Use of current meters in aquatic research & engineering
Advantages, Sensor Limitations;’Applications; Ocean Circulation; Velocity, Oxygen, Sediment & Transport Factors; Modelling & Monitoring
The increase in exploitation and use of the aquatic environment by humans with increasing population enhances the need for knowledge for future sustainable management of lakes, coastal zones and the open ocean. A frequently used tool for process understanding and management are mathematical models that simulate global, regional and local scenarios. To be reliable models have to be calibrated and validated with high-quality field measurements. The more good data that is used the better the model becomes and forecasting possibilities increase, reducing the risk of taking ill-founded environmental and/or economic-political decisions.
Most of the Earth’s aquatic environments-shallow lakes and the coastal zone to the deep sea (Smith and Druffel, 1998-are in constant change. The time scales for these variations can vary from seconds and minutes to seasonal, annual, over decades and longer. In research and surveying, it is becoming increasingly clear that the most efficient, practical and economical way to gain a better understanding of the ongoing variations is to be constantly present with instruments such as: buoys, drifters, moored instruments, landers and on-line long-term observatories. Even the most ambitious sampling programs cannot sample as frequently as in-situ instruments. Consequently, the risk of overlooking major events is big in “traditional” studies.
For more than 30 years oceanographers have used moored instruments to measure physical parameters-current velocities, temperature and salinity. Recently, a new generation of these instruments has extended the possibilities for investigators by measuring other parameters such as: amount of suspended particles (turbidity), dissolved oxygen, pH, organic material (fluorescence), redox potential, water level and waves (direction and height). Also, the ongoing development in this field will undoubtedly offer users more possibilities in the future.
This article presents investigations and results obtained with Aanderaa Instruments AS (Bergen, Norway) RCM9 current meters; it indicates the performance, weaknesses and the vast possibilities offered by such instruments.
Sensors Advantages, Limitations
Since the RCM9 entered the market in the spring of 1996, a total of about 600 instruments have been delivered.
The instruments can autonomously measure and record current speed and direction (velocity), conductivity, temperature, depth, dissolved oxygen and turbidity for more than three years (depending on the measuring interval). More technical information about the RCM9 is available at http://www.aanderaa.no/.
Current velocity. The “classical” way to measure current speed is to use rotor/propeller equipped instruments (e.g., Emerey and Thomson, 1997; Mitsuzawa and Holloway, 1998; Woodgate et al., 1999). A mounted vane assembly that directs the instrument into the current determines its direction. Instruments with moving parts are limited in measuring low current speeds-a minimal current, normally around 2 centimeters/second, is needed for the rotor to start. Also in areas with high loads of particles or high growth rates, the rotor might be affected.
The new generation of current meters measure velocity without moving parts by using acoustic frequency– shift or electromagnetic techniques (for further electromagnetic sensor information, see www.interoceansystems.com/; http://www.marshmcbirney.com/ and http://www.valeport.co.uk/).
Acoustic current meters transmit hydroacoustic signals (normally at frequencies of 38 to 10 MHz) and use the acoustic frequency shift to measure the current. In principle, two different techniques are possible: the “sing around technique” and the “backscatter technique.” The former uses the phase-shift of “acoustic paths” to measure the current. Sending a signal between narrowly placed transmitter and receiver pairs creates an acoustic path (the latest commercial information on this technology is available at http://www.falmouth.com/, http://www.nobska.net/mavs.htm and http://www. sensortec.com/).
In the acoustic backscatter technique, the instrument transmits an acoustic signal (from at least 0.4 meters away), which is reflected by particles and/or gas bubbles moving with the water. Depending on the velocity of the suspended particles, a frequency shift (Doppler) in the acoustic signal will occur which is used to calculate the current velocity.
This technique is used on the Aanderaa RCM9. Other single-point current meters employing the same technique are presented on http:// www.nortek-as.com/ and http:H www.sontek.com/index.html.
