into the interior of the tank. Dye can be introduced into the interior of the tank and water can be sampled via two PVC tubes (Ø 3 mm) (Fig. 4 (1)) leading into the outflow region of the pump.
Application of equipment and processes
The current meter was calibrated against known current velocities in sections of two rivers. Current velocities were determined prior to calibration through single particle trajectories. Both rivers were wider than 4 meters (depth > 1m) and floated evenly. The width and depth of the rivers prevented a deformation of the pressure drag in front of the current meter that otherwise occurs in little flow tanks. The current meter has a stall speed of 0.06 m s-1 (Fig. 8). Its signal output is linear between 0.06-1 m s-1. The reading of the current direction during the calibration did not oscillate and pointed always downstream.
Within the incubation tank the ambient current velocity can be simulated. The incubation tank was immersed in a water tank for calibration of the current velocity and profile inside the tank. A sheet of acrylic glass simulated an ice cover against which the tank was pressed. Through the acrylic glass the current could be observed. To quantify the flow in the tank neutrally buoyant particles (approx. Ø 1 mm) were added to the water. Current velocities were determined from single particle trajectories passing a defined distance inside the tank. To visualise the current profile ink was added to the water through the opening of an auxiliary tubing (Fig. 4). The ink front moved similarly fast within the tank with a slight faster tendency at one side. The current was turbulent directed which resembles closest the natural situation. The velocity inside the incubation tank was set to 0.03 m s-1 which closest resembled averaged current profiles for all 19 field stations (Fig. 9).
Presentation of data and results
Data were obtained during the Arctic expedition ARK XI/1 (July 7 to September 20 1995) of the RV Polarstern into the Laptev Sea between 75° to 82° North and 95° to 150° East. Ice cores were drilled with gasoline engine driven CRRELL ice auger. A sketch of the combined equipment on the ice during ARK XI/1 is shown in Fig. 10. Electricity for the incubation tank was provided by a diesel generator. Both constructions were simple and operated well during the entire expedition.
Strategy of sampling:
Current profiles and directions under ice floes were measured with a vertical resolution of 10 cm to a depth of 183 cm below the interface. Velocities were averaged from 1 min measuring intervals. Profiles were measured on 19 stations. During its operation the new inflatable sleeve system was tested. The current meter was equipped with two diffusion rings and left in the ice (1.30 m thickness) during the formation of new ice and air temperatures below -3°C. After 24 hours the instrument was operating well and could be easily retrieved.
Horizontal diffusion of fluorescent dye tracers in the ice were determined with the current meter that allowed to incubate the ice with tracers in defined strata. To investigate the lateral transport of dissolved nutrients in the ice the current meter was used to position two tracers (Uranin, Rhodaminchloride) at two different levels in the ice. Along a transect holes which penetrated 1 m into the ice (sack holes) were drilled. After 2 h intervals subsamples from the infiltrated brine were taken. The tracer concentration was determined fluorometrically.
Spatial variability of water exchange between the water and the ice in the lowermost meter of the ice floe was determined with the incubation tank. The tank was installed at the underside of the ice floe through a 12 cm core hole. The vertical exchange of dye across the ice-water interface and further transport in the ice, was determined from segments of melted ice cores. The current velocity inside the tank was adjusted to ambient velocities. Dye (Rhodaminchloride) was introduced through the auxiliary tubing into the interior of the tank. The ice was incubated for 2.5 hours during which natural pressure waves outside the tank freely communicated across the flexible bottom with the interior of the tank. After incubation the tank was removed from the ice and cores were drilled from the ice. Ice cores were recovered and divided into equal 10 cm segments and melted.
To determine the distribution of algae inside the ice and their coupling with the horizontal exchange of water between the water and the ice standard methods of marine biologists were used (Gradinger et al. 1992). Photopigments (Chl a and Phaeophytin) were used as indices for algal concentrations. The remaining sample was analysed fluorometrically (Perkin Elmer Luminiscence Spectrometer LS 50B; Excitation 440 Emission 540 nm) to determine the tracer concentration in the melted ice. Subsamples were used to determine salinity concentrations in the water. Zones of elevated tracer and salinity concentrations point to areas in the ice which provide good nutrient supply to the algae.
Results:
Current velocities in the first 2 m under the ice varied on all stations between <0.06 and 0.25 m s-1. Current profiles showed turbulent components which can be seen from the large standard deviations of mean current velocities at various depths (Fig. 11). Current directions also strayed in relation to depth between 0-90°.
Preliminary data on the horizontal diffusion of fluorescent dye tracers in the ice indicate that lateral transport of brine was on the order 40 cm h-1 which roughly equals preliminary results found by Weissenberger (1994) for the identical oceanic region.
An ice thickness profile and the vertical temperature and salinity gradient during the incubation of ice with dye tracer is shown in Fig. 12. The distribution of algae inside the ice and their relation to the horizontal exchange of water between the water and the ice can be seen by the vertical profiles of tracer, Chl a, Phaeophytin, and salinity (Fig. 13). The horizontal and vertical varying intensity of all four parameters point to areas for optimal algae growth inside the ice. The tight coupling of all four parameters illustrates their significance for improved algal growth in the ice.
