R/V MOANA WAVE - FLUX DATA (Release 2.5)

C. Fairall (NOAA/ERL, USA)

Data Access

After June 1996, data will also be archived at NOAA's National Climatic Data Center (NCDC) in Asheville, NC (http://www.ncdc.noaa.gov) and at the National Center for Atmospheric Research (NCAR) in Boulder, CO (http://dss.ucar.edu/pub/toga_coare/).

Background

Flux data were taken during three legs: 11 November to 3 December 1992, 17 December 1992 to 11 January 1993 (most of the 24 December data is missing due to repairs), and 28 January to 16 February 1993. The ship was generally located at 1.7 S, 156E, except for a few days at the beginning of leg 2 and the end of leg 3 when it was at 0, 156E.

Data File Information

Data Processing

These tables are recent productions, but the numbers given here are still based on the real-time computations done at sea. Reprocessing of the data did not result in significant changes. The data format is similar to that used for the COARE Pilot, TIWE, and ASTEX data, with some notable exceptions.

(1) The ship speed and course is taken directly from GPS. The true wind has been corrected for the surface current based on IMET data supplied by Bob Weller (WHOI). Thus, these winds are referenced to the surface of the water, not the fixed earth. Typical corrections are on the order of 0.5 m/s. A mean correction of 0.2 m/s has been subtracted from the wind speed based on comparisons with the R/V Franklin.

(2) The ship's heading was taken from the ETL gyrocompass.

(3) Air temperature and humidities are given from two different sensors (the Vaisala HMP-35 and the OPHIR IR-2000). Both have crude corrections for daytime heating errors based on measured wind speed and solar radiation. The OPHIR mean humidity has been increased by 3% based on calibrations of the OPHIR hygrometer against the CSIRO psychrometer during TOGA COARE. The factory calibration for the Vaisala humidity sensor has been reduced 2% to force average agreement with the OPHIR.

(4) The humidity flux has been increased by 3% as discussed above. A humidity flux correction of an additional factor of 3.5% applied to previous releases to account for the physical separation (1 m) of the hygrometer from the anemometer (as per Kristensen, 1993) has been retracted.

(5) Several data quality indicators are given: relwind, J, Oph, Jm, and tilt. Good turbulence data is obtained only when the relative winds are within +/- 90 degrees of 0. (Note, sensible heat flux and stress are more sensitive to contamination and flow distortion and this 90-degree limit may not be strict enough.) J=1 implies the small-scale temperature fluctuations greatly exceeded a reasonable threshold. This implies contamination of the turbulence data by the ship's plume even though the relative wind direction is satisfactory. Oph is the standard deviation of the OPHIR hygrometer clear channel (in counts). When the optics are clean, this is around 5. Values exceeding 15 indicate serious contamination of the latent heat fluxes, either by sea salt or rain. This also occurs with clean optics when sun glint gets into the optics, usually for high solar zenith angles, particularly when the ship is rolling a lot. A tilt exceeding 10 degrees indicates a problem with the covariance fluxes. Jm is a velocity index indicating a maneuver of the ship during the record. Values of Jm exceeding 2 usually result in poor values for covariance stress; latent and sensible heat flux may be unaffected. Note that the turbulent fluxes are not to be used during precipitation exceeding 1 mm/hr. Both the OPHIR hygrometer and the sonic anemometer are subject to rain problems.

(6) Structure function parameters for temperature, humidity, and streamwise and vertical velocity are given. A Taylor's hypothesis correction has been applied (Wyngaard and Clifford, 1977). These have been determined by a very simple algorithm using the variance spectrum. They have not yet been eyeball checked. The ratio Cw/Cu should be 1.33; deviations of more than 30% from this value suggest problems with the data.

(7) Covariance, inertial-dissipation, and bulk values are given for stress and the heat fluxes.

(8) The bulk computations use the latest version of the Fairall-Bradley-Rogers algorithm (Fairall et al., 1996a) developed for the COARE program. It uses Smith (1988) for the drag coefficient, expressed as a Charnock relation for the roughness length:

zo =.011 u*^2/g+.11*nu/u*

Latent/sensible heat transfer coefficients are from Liu et al. (1979), but adjusted as per Fairall et al. (1996a). The mean vector wind is combined with a gustiness velocity equal to 1.25 times the convective scaling velocity (Godfrey and Beljaars, 1991), which is used to give more accurate fluxes at low mean wind speeds. The sea-surface humidity is 0.98 times the saturation humidity for pure water at the sea-surface temperature to account for the effect of sea water salinity. No Webb correction to the latent heat flux is included (it increases Hl an average of 4 W/m^2 for COARE). A cool skin correction to the sea surface temperature has been applied (Fairall et al., 1996b). Because the Moana Wave used a floating sea surface temperature sensor, no warm layer correction has been used.

9) The solar radiation has been increased 3% based on intercomparisons and postcalibrations.