All of these techniques, including that of the rotor, have limitations. The electromagnetic and sing-around techniques increase the risks of “hydrodynamic noise” (disturbances of the water current at the point of measurement), especially if fouling occurs or if any objects (e.g. algae and/or seaweed) get trapped on the instrument or in its frame. The backscatter technique normally uses the upstream components to calculate the current, which eliminates the risk of hydrodynamic disturbances from the instrument itself. These instruments are also almost insensitive to fouling (see Figure 1). An inconvenience with the technique is that the signal-to-noise ratio increases in clear waters without particles. However, good signal-processing hardware and software can compensate for it.
Many investigators have compared and intercalibrated current meters of different makes and measuring principles. Baring hardware or software failures, in most cases, the various instruments and measuring principles compare well. One ongoing current-meter test is presented at the home pages of Woods Hole Oceanographic Institute (http://www.whoi.edu/science/ultra– moor/).
Other such instrument inter-comparative tests have recently been done by research institutes such as: IFREMER (France), JAMSTEC (Japan), Alfred Wegener Institute (Germany), NOAA-PMEL
(Seattle, USA) and the Institute of Marine Research (Norway).
Conductivity and temperature. Water density differences, together with winds, are the main driving forces of circulation in oceans and lakes. Temperature and salinity mainly determine the water density. Temperature is a straightforward parameter to measure, and most manufacturers equip their instruments with reliable platinum-wire sensors (such as Pt 100 or Pt 2000). If a reasonably fast response to temperature changes is needed, necessary in for example in profiling applications (see below), care should be taken in placing the temperature sensor in an external thin-walled metal housing. The temperature response time of some current meters was discussed in the noted Woods Hole Ultramoor tests. In general selfrecording current meters do not have the same rapid response time and high-sampling rates as profiling CTD instruments (e.g. http://www.ocean sensors.com, www.seabird.com/, http://www.idronaut.it/index.html). In Figure 2 comparative profiles of temperature, salinity and turbidity/light transmission between a Sea Bird CTD instrument (for more information see www.seabird.com/) and an RCM9 are presented.
The difficulties involved in estimating salinity/density of natural waters have been subject to numerous discussions and methodological revisions during the past 150 years (for a review, see Wilson, 1981). For in-situ measurements today, conductivity is by far the most dominant method in use. The conductivity indicates the complex mixture of dissolved salts, but it is also affected by biological and chemical parameters (e.g. calcium-carbonate dissolution and organic-carbon oxidizing to C02 contribute to the conductivity). Consequently, especially in coastal areas, investigators should be careful in using conductivity as an absolute measurement of salinity/density, if not accounting for the exact composition of the seawater (Wilson, 1981).
Two methods have been used to measure conductivity in-situ: (1) electrode technique: a water sample is “trapped” into a closed, geometrically well defined conductivity cell. The forcing of a current between electrodes in the cell generates an electrical field. The voltage drop in this field is dependent on the conductivity of the water; (2) inductive method: an open conductivity cell consisting of two coils is placed in the water. The conductivity of the water within the cell is measured.
A weakness of both methods is that a slightly different conductivity will be measured for water circulating through the cell compared with immobile water. For higher accuracy and precision some manufacturers pump water through their cells when measurfing (see e.g. http://www.oceansensors.com and www.seabird.com/). When using moored instruments in environments with high biological growth, care must be taken since conductivity cells are sensitive to fouling (e.g. to changes in the geometry of the measurement cell).
Water conductivity is complexly related to pressure and temperature, which consequently have to be measured in parallel and to be accounted for (by using empirical formulas; UNESCO, 1980) when converting to salinity. Both temperature sensors and conductivity cells may drift with time and regular recalibrations are recommended.
Pressure. It is necessary to measure pressure when using the instrument for profiling or investigations that involve variations in the water level. As indicated, pressure is needed as an input parameter for salinity calculations and is a useful parameter when controlling instruments moored at a specific depth.
For profiling in shallow water or monitoring of small water-level variations, high precision sensors are needed while for instrument positioning, in e.g. a mooring, wide-range sensors with lower precision is normally sufficient. In most cases, pressure is measured by means of a silicon piezoresistive bridge.
Turbidity. Measurements of suspended sediment concentration started more than 30 years ago. Usual methods are: – gravimetric (water is vacuum-filtered, in the laboratory, into a pre-weighed 0.4- to 0.8-micrometer mesh filter, which is dried and weighed again), * optical and more recently * acoustic.