Interpretation of data:
Conclusions
Both instruments in combination offered a versatile tool to study exchange processes at the ice-water interface under in situ conditions. The collected data met well our requirements. The current meter was sensitive enough to detect current velocities down do 0.06 m s-1 which was sufficient for our purposes but could be improved in the future. The temperature measurement was not sensitive enough. We used a Microprocessor Conductivity Meter LF 196 which we deployed through a drilled ice hole next to the meter, instead. A more sensitive sensor should be used in the future. The additional water-sampling option could be used very comfortably and allowed us to sample defined layers of water below the ice (data were not presented). The design of the two instruments with the inflatable sleeve-system fix them firmly in the ice. The design offers a multifold improvement. Instruments are well positioned but no longer froze into the ice. In addition they are neutrally buoyant and could be combined successfully with other techniques to gain further insights into processes operating in, under and at the ice. The incubation tank offered a unique approach to determine the variability of water exchange across the ice-water interface under in situ conditions. The retrieval of the tank was always easy. Both instruments are prototypes that were a little long. We suggest to construct the shaft system such that it can be disassembled into shorter units in the future. However their light weight was very advantageous so that one person could operate both instruments alone.
Nomenclature
Acknowledgements
We thank the crew of Polarstern for their excellent support, I also thank the ice working group: A. Bartel, A. Darovskih, H. Eicken, J. Freitag, M. Gleitz, S. Grossmann, C. Haas, F. Haubrich, P. Jochmann, J. Kolatscheck, F. Lindemann, J. Lobbes, T. Scherzinger, S. Searson, S. Timofeev, F. Valero Delgado, I. Werner and A. Zatchek for their good team spirit and their helping hands. This research was funded by the "Deutsche Forschungsgemeinschaft" (DFG), Grand SP 377/41.
Appendices
References:
Ackermann, N. L., Shen, H. T. & Sanders, B. E. (1990): Sediment enrichment of coastal ice covers. In: Proceedings of the International Association of Hydraulic Research Ice Symposium, Espoo, Finland, 86-96
Ackley, S. F., Dieckmann, G. & Shen, H. (1987): Algal and foram incorporation into new sea ice. EOS 68, 1736
Alexander, V., Chapman, T. (1981): The role of epontic algal communities in Bering Sea ice. In: Hood DW, Calder JW (eds.) The eastern Bering Sea shelf: Oceanography and resources, NOAA, 2, 773-783Cota, G. F. (1985): Photoadaption of high Arctic ice algae. Nature, London. 315, 219-222
Cota, G. F., Prinsenberg, S. J., Bennett, E. B., Loder, J. W., Lewis, M. R., Anning, J. L., Watson, N. H. F. & Harris, L. R. (1987): Nutrient fluxes during extended blooms of Arctic ice algae. J. Geophys. Res. 92 (C2), 1951-1962
Cota, G. F., & Horne, G. F. (1989): Physical control of Arctic ice algae production. Mar. Ecol. Prog. Ser., 52, 11-21
Eicken, H. (1992): The role of sea ice in structuring Antarctic ecosystems. Polar Biol 12, 3-13
Eide, L. I. & Martin, S. (1975): The formation of brine drainage features in young sea ice. J. Glaciol. 14, 137- 152
Frankenstein, G. & Garner, R. (1967): Equations for determining the brine volume of sea ice from -0.5° to -22, 9° C. J. Glac. 6, 237-239
Gradinger, R., Spindler, M. & Weissenberger, J. (1992): On the structure and development of Arctic pack ice communities in Fram Strait, a multivariate approach. Polar Biol. 12, 727-733Horner, R. (1985): Sea ice biota. CRC press, Boca Raton/Fl. 215p.
Horner, R. (1990): Ice associated ecosystems. In: Polar marine diatoms. Medlin, L, K., Priddle, J. (eds.), British Antarctic Survey, Cambridge, 9-14
Jezek, K. C., Stanton, T. K., Gow, A. J.& Lange, M. A. (1990): Acoustical and morphological properties of undeformed sea ice, laboratory and field results. In: Proceedings of the W. F. Weeks Sea Ice Symposium. U.S. Army Cold Regions Res. and Eng. Lab. Monograph 90-1, Hanover, NH, 67-75.
Maykut, G. A. (1985): The ice environment. In: Sea Ice Biota. Horner, R. A. (ed.), CRC Press, Boca Raton, 21-82
Meguro, H., Ito, K. & Fukushima, H. (1967): Ice flora (bottom type): a mechanism of primary production in polar seas and the growth of diatoms in sea ice. Arctic 20, 114-133
Niedrauer, T. M. & Martin, S. (1979): An experimental study of brine drainage and convection in young sea ice. J. Geophys. Res. 84, 1176-1186
Poulin, M (1990): Sea ice diatoms (Bacillariophyceae) of the Canadian Arctic. I. The genus Stenoneis. J. Phycol. 26, 156-167Reeburgh, W. S. (1984): Fluxes associated with brine motion in growing sea ice. Polar Biol. 3, 29-33.
Shen, H. T. & Ackermann, N. L. (1990): Wave-induced sediment enrichment in coastal ice covers. In: Proceedings of the W. F. Weeks Sea Ice Symposium. U.S. Army Cold Regions Res. and Eng. Lab. Monograph 90-1, Hanover, NH, 100-102
Spindler, M. & Dieckmann, G.S. (1986): Distribution and abundance of the planktonic foraminifer Neogloboquadrina pachyderma in sea ice of the Weddell Sea. Polar Biol. 5, 185-191
Weissenberger, J., Dieckmann, G., Gradinger,
R. & Spindler, M. (1992): Sea ice: a cast technique to examine and analyze
brine pockets and channel structure. Limnol. Oceanogr. 37, 179-183Weissenberger,
J. (1994):Stain dilution experiments: In: Reports on Polar Research, The Expedition
ARCTIC 93 Leg ARK-IX/4 of RV "Polarstern" , 65-68
