A systematic examination of measurement error in acoustic Doppler current profiler (ADCP) velocity estimates, in the limit of large signal-to-noise ratio, is made using a system model and sonar signal simulations coupled into an ADCP. The model is extremely successful in predicting ADCP performance. The signal simulations provide model validation. Three main sources of error are examined: frequency tracking, measurement variance (inherent variance of pulse-to-pulse incoherent volume reverberation), and measurement bias. A theoretical lower bound on measurement variance is directly tested by coupling simulated signals into an ADCP. Trade-offs between velocity error, vertical and temporal resolution, and maximum range are discussed, with specific focus on optimizing parameters available to users, of commercial instruments.
A pilot intercalibration study was performed to evaluate the ability of RD Instruments' 307 kHz acoustic Doppler current profiler to measure zooplankton abundance. The intercalibration was between a bottom-mounted acoustic profiler deployed in 150 m of water at the edge of the New England [USA] shelf and zooplankton samples collected with a MOCNESS net system which was towed near the profiler. The results of the study showed significant correlations between backscattered signal intensity and total zooplankton volume, cross-sectional area, and dry weight. From the correlations it appears to be possible to predict zooplankton biomass to approximately .plus-minus. 15 mg m-3 over the range of abundance spanned by our data with enormous resolution in time and space. However, experience with processing the acoustic data clearly indicates that significant results are only possible after careful attention to the calibration of the transducers/electronics and to the quality of the data produced by the individual profiler beams.
Acoustic Doppler current profiler (ADCP) velocity measurements are subject to bias due to the effect of the signal processing filters on the spectrum of the Doppler-shifted signal and on the noise. Bias will occur when the filter is not centered on the signal. Numerical models of the received signal and the processing filter used in RD Instruments profilers show that biases on the order of 10 cm s/sup -1/ can occur in the lower half of the current profile in regions of high current shear. Errors tend to increase with the width of the acoustic beam and with the speed of the ship, and decrease with the pulse length. These biases are identified in ADCP velocity measurements made in the high shear of the equatorial undercurrent. The authors suggest criteria for editing existing ADCP data to remove excessive bias, and recommend changes in profiler parameters which should greatly reduce the bias in future datasets.
A systematic examination of measurement error in acoustic Doppler current profiler (ADCP) velocity estimates, in the limit of large signal-to-noise ratio, is made using a system model and sonar signal simulations coupled into an ADCP. The model is extremely successful in predicting ADCP performance. The signal simulations provide model validation. Three main sources of error are examined: frequency tracking, measurement variance (inherent variance of pulse-to-pulse incoherent volume reverberation), and measurement bias. A theoretical lower bound on measurement variance is directly tested by coupling simulated signals into an ADCP. Trade-offs between velocity error, vertical and temporal resolution, and maximum range are discussed, with specific focus on optimizing parameters available to users, of commercial instruments.
The Ekman transport is estimated from two sets of hydrographic and shipboard acoustic Doppler current profiler (ADCP) velocity observations made during June and September 1995, during the southwest monsoon in the Arabian Sea. Both sets of measurements were made along latitude 8 degrees 30' N, designated as World Ocean Circulation Experiment (WOCE) line I1W, from Somalia to Sri Lanka. The Ekman transport estimates calculated from ageostrophic velocity were southward: 17.6+or-2.4 10/sup 6/ m/sup 3/ s/sup -1/ in June and 7.9+or-2.7 10/sup 6/ m/sup 3/ s/sup -1/ in September. These direct estimates were in good agreement with those predicted by the Ekman balance using both shipboard and climatological winds. The vertical structure of the ageostrophic velocity and the stratification were quite different between the two occupations of the transect. The wind-driven momentum was confined to a very shallow layer in June (about 50 m) and the surface layer was strongly stratified, with a maximum salinity layer at depths between 50 and 70 m. The ageostrophic velocity penetrated much deeper in September (to about 160 m) and the pycnocline was correspondingly deeper. In both cases, the Ekman transport penetrated beneath the mixed layer, to the top of the pycnocline.
Moored acoustic Doppler current profiler velocity estimates and buoy wind observations made during a period of moderate southward winds were used to test the Ekman balance at a site in the California Current. As in prior studies, the wind-driven flow was separated from the total flow by subtraction of a deep reference current. The wind-driven flow was shown to be in an Ekman balance on daily timescales over a period of several months. The mean observed transport was to the right of the wind and agreed to within 3% in magnitude and 4 degrees in phase with the predicted Ekman transport, although the error bar was about 20%. The mean velocity profile was a smooth spiral, qualitatively similar (although flatter) than the theoretical Ekman spiral. From the observed mean momentum balance, profiles of the turbulent stress and eddy viscosity were inferred. Eddy viscosity estimates within the wind mixed layer were O(100 cm/sup 2/ s/sup -1/).