The signal strength of the acoustic backscatter from current sensors has been used to monitor suspended sediment concentrations. However, since the signal strength from backscattering particles depends on complexly interrelated parameters (e.g. sound absorption, signal strength, acoustic pulse length, range between the transducer and the particle size, shape and composition etc.), this technique has frequently proven difficult to use.
Optical instruments offer a more straightforward method to obtain highresolution time series of suspended sediment concentration, although also sensitive to particle size, shape and particles in suspension are: * by measuring the light dampening caused by suspended particles in a light path (transmissometer) or * by optical backscatter. To be sensitive, the light transmitter and receiver (the light path) of transmissiometers have to be sufficient, which leads to rather big instruments inappropriate for mounting on current meters. The Instituto de Ciencias del Mar de Barcelona (CSIC) participated in several projects (www.icm.csic.es/inve/sinvedepgof.ht ml.se) where extensive comparisons between three different types of optical instruments were made: – Sea Tech transmissometer (currently Wet Labs Inc. (Philomath, Oregon), www.wetlabs.com) * the optical backscatter sensor (by D&A Instrument Co. (Townsend, Washington); www.d-ainstruments.com) and * RCM9 turbidity sensor (also an optical backscatter sensor). Calibrations between turbidity measurements expressed as “beam attenuation coefficient” (BAC) or Formazin Turbidity Units (FTU) and suspended sediment concentration by gravimetric methods were made in estuary, shelf and slope environments (Guillen et. al., in press). A general conversion relation between turbidity units (BAC and FTU) and suspended sediment concentration does not exist. Specific calibration curves for each instrument and for each environmental condition are required in order to obtain accurate transformations. General calibrations can only be used for qualitative or semiquantitative purposes when they are used for suspended sediment with similar characteristics. The suspended sediment concentration in the marine environment can range from less than 0.1 milligram/liter to several thousands of milligrams/liter. Thus, the selection of the turbidity range appropriate to each condition is a relevant factor for successfully quantifying suspended sediment concentration and fluxes. It is recommended to use two or more equivalent sensors simultaneously with different calibration ranges.
Oxygen. In most biological and chemical processes, oxygen is the single most important parameter to measure. Furthermore, for environmental reasons it is critical to monitor oxygen in areas where the supply of oxygen is limited compared to the need (e.g. Fjords, the Baltic Sea, the Black Sea, around fish farms, in areas prone to the dumping of mine waste, in shallow coastal areas with frequent algae blooms etc.).
Until today almost only electrochemical oxygen sensors (of Clark type, Clark, 1959) have been used.
Regardless of the make, these sensors are affected by several complexly interrelated factors (Berntsson et al., 1997) such as temperature, pressure, water circulation in front of the sensor head and aging of membranes and the electrolyte. Therefore, care must be taken when using them for absolute long-term measurements. However, if the sensors are calibrated against Winkler titration, or if relative changes in oxygen (instead of absolute values) are of interest, useful information can be obtained with them (examples below).
In the near future, commercially available optical oxygen sensors (soon available from Aanderaa and PreSens http://www.presens.de) might offer users new possibilities for accurate long-term measurements (Klimant et al., 1995; Holst et al., 1995; Glud et al., 1999).
Ocean circulation. Freezing and melting of sea ice in the Arctic Ocean and around Antarctica are the main driving forces of ocean circulation on Earth. A good knowledge of the present circulation in these areas is essential to be able to understand how climate changes are and will affect this circulation.
As part of the European Union MAST III project VEINS (Variability of Exchanges in the Northern Seas, http://www. ifm. uni-hamburg.de/-veins/index.html), five moorings with a total of 20 current meters (RCM7 and some RCM9) were deployed in the area between Norway and Bear Island in the Barents Sea to measure the variability in the volume flux in and out of the Barents Sea (Ingvaldsen et al., 1999, Ingvaldsen et al., 2000). The amount of Atlantic water flowing via the Barents Sea to the Arctic Ocean is approximately the same amount as the one flowing west of Spitsbergen. It is therefore important to know the variability in the volume flux in order to understand the temperature changes that now take place in the Arctic Ocean. In addition, the amount of Atlantic water flowing into the Barents Sea is of great importance for the fish stocks living there. Therefore, current meters are a very important tool in the modeling and monitoring process of Atlantic inflow.
