3:45 PM - 4:00 PM
[SCG55-31] Short-period fluctuation of sound velocity field in ocean observed in GNSS-A surveys and internal gravity wave
Keywords:sound velocity, internal gravity wave, GNSS-acoustic survey, XBT
The GNSS-Acoustic (GNSS-A) observation method, one of the seafloor geodetic observation methods, combines GNSS positioning by onboard antennas and undersea acoustic ranging to capture crustal deformation of the seafloor as a change in the position of an array consisting of multiple seafloor transponders. Since acoustic ranging is used, the positioning accuracy depends on the assumed undersea sound speed field. Sound speed varies with time due to oceanic phenomena. Assuming a time-dependent stratified sound speed structure in the ocean, changes in mean sound speed throughout the water column can be expressed as changes in NTD (Nadir Total Delay), which is the vertical projection of the traveltime residual. The short period oscillations with a period of 30-60 minutes in the NTD time series, superimposed on oscillation with tidal frequency, are often observed and are thought to be due to internal gravity waves. It is currently difficult to represent such short period oscillation to the GNSS-A analysis because wavelength is too short and no longer regarded as linear gradient within the array dimension. Therefore, it is crucial to know the wavelength of typical internal gravity wave at the survey site.
GNSS-A observations were conducted during a cruise in the Kumano-nada on Dec. 19 –24 in 2016. Temperature profiles were measured successively every 10 minutes for 2 hours using XBTs (eXpendable Bathy Thermograph) during a GNSS-A point survey at the center of the six transponder array. The purpose of this study is to extract the characteristics of the internal gravity wave from the temporal variation of the temperature data and to investigate how it affects the undersea acoustic ranging.
First, we calculated XBT-derived sound speed profiles and corresponding NTD for two hours from the temperature profiles. The nominal accuracy in XBT temperature, ~0.2 °C, results in larger error evaluation in NTD than its signal. However, temperatures in the deepest part of the individual profiles, where small temperature variation is expected, have only 0.02 °C in standard deviation owing to the same manufacture lot is used, in which the error is reduced by one order. To reduce the bias error among XBT profiles, we calibrated the XBT data using the more accurate XCTD (eXpendable Conductivity, Temperature and Depth) measurement, which is verified by the fact that scatters both in deepest part and surface mixing layer are simultaneously reduced by this procedure. We compared the NTD based on thus calibrated temperature profiles and acoustic derived NTD and found in good agreement. A small difference between them can be explained by ~10 cm errors in GNSS ellipsoidal height and predicted sea surface height by geoid and tide models.
Next, we characterize the oscillation of XBT temperature profiles by plotting iso-thermal depth every 0.2 °C and found oscillations with a period of about 1 hour. The observed period satisfies the minimum limits (3700 s in deeper and 740 s in shallower parts) predicted by Brunt-Väisälä frequency estimated using the vertical gradient of potential density profile calculated from temperature and salinity obtained from XCTD measurement. Such short period oscillation in NTD is often observed in GNSS-A measurement, indicating that it might be a universal phenomenon. In the future, we will calculate the possible mode of the internal gravity wave from the vertical density gradient and identify the dominant mode in the observation. Then, we will determine the wavelength of the internal gravity wave from the period and phase velocity of the identified mode and discuss how it affects the GNSS-A analysis.
GNSS-A observations were conducted during a cruise in the Kumano-nada on Dec. 19 –24 in 2016. Temperature profiles were measured successively every 10 minutes for 2 hours using XBTs (eXpendable Bathy Thermograph) during a GNSS-A point survey at the center of the six transponder array. The purpose of this study is to extract the characteristics of the internal gravity wave from the temporal variation of the temperature data and to investigate how it affects the undersea acoustic ranging.
First, we calculated XBT-derived sound speed profiles and corresponding NTD for two hours from the temperature profiles. The nominal accuracy in XBT temperature, ~0.2 °C, results in larger error evaluation in NTD than its signal. However, temperatures in the deepest part of the individual profiles, where small temperature variation is expected, have only 0.02 °C in standard deviation owing to the same manufacture lot is used, in which the error is reduced by one order. To reduce the bias error among XBT profiles, we calibrated the XBT data using the more accurate XCTD (eXpendable Conductivity, Temperature and Depth) measurement, which is verified by the fact that scatters both in deepest part and surface mixing layer are simultaneously reduced by this procedure. We compared the NTD based on thus calibrated temperature profiles and acoustic derived NTD and found in good agreement. A small difference between them can be explained by ~10 cm errors in GNSS ellipsoidal height and predicted sea surface height by geoid and tide models.
Next, we characterize the oscillation of XBT temperature profiles by plotting iso-thermal depth every 0.2 °C and found oscillations with a period of about 1 hour. The observed period satisfies the minimum limits (3700 s in deeper and 740 s in shallower parts) predicted by Brunt-Väisälä frequency estimated using the vertical gradient of potential density profile calculated from temperature and salinity obtained from XCTD measurement. Such short period oscillation in NTD is often observed in GNSS-A measurement, indicating that it might be a universal phenomenon. In the future, we will calculate the possible mode of the internal gravity wave from the vertical density gradient and identify the dominant mode in the observation. Then, we will determine the wavelength of the internal gravity wave from the period and phase velocity of the identified mode and discuss how it affects the GNSS-A analysis.