<< Please see special notes for V1.0 below>> SHEBA OCEAN TURBULENCE MAST DATA 1. Experiment Description The oceanography mast experiment was designed to measure mean and turbulent quantities at multiple levels in the ocean boundary layer beneath the SHEBA ice floe, using instrument clusters comprising mechanical current meters mounted along 3 mutually orthogonal axes, with nearby SBE temperature and conductivity sensors. The clusters were mounted 4 m apart on a torsionally rigid, self-contained mast that could be lowered to arbitrary depths in the upper ocean. Each cluster feeds signals to a special SBE 9 underwater unit, to be multiplexed to a deck unit and subsequently recorded via computer. The underwater unit includes compass, tiltmeters, and pressure sensor. The system samples 6 times per second, with averaging in the deck unit reducing the data collection rate to 1 or 0.5 samples per second, depending on current velocity. Specification and performance of the system are described in McPhee (1992;1994) and McPhee and Stanton (1996). Measurements were made from 8 Oct 97 to 28 Sep 98, with sizable gaps in February and March do to ice camp breakup and redeployment. Prior to the redeployment in late March, there were 4 clusters (from nominally 4 m to 16 m beneath the ice/ocean interface). After that, two clusters were maintained, nominally at 4 and 8 m below the interface (but raised to 2 & 6 m during the summer). The standard SBE instruments were calibrated prior to the experiment, and those calibrations were used to produce data here. Post-project calibrations were done, and it is possible to intercalibrate in the field, however, any such corrections are not included in the present data set. When 4 clusters were deployed, the uppermost cluster included a SBE7 microstructure conductivity instrument in lieu of the standard SBE4 conductivity cell. The microstructure instrument has much better small scale response, but is subject to significant drift in absolute calibration. Two factors have important impact on the quality of the measurements. First, if the velocity in the boundary layer relative to the drifting ice is less than about 5 cm/s, the geometry of the current meter triad usually precludes all three rotors turning consistently. Second, especially during the summer months, biofouling often affects one or more of the current meters and/or the SBE sensors. 2. Data presentation a. Data are screened objectively for outliers by high-pass filtering and eliminating values lying beyond three standard deviations (3.5 for velocity) from the origin, with gaps filled by linear interpolation. b. Data are then grouped in "15-min realizations" of the turbulent flow. The 15-min sample is considered valid if the flow is strong enough that all current meters turn at least 90% of the time. (For discussion of flux measurements in low energy flows, see McPhee and Stanton, 1996.) c. For valid realizations, the velocity data are rotated into a coordinate system aligned with the mean streamline (assumed horizontal) so the instantaneous flow is (u,v,w) where =U; ==0. By Taylor's hypothesis, an estimate of the ensemble Reynolds averaged momentum and kinematic heat fluxes is then given by the covariances: , and ; where the first is the component of the horizontal kinematic stress traction vector in the direction of the mean streamline, the second is the component 90 deg clockwise from the mean streamline (i.e., u, v, w are components in a right-handed system with w positive upward), and the third is the kinematic vertical heat flux. In principle, the salinity flux is similarly furnished by , but the ducted design of the SBE 4 conductivity meters often precludes measuring into the inertial subrange of the salinity spectrum, and they are not suitable for direct flux measurement without adjustment. When proper calibration is completed the salinity flux from the cluster equipped with the SBE 7 microstructure conductivity will be made available. For now, in the 4-cluster configuration, the salinity of Cluster 1 (topmost) is assigned -99.9 for mean salinity. The variance of salinity IS listed and should be approximately valid, pending any change in SBE 7 calibration from the original. d. Data files are grouped by month and by cluster. For example, mastlistnov_1 contains all valid 15-min realizations during the month of November for Cluster 1. The data for each realization is given on one line with entries separated by spaces. Column 1. Date/time in format yyyymmddhhmmss, sample centered UT time 2. Decimal year day of 1997 (e.g., 1200UT on 1 Jan 98 is 366.5) 3. Water temperature (deg C) <> 4. Water salinity (practical salinity scale)<> 5. Average current speed (m/s) 6. Magnetic heading of the mean horizontal current vector (streamline heading), degrees 7. Depth of the midpoint of the cluster, m. (NB-- NOT distance from ice) 8. Twice the TKE per unit mass: q^2= ++, m^2 s^-2 9. , m^2 s^-2, Reynolds stress in direction of mean streamline 10. , m^2 s^-2, 90 deg clockwise 11. , K m s^-1, turbulent heat flux is rho*cp* 12. , K^2, temperature variance <> 13. , psu^2, salinity variance <> 3. Cautions/Contacts a. Adjustments to the SBE calibrations