Variations in oxygen concentrations and their implications. Oxygen uptake by sediments. One of the goals in the ongoing European Union MAST III project KEYCOP (Key Coastal Processes, http://biologi.uio.no/keycop/) is to understand and compare biogeochemical processes at the sediment-water interface in the nutrient-rich (mesotrophic) Skagerrak (North Sea) and the nutrient-poor (oligotrophic) Northern Aegean Sea.
In-situ studies of biological and chemical activities are done with bottom landers (Tengberg et al., 1995). Descending to the seafloor, landers measure chemical exchanges between the sediment and the overlying water by “benthic chambers” gently closing off 4 x 400 square centimeters of sediment and the overlying water. By measuring with electrodes or by taking water samples in the entrapped water, solute exchanges between the sediment and the water can be estimated. One example is the sediment oxygen uptake (mainly done by bacteria) that can and has been measured directly with RCM9-type electrodes.
Oxygen variations in shallow, partly reed-covered lakes. An extreme example is Lake Ensiled at the AustroHungarian boarder that is covered by reed over more than 100 square kilometers. The mean depth of the lake is around 1.5 meters, but in most of the reed-covered zones it decreases to 0.30.5 meters. Prevailing windstorms can induce strong seiche and circulatory motion with temporary water-level changes of up to 30 to 40 centimeters. This water mass back-and-forth motion results in water-mass exchange between the pelagic and littoral areas being the main source of oxygenated water to the reed. The seiche motion is considerably damped by the dense reed; therefore, the flushing of the inner part of reed-covered zones is poor. In long, calm, late summer periods total depletion of dissolved oxygen can be observed. As a result, phosphorus is released, which enhances the eutrophication processes.
RCM9 current meters equipped with turbidity and dissolved oxygen sensors have recently been used (in a Finnish-Hungarian pilot study) in a 10-meter wide and 2-meter deep section of a main canal some 300 meters away from its entrance to the pelagic area. The bottom-mounted instrument captured several prevailing wind– induced events, showing clear features of the hydrodynamic-driven dissolved-oxygen variations (see Figure 3).
Measuring and modeling of sediment transport processes. Lake Balaton (Hungary), the largest shallow lake in Central Europe, from time to time, is high in nutrients from resuspended sediment which results in poor water quality. One way to improve the conditions is to remove the uppermost-polluted bed layer by dredging. To plan interventions, detailed information on wind-induced hydro- and sediment dynamics in prevailing winds is essential.
Wind, current and sediment transport measurements, laboratory analysis and related numerical modelling (Jozsa et al., 1998) have been combined to plan dredging activity in the west-most 40-square-kilometer large, 2- to 3-meter deep bay of the lake, characterised by spatially irregular, high internal nutrient loads in the cohesive bed. Long-term simultaneous measurements of wind, current and turbidity at multiple depths and sites revealed the main features of prevailing water and sediment exchange processes. Wind-induced flow data showed the strong presence of largescale, primarily horizontal circulation patterns, which could be reproduced with a depth-integrated finite-difference numerical flow model by proper scaling of the surface wind shear stress field (J6zsa et al., 1990; Sarkkula et al., 1991). Turbidity data converted to suspended solids concentration by laboratory calibration showed the overall dominance of wind-wave induced resuspension, enhanced by current– induced bottom shear stress at some places.
The collected data have been used then to tune a depth-integrated finite difference sediment transport model. In the model, spatially different threshold bed erosion stresses were estimated based on bottom mapping by a penetrometer. A tri-fraction sediment transport model provided acceptable prediction of the fine-grained mud accumulation zones. More information is available at http://www.aanderaa.no.