A comparison of measured and wind-derived ageostrophic transport is presented from a zonal transect spanning the Atlantic Ocean along 11 degrees N. The transport per unit depth shows a striking surface maximum that decays to nearly zero at a depth of approximately 100 m. The authors identify this flow in the upper 100 m as the Ekman transport. The sustained values of wind stress and the penetration depth of the Ekman transport are discussed. The transport, calculated from the difference of geostrophic shear and shear measured by an acoustic Doppler current profiler, is compared with that estimated from shipboard winds, and from climatology. The mean depth of the Ekman transport extended below the mixed layer depth, which varied from 25 to 90 m. The profile of ageostrophic transport does not appear consonant with slablike behavior in the mixed layer, even when spatial variations in mixed layer depth are taken into account.
Mixed-layer drifter measurements and hydrographic examinations of the 3-D circulation of the Ensenada Front near northern Baja California, Mexico, reveal the presence of a narrow, surface-intensified, high speed filament and help trace the flow and sub-mesoscale movement. Low-salinity and cool water of subarctic origin constitute the filament. The filament turns toward the southeast after flowing for 200 kilometers. The mesoscale eddy field influences the course of the filament.
The spatial covariances of the time-dependent density and geostrophic velocity fields off southern California are determined from a unique set of repeated hydrographic observations (44 cruises) made by the California Cooperative Oceanic Fisheries Investigations from 1984 to 1994. The covariances and objective analysis are used to combine direct velocity observations, from shipboard acoustic Doppler current profiler (ADCP) measurements made on a recent survey (October 1993), with hydrographic observations. The analysis reduces ageostrophic noise in the ADCP velocities by smoothing and enforcing horizontal nondivergence; additionally, the velocities are mapped over scales that are dynamically consistent with the hydrography. Maximum surface geostrophic flow in the California Current in October 1993 is about 35 cm s/sup -1/, 50% larger than that estimated assuming a 500-m level of no motion. Absolute flow at 500 m is O(10 cm s/sup -1/) and indicates that the surface eddy field penetrates through the thermocline. Uncertainty in the geostrophic reference calculated from the ADCP measurements is of O(4 cm s/sup -1/). The velocity residual (objectively analyzed minus raw ADCP estimates) exhibits smaller correlation scales than the geostrophic flow.
The authors evaluate the temporal and horizontal resolution of geostrophic surface velocities calculated from TOPEX satellite altimeter heights. Moored velocities (from vector-averaging current meters and an acoustic Doppler current profiler) at depths below the Ekman layer are used to estimate the temporal evolution and accuracy of altimeter geostrophic surface velocities at a point. Surface temperature gradients from satellite fields are used to determine the altimeter's horizontal resolution of features in the velocity field. The results indicate that the altimeter resolves horizontal scales of 50-80 km in the along-track direction. The rms differences between the altimeter and current meters are 7-8 cm s/sup -1/ much of which comes from small-scale variability in the oceanic currents. The authors estimate the error in the altimeter velocities to have an rms magnitude of 3-5 cm s/sup -1/ or less. Uncertainties in the eddy momentum fluxes at crossovers are more difficult to evaluate and may be affected by aliasing of fluctuations with frequencies higher than the altimeter's Nyquist frequency of 0.05 cycles d/sup -1/, as indicated by spectra from subsampled current meter data. The eddy statistics that are in best agreement are the velocity variances, eddy kinetic energy and the major axis of the variance ellipses. Spatial averaging of the current meter velocities produces greater agreement with all altimeter statistics and increases the authors' confidence that the altimeter's momentum fluxes and the orientation of its variance ellipses (the statistics differing the most with single moorings) represent well the statistics of spatially averaged currents (scales of 50-100 km) in the ocean. Besides evaluating altimeter performance, the study reveals several properties of the circulation in the California Current System: (1) velocity components are not isotropic but are polarized, strongly so at some locations, (2) there are instances of strong and persistent small-scale variability in the velocity, and (3) the energetic region of the California Current is isolated and surrounded by a region of lower energy starting 500-700 km offshore. This suggests that the source of the high eddy energy within 500 km of the coast is the seasonal jet that develops each spring and moves offshore to the central region of the California Current, rather than a deep-ocean eddy field approaching the coast from farther offshore.
Low-pass-filtered velocities obtained from World Ocean Circulation Experiment (WOCE) surface drifters deployed in the California Current off northern California during 1993-1995 have been compared with surface geostrophic velocity estimates made along subtracks of the TOPEX/POSEIDON altimeter and with moored acoustic Doppler current profiler (ADCP) data. To obtain absolute geostrophic velocities, a mean sea surface height (SSH) field was estimated using the mean drifter velocities and historical hydrographic data and was added to the altimetric SSH anomalies. The correlation between collocated drifter and altimetric velocities is 0.73, significant at the 95% level. The component of the drifter velocity which was uncorrelated with the altimetric velocity was correlated with the wind in the Ekman transport sense. Monthly averages of eddy kinetic energy (EKE), estimated using all drifter and altimeter data within the domain (124 degrees -132 degrees W, 33 degrees -40.5 degrees N), show energy levels for the drifters that are about 13% greater than those for the altimeter. Drifter, altimeter, and ADCP measurements all exhibit similar seasonal cycles in EKE, with the altimeter data reaching maximum values of about 0.03 m super(2) s super(-) super(2) in late summer/fall. Wavenumber spectra of the altimeter velocity indicate that the velocity fluctuations were dominated by features with wavelengths of 240-370 km, while the ADCP data suggest that the temporal scales of these fluctuations are at least several months. Between 36 degrees and 40.5 degrees N, the region of monthly maximum EKE migrates westward to about 128 degrees W on a seasonal timescale. This region of maximum EKE coincides with the maximum zonal SSH gradient, with increased EKE associated with increased southward flow. A simple model shows that much of the seasonal cycle of the SSH anomalies can be produced by linear processes forced by the curl of the wind stress, although the model cannot explain the offshore movement of the front.