may follow cross-calibration. b. The 15-min realization averaging time represents an estimate of a "spectral gap" between turbulence and lower frequency variability. Since the time scale of individual turbulent eddies is often a significant fraction of the averaging time, one can expect large variation in flux estimates from one realization to the next, and should average several realizations for stable flux estimates. c. Original time series used in the 15-min realizations can be obtained by special arrangement. d. Contact: Miles G. McPhee McPhee Research Co. 450 Clover Springs Road Naches WA 98937 USA Voice/Fax both 509-658-2575 email: miles@wolfenet.com REFERENCES McPhee, M.G., 1992: Turbulent Heat Flux in the Upper Ocean under Sea Ice, J. Geophys. Res., 97, 5365-5379. McPhee, M.G., 1994: On the Turbulent Mixing Length in the Oceanic Boundary Layer, J. Phys. Oceanogr., 24, 2014-2031. McPhee, M. G., and T. P. Stanton, 1996. Turbulence in the statically unstable oceanic boundary layer under Arctic leads, J. Geophys. Res., 101, 6409-6428. NOTES (1) The following matlab routine will read the ascii files and plot all of the fields: ****************************************************************** ****************************************************************** function [t,mnq,trbq]=check_list(month,cluster) % [t,mnq,trbq]=check_list(month,cluster); % 27 Aug 99 MGM, check the ascii list files % INPUT: 3 letter month string (eg 'apr') % cluster number % OUTPUT: t time vector in days of 1997 % mnq mean quantities: T, S, U, magbrg, depth % trbq turbulent quantities q^2, , , , , s=sprintf('load -ascii mastlist%s_%1d',month,cluster);eval(s); s=sprintf('y=mastlist%s_%1d;',month,cluster);eval(s); t=y(:,2); mnq=y(:,3:7);trbq=y(:,8:13); ans=input('Plot mean quantities? ','s'); if ans=='y' mcap=['Temp (\circC) '; 'Sal (psu) '; 'Speed (m s^{-1})'; 'mag hdg (\circ) '; 'depth (m) ']; hdg=sprintf('Month: %s, Cluster %1d',month,cluster); for j=1:5 plot(t,mnq(:,j),'+');grid title(hdg);xlabel('Day of 1997'); ylabel(deblank(mcap(j,:))); ans=input('Print? ','s'); if ans=='y',print,end end end ans=input('Plot turbulence quantities? ','s'); if ans=='y' mcap=['q^2 (m^2s^{-2}) '; ' (m^2s^{-2})'; ' (m^2s^{-2})'; ' (Kms^{-1}) '; ' (K^2) '; ' (psu^2) ']; hdg=sprintf('Month: %s, Cluster %1d',month,cluster); for j=1:6 plot(t,trbq(:,j),'+');grid title(hdg);xlabel('Day of 1997'); ylabel(deblank(mcap(j,:))); ans=input('Print? ','s'); if ans=='y',print,end end end ****************************************************************** ****************************************************************** Special Notes: Version 1.0 October: Noise on cluster 2 may degrade T/C performance. T and C were set to -99.0 on some 15-min samples but mechanical properties (stress, q^2) ok. November: Noise on cluster 2 may cause degradation early in Nov, some T/S values replaced by -99.9 December: quality checks showed that cluster 3 exhibited inconsistent U/u* ratios, current speed, and theta (angle of rotation about the horizontal ym axis). Spot checks indicated a problem with the wm current component. mastlistdec_3 is not included in V1.0, though some data should be ok. January: Some problems with T, cluster 2, -99.9 substituted for T,S 29 samples were dropped for cluster 3 because of large theta deviation. Problem appears to be corrected after day 23. Note half meter step changes in mast depth. February: Some problems with T1, -99.9 values substituted No signal from cm 1 of cluster 4 after day 37.35 of 1998 March: No signal from cm1 of cluster 4, no mastlistmar_4 file April: Mast deployment reduced to 2 clusters for duration Fouled conductivity on Cl 1 for t<92.1, replaced with -99.9 June: look for depth change late in month July: disappointing lack of good data due to biofouling and low currents until about day 210 (29 Jul). The SBE 9B fish that had been in use since October developed corroded connectors, and was replaced with an older unit (SBE 9A) on 27 Jul. This introduced some problems with the data not seen previously. One was the intermittent introduction of spurious zeros in the magnetic compass reading. In the July files where it occurred (4), it was possible to remove the zeros by linear interpolation, but for some of the files in August and September, the problem was too severe to get decent current direction-- these were dropped from the ascii listing, but will still furnish good vertical velocity and fluxes. The correction for compass cosine response is significantly larger for the SBE 9A instrument, and in general the direction measurements are considered less accurate. August: Biofouling problems persisted in August. A comparatively large number of 15-min realizations otherwise meeting the objective screening criteria were discarded because the angle of rotation needed to bring mean streamline to horizontal exceeded the mean angle by +/- 15 deg. This indicates either horizontal misalignment of entire cluster, or that the um or wm component is slowed by fouling. See also the notes for July regarding current direction. Watch for a short period of deeper deployment around 243.12 (31 Aug). September: See notes for July regarding current direction.