Sediment transport and erosion at an intertidal mudflat. In North Sea tidal areas, the transported particles are generally fine-grained and reactive and can serve as important sinks of organic compounds and metals. The main objective of the European MAST III project INTRMUD (http://www.science.plym.ac.uk/KDYER/) was to establish a detailed classification scheme for intertidal mudflats based on physical, chemical and biological parameters. Mudflat investigations in the Danish Wadden Sea included determinations of: sediment height variations; grain size; organic content and fecal pellets; suspended particles; equivalent settling diameters of suspended material; sedimentation rates; erosion thresholds; chlorophyll content and dominant biological species.
To more fully understand how suspended sediment is brought to the tidal flats, current velocity, water depth and suspended material (see Figure 4) were examined, by burying an RCM9 in the mud so that only the upper part of the instrument was sticking up, under different wind conditions. The example shows that current velocities were almost symmetric around high water slack but the suspended material generally had higher concentrations in low tide than in high tide due to local onshore winds (Andersen, 1999).
Sediment resuspension and release of metals from mining waste. In 1998, five million tons of toxic mud from a pond of pyrite mining waste accidentally spilled into the Guadiamar River (South Spain), which conjuncts with the bigger Guadalquivir River before entering the Atlantic. The estuary did not receive any particulate pyrite but it was affected by spill wastewaters with high concentrations of dissolved zinc. Due to the drastic pH increase in the estuary (where the river meets the sea, the pH increases from 4.5 to almost 8) most of the dissolved zinc precipitated out into the sediments and/or onto suspended particles. This resulted in a 20fold increase in the zinc content of the estuary surface sediments compared to background values in unpolluted areas. Tides cause constant periodic resuspension of contaminated sediment (Fgure 5), potentially harmful to benthonic and planctonic fauna, fish, shellfish and other estuary organisms (Palanques et al., 1999).
Trawling and sediment resuspension. The direct effect of bottom trawling (ploughing and scraping of the seabed) is sediment resuspension. In an experiment to create a disturbance by means of experimental fishing in an unfished sheltered area in the northwestern Mediterranean Sea (Palanques et al., in press), water turbidity was recorded before and during trawling. Trawling produced a slow and gradual turbidity increase in the water column in the first hours after trawling began, which generally lingered less than 2 meters above the bottom, then progressively resuspended particles were dispersed several meters above the bottom, and water turbidity 2 meters above the bottom increased significantly between one and four days after. Trawling on continental shelves has a significant effect on water turbidity, which must be superimposed on natural processes.
Spill monitoring from dredging activities. Water-suspended particles, which have a direct impact on the ecological system, may enhance the releases of toxic substances and/or nutrients and/or increase the consumption of oxygen (Wainright, 1987; Wainright and Hopkinson, 1997). In a two-day spill-monitoring project in Trelleborg Harbor (South Sweden) during dredging, profiles at different distances from the dredging platform were obtained every second meter of water depth by lowering an RCM9. To calibrate the turbidity sensors, a Niskin bottle was mounted just above the RCM9 for water samples. The circulation in the harbor is mostly driven by ferries except for occasional strong winds. As expected, water suspended material decreases with distance from the dredging platform. It was noticed that a departing ferry created high current speeds and particle resuspension, similar to the highest values measured around the dredging platform.
Navigational safety around offshore platforms. Recording current meters can also be used to improve navigational safety and ROV operations around offshore platforms. On the Norwegian platform Polar Pioneer, an RCM9 is used for continuous online profiling of salinity and temperature as well as current velocity.
Since RCM9 current meters are good for long-term deployments (up to three years, depending on intervals), reliable, robust, easy to use and since they combine several different sensors on the same instrument, they have proven suitable for a large variety of investigations, some of which are presented here.