Moored observations of currents and temperatures made in the upper 600 m on eddy-resolving scales over a 2-year period are used to examine the spatial and temporal characteristics of the California Current mesoscale circulation. The observations were made at three principal longitudes: 124 degrees W, 126 degrees W, and 128 degrees W in the vicinity of Point Arena. They bracket the 600-km-wide band of high mesoscale variability found along the eastern boundary of the North Pacific. At all locations, the mesoscale variability was larger than the mean flow, and the spatial modes of variability as determined from empirical orthogonal function analysis consisted of an alongshore mode, a cross-shore mode, and a rotational mode. Observations made near the continental slope (124 degrees W) were dominated by the poleward flowing California Undercurrent, with mesoscale eddies and meanders superposed. The nearshore eddy kinetic energy peaked in a band centered around 60 days. Observations made at 128 degrees W, near the offshore boundary between the energetic mesoscale band and the ``eddy desert'' of the northeast Pacific, were characterized by small means, fewer eddy events, and a peak in eddy kinetic energy at 120-180 days. The good horizontal resolution of the current meter arrays allowed us to estimate the relative vorticity, horizontal divergence, and Rossby number and therefore to evaluate the relative strength and occurrence of anticyclones and cyclones. We found the mesoscale eddy field to be strongly nonlinear, with Rossby numbers ranging from 0.1 to 0.5. All of the eddies observed at the offshore site were nonlinear, deep, warm anticyclones. Shipboard hydrography revealed the origin of one of these anticyclones to be the California Undercurrent, and this eddy retained its strong anomalies after several months and several hundred kilometers of propagation. Despite the lower incidence of eddies as one moves west from the coast, the eddies that we observed offshore provide evidence for propagation and transport of properties from the coast to the central North Pacific across the California Current System.
A deep, nonlinear warm eddy advecting water that was also anomalously saltier, lower in oxygen, and higher in nutrients relative to surrounding waters was observed in moored current and temperature measurements and in hydrographic data obtained at a site similar to 400 km off the coast of northern California. The eddy was reproduced using a nonlinear quasi-geostrophic model, initialized by an iterative procedure using time series of 2-day averaged moored current measurements. The procedure demonstrates how a data assimilative technique synthesizes and enhances the resolution of a relatively sparse data set by incorporating time-dependence and model physics. The model forecast showed significant skill above persistence or climatology for 40 days. Our hypothesis, that the eddy was generated at the coast in winter and subsequently moved 400 km offshore by May, is consistent with the eddy movement diagnosed by the model and with the observations and coastal climatology. The model evolution significantly underpredicted the temperature anomaly in the eddy owing in part to unmodeled salinity compensation in trapped California Undercurrent water. Together, observations and model results show a stable nonlinear eddy in the California Current System that transported water and properties southwestward through the energetic eastern boundary region. Coherent features such as this one may be a mechanism for property transfer between the eddy-rich coastal zone and the eddy desert of the eastern North Pacific Ocean.
A pair of high-resolution oceanographic surveys in March and April 1995 revealed a large and rapid transition from late winter to spring conditions in the coastal zone off central and southern California. These data are unique in capturing the detailed three-dimensional physical structure of and biological response to the spring transition in the southern California Current System (CCS). Changes associated with the transition included a strong tilting of isopycnals, which lifted by up to 60 m near the coast and dropped 20-40 m offshore, a subsequent increase in cross-shore density gradients, the development of a strong nearshore equatorward jet, and an increase in net equatorward transport from the shelf break out to 300 km offshore. The most dramatic physical changes were confined to the shoreward 150 km and extended at least to the depth of the core of the California Undercurrent ( similar to 300 m). In response to these physical changes, there was an apparent strong increase in primary productivity, as indicated by changes in nearshore vertically integrated fluorescence and beam attenuation coefficient. Atmospheric and oceanic conditions in the CCS were near seasonal norms in the winter and spring of 1995, implying that a transition of the magnitude and rapidity observed here may be an annual event. Furthermore, the development of the coastal upwelling jet was independent of the winter manifestation of the main core of the California Current, which was maintained well off shore. This suggests that the California Current is regenerated seasonally through the development and offshore evolution of the coastal upwelling jet. It is not known whether the new jet joins and strengthens or replaces the offshore core of the previous winter.