In the present study, the main objective was to investigate the characteristics of turbulent fluxes and to modify the flux algorithm. Turbulent flux data was observed at the Ieodo Ocean Research Station (IORS) for several years. This data explains the...
In the present study, the main objective was to investigate the characteristics of turbulent fluxes and to modify the flux algorithm. Turbulent flux data was observed at the Ieodo Ocean Research Station (IORS) for several years. This data explains the characteristics of middle-latitude open sea such as seasonal variation.
Reliable observation data was required to modify the flux algorithm. In this study, the quality control methods for flux measurement at the Ieodo Ocean Research Station (IORS) included weather check, the quality control algorithms of Vickers and Mahrt (1997, VM checks), flag check, and direction check. Generally, rain is recognized as a main cause of flux measurement problems. But errors are also occurred due to strong fog, and high waves, and strong wind. The sensor used to observe H2O is especially sensitive in these environments. The effects of the three tilt correction method (double rotation, triple rotation, and planar fit) were also compared. When the observation period was shorter than weeks, the double rotation method was more appropriate instead of planar fit method. But the analysis of momentum flux corrected by the double rotation method required needed more attention than turbulent heat fluxes.
The periods of IORS observation were sufficiently long to categorize various criteria. Turbulent flux fluxes were analyzed according to stability level and season. Stable (z/L ≥ 0.4) cases compared about 5% of cases, near neutral (-0.4 < z/L < 0.4) cases comprised about 70%, and unstable (z/L ≤ -0.4) ratios comprised about 25%. Seasonal variation was especially important in middle-latitudes.
The relationship between turbulent fluxes and wind speed exhibited different features, depending on the season. In the winter (DJF), latent heat flux is rapidly increased when wind speed increased. In SON, latent heat flux was more scattered than in DJF. Sensible heat flux was similar to latent heat flux. Friction velocity showed good agreement in the relationship between wind speed and friction velocity in a neutral state. The seasonal relationship between turbulent fluxes and wind speed was again classified according to stability. In an unstable state, heat fluxes were closely correlated with wind speed. Friction velocities in near neutral states were greater than friction velocities in other states.
The transfer coefficient also varied seasonally. In the unstable condition of DJF, CE increased as wind speed increased. CH exhibited a clear relationship with wind speed. Generally, CE and CH were treated as the same value in the model. But IORS results data denoted CE and CH as begin very different. However, CD did not show seasonal variation, but it did exhibit different features based on stability.
Droplets played a large role in latent heat transfer between the ocean and the atmosphere. Sea spray was not included in the COARE algorithm, though it could influence air-sea heat fluxes. In this study, COARE3.0 was modified to include the sea spray process. The transfer coefficients for heat fluxes were adjusted by scale factor. Significant wave height was modified to use different parameterizations based on the wave peak period. Modified turbulent heat fluxes showed good agreement with the observation data and the modified drag coefficient was well-fitted to the observation data. In low wind conditions, however, the drag coefficients still disagreed with the observation data.
Spray effects were most evident in the latent heat flux. When wind speed was greater than 10ms-1, spray significantly impacted the total latent heat flux. The spray effects of latent heat flux were controlled by wind speed, the difference between surface humidity and air humidity and significant wave height.
To reduce uncertainty in the air-sea flux algorithm, new parameterizations were developed based on field and laboratory experiments conducted by many researches. However flux algorithms were limited to certain conditions (calm to light winds and high wind speeds). The results of present study should help facilitate some understanding of air-sea interaction in the ocean.