Since it is insensitive to fouling, backscatter technology is the most suitable technique for any long-term current velocity measurements. The “traditional” weakness of high signalto-noise ratios in clear water of this technology have been overcome with new hardware and software on the Aanderaa current meters. The RCM9s and RCMlis (deep-sea version of RCM9) are successfully used in clear waters of the arctic winter lakes and the deep sea.-Dr. Anders Tengberg, Department of Analytical & Marine Chemistry, Goteborg University, Goteborg, Sweden; Dr. Morten Pejrup, associate professor, and Thorbjrn J. Andersen, oceanographer, Institute of Geography, University of Copenhagen (Copenhagen, Denmark); Dr. Jorge Guillen, senior scientist, and Dr. Albert Palanques, researcher, Institute of Marine Sciences of Barcelona (Spanish Research Council); Jostein Hovdenes, senior engineer, and Helge Minken, vice president and head of the technology department, Aanderaa Instruments A’S (Nesttun, Norway); Randi Ingvaldsen, physical oceanographer and Harald Loeng, senior scientist, Institute of Marine Research (Bergen, Norway); Dr. Janos Jozsa, associate professor Department of Hydraulic & Water Resources Engineering, Budapest University of Technology & Economics (Budapest, Hungary); and Dr. Juha Sarkkula, coordinator and manager of national and international projects, Impacts Research Division, Finnish Environment Institute (Helsinki, Finland).
Andersen T.J., Meddelelser fra Skalling-Laboratoriet, XXXVII, ISBN: 87-87945-39-8, Copenhagen, 1999.
Berntsson M. et al., Analytica and Chimica Acta, 355: 43-53, 1997.
Butcher S.S. et al., Wolfe editors, Global biogeochemical cycles. Academic Press Limited, London, UK., pp. 367, 1992.
Clark L.C. Jr., US. Patent Number 2,913,386,1959.
Emery W.J. and R.E. Thomson, Data analysis methods in Physical Oceanography, Elsevier Science Ltd., Oxford, UK, 1997.
Glud R.N. et al., Deep-Sea Research, 46: pp. 171-183, 1999.
Guillen J. et al. Scientia Marina (in press), 2000.
Holst G. et al., Proceedings SPIE 2508, 45, pp. 387-398, 1995.
Ingvaldsen R. et al. (Submitted to Continental Shelf Research), 2000. Ingvaldsen R et al., ICES C.M 1999
(L:05): 12 pp (mimeo), 1999. Jahnke R.A. Reviews of Geophysics, 28, pp. 381-398, 1990.
J6zsa J. et al., Proceedings VIII. International Conference on Computational Methods in Water Resources, Venice, Italy, CMP/ Springer, 1990.
Jozsa J. et al., Proceedings 3rd International Conference on Hydro-Science and Engineering, Cottbus, Germany, The University of Mississippi, Center for Comput. Hydro-Science and Engineering, 1 GOR
Jozsa J. et al., Proceedings XXVIII. IAHR Congress, Graz, Austria, 1999.
Klimant I. et al., Limnology and Oceanography, 40, pp. 1159-1165, 1995.
Loffler H., Neusiedlersee: The limnology of a shallow lake in Central Europe. Dr. W. Junk bv Publishers, The Hague, The Netherlands, 1979.
Mitsuzawa K. and G. Holloway, Journal of Geophysical Research, 103, pp. 13085-13092, 1998.
Palanques A. et al., Limnology and Oceanography, In press, 2000. Palanques A. et al., The Science of
Total Environment, 242, pp. 211220, 1999.
Rakoczi L. and J. Jozsa. Proceedings XXVIII. IAHR Congress, Graz, Austria, 1999.
Sarkkula J. et al., Proceedings Vienna Symposium, IAHS Publication No. 206, 1991.
Smith K.L. Jr. and E.R.M. Druffel. Deep-Sea Research II, 45, pp. 573586, 1998.
Tengberg A. et al., Progress in Oceanography, 35, 253-292, 1995. UNESCO Report #37 on practical
salinity scale 1978: E. L. Lewis, Ocean Engineering, 1980. Wainright S.C. Science, 238, pp. 17 101712, 1987.
Wainright S.C. and C.S. Jr. Hopkinson, Journal of Marine Systems, 11, pp. 353-368, 1997.
Wilson T.R.S., Conductometry. In M. Whitfield and D. Jagner editors, Marine Electrochemistry, John Wiley and Sons Ltd., UK., pp. 145185, 1981.
Woodgate R.A. et al., Journal of Geophysical Research, 104, 1805918072, 1999.
Copyright Compass Publications, Inc. Feb 2001
Provided by ProQuest Information and Learning Company. All rights Reserved