CO2 transfer velocity at air-sea interface associated with different mechanisms such as turbulence and wave breaking, which are closely related with wind speed, has been studied from satellite data in this work. The determination of wind friction velocity from satellite-derived wind data will take an important role in computation of CO2 flux transfer. It is necessary for relation between wind speed and wind friction velocity to determine that of relation between non-dimensional roughness length and wave age, included with all parameters (wind, wave). Because of the large variation of the non-dimensional roughness length, it is difficult to describe wave influence by the single wave age.
In this study we observed directional wave spectra at the Hiratsuka Tower of National Research Institute for Earth Science and Disaster Prevention (NIED) in Sagami Bay, Japan. As the result, the data set was separated into two kinds of directional spectrum types, Case 1 contains only swell and wind waves, but Case 2 consists of a lot of component waves due to several different wind fetch in the bay. The results show that Case 2 is associated with the large variation in the relation between non-dimensional roughness length and wave age. As wave age is associated with fetch, and a new relationship between non-dimensional roughness length and wave age was proposed, based on in situ wind and wave data after considering wind fetch effect.
A new method was proposed to estimate u*, which is based on the new relationship between non-dimensional roughness and wave age, after considering fetch and wave directionality. We estimated the wave frequency from two methods: 3/2 powers law (Toba, 1972) and WAM model (WAMDI, 1988). The results are compared with the results estimated from Charnock formula (1955), and the influence of wave effects on the wind stress is also discussed. A new relationship was established to determine CO2 exchange coefficient based on whitecap model (Monahan and Spillane 1984), using U10-u* relationship in the North Pacific Ocean, satellite data of NOAA/AVHRR (SST) and DMSP/SSM/I (wind speed) in Oct., Nov., and Dec. 1991. The CO2 exchange coefficient estimated by other models (Wanninkhof, 1992; Liss and Merlivat, 1986; Tans et al., 1990) is also compared with these results. Also, we investigated wave effect for CO2 exchange coefficient in numerical simulation using Whitecap model (Monahan and Spillane, 1984) and the new relationship between non-dimensional roughness length and wave age. These results show the importance of wave breaking effect. |
|
|