10) The rain rate has been decreased to account for non-cosine response as per rain tower measurements done by Bradley. The interpretation of the headings on the tables is as follows:

Date: YYMMDDHHmmss, YY=year, MM=month, DD=day, HH=hour, mm=minute,ss=sec
Us:   ship speed (as described above)
U:    true wind speed at 15-m height
Tru:  true wind direction rel. to N (meteorological convention)
Rel:  relative wind direction
Hed:  the direction the ship's bow is pointing
Ts:   sea surface temp (no cool skin correction)
T:    Vaisala air temperature (about 15 m)
qs:   sea surface specific humidity  (g/kg) (no cool skin correction)
q:    Vaisala air specific humidity (about 15 m)
Hsc:  covariance sensible heat flux
Hsi:  inertial sensible heat flux
Hsb:  bulk sensible heat flux
Hlc:  covariance latent heat flux
Hli:  inertial latent heat flux
Hlb:  bulk latent heat flux
Tuc:  covariance surface stress (-wu part only)
Tui:  inertial-dissipation surface stress
Tub:  bulk surface stress
Rs:   solar irradiance
Rl:   longwave irradiance
Rain: precipitation (mm/hr)
J:    ship plume/contamination index (0 implies good conditions)
Oph:  standard deviation of OPHIR hygrometer clear channel counts (<15 implies
      reasonably clean optics).
Tlt:  mean wind vector tilt, degrees (<10 ok covariances)
Jm:   ship maneuver/contamination index, m/s (<2 implies good conditions)
Ct:   sonic temperature structure function parameter (K^2/m^.667)
Cq:   water vapor structure function parameter ((g/m^3)/m^.667)
Cu:   streamwise velocity structure function parameter ((m/s)^2/m^.667)
Cw:   vertical velocity structure function parameter ((m/s)^2/m^.667)
Hr:   sensible heat flux due to precipitation at droplet wet-bulb T
To:   OPHIR air temperature
Qo:   OPHIR specific humidity
Lat:  Latitude
Lon:  Longitude

The following lines of FORTRAN code can be used to read the file. Note that the FX.0 format will read whatever decimal places are already in the file.

      READ(LUNIN,505)ADUM     ! skip header line
505   FORMAT(A1)
700   READ(LUNIN,705,END=999)IYR,MON,IDAY,IHR,IMIN,ISEC,US,U,TRU,REL,
     1 HED,TS,T,QS,Q,IHSC,IHSI,IHSB,IHLC,IHLI,IHLB,TUC,TUI,TUB,RS,RL,
     2 RAIN,J,IOPH,ITLT,JM,CT,CQ,CU,CW,HR,T0,Q0,XLAT,XLON
705   FORMAT(6I2,F3.0,F6.0,3F5.0,4F6.0,6I5,3F7.0,3F6.0,4I4,
     1 4F9.0,3F7.0,2F9.0)
      GO TO 700
999   CONTINUE

This file contains covariance, inertial-dissipation, and bulk flux estimates. For the purposes of producing a smooth estimate of the fluxes for surface energy budget or model inputs, I suggest using the median of the three values. Be warned that the inertial-dissipation sensible heat flux is not very good because of light winds and ship plume contamination. If you are interesting in producing your own bulk parameterization, then you will want to use the turbulent fluxes. In this regard, the covariance fluxes are considered the standard. However, as I mentioned on the previous page, much of the direct turbulence data is invalidated by experimental conditions. As a share resource, the ship was operated to permit simultaneous oceanographic measurements rather than optimizing atmospheric flux measurements. Because the ship was operating in a drift mode for most of the time, turbulence data for wind speeds less than about 2 m/s are difficult to guarantee.

The following are a set of rules I suggest to get the best flux estimate from the data file:

IF 270>Rel>90 OR J=1 OR Rain>.5 OR Tilt>10 or Jm>2 THEN
      Use the bulk flux for all three fluxes

IF Oph>10 THEN
      Use the bulk flux for the latent heat flux

OTHERWISE
      Use the median of the three fluxes

Chris Fairall
NOAA/ERL/WPL
325 Broadway
Boulder, CO 80303
USA

email: cwf@etl.noaa.gov
Phone: (303) 497-3253
FAX: (303) 497-6978

References

Fairall, C.W., E.F. Bradley, J.S. Godfrey, G.A. Wick, J.B Edson, and G.S. Young, 1996b: The cool skin and the warm layer in bulk flux calculations. J. Geophys. Res., 101, 1295-1308

Godfrey and Beljaars, 1991, JGR, 96, 22043-22048

Kristensen, 1993

Kristensen, L., J. Mann, S.P. Oncley and J.C. Wyngaard, 1997: How close is close enough, when measuring scalar fluxes with displaced sensors? J. Atmos. Oceanic Technol. (in press)

Liu et al. 1979, J. Atmos. Sci., 36, 1722-1735

Smith, 1988, JGR, 93, 15467-15472

Wyngaard, J.C., and S.F. Clifford, 1978: Estimating momentum, heat and moisture fluxes from structure parameters. J. Atmos. Sci., 35, 1204-1211

 

http://www.ncdc.noaa.gov/oa/coare/catalog/data/air_sea_fluxes/moana_flux.html
Downloaded Thursday, 24-Jul-2008 04:20:20 EDT
Last Updated Thursday, 26-May-2005 17:01:02 EDT by David.P.Smith@noaa.gov
Please see the NCDC Contact Page if you have questions or comments.