5:15 PM - 6:30 PM
[MIS19-P04] Detecting vertical movement in GNSS-acoustic measurement with various survey situations
Keywords:GNSS-A, seafloor vertical movement , sound speed in seawater
In GNSS-Acoustic geodetic survey, vertical motion and temporal change in sound speed are nearly in a trade-off relation, which makes it difficult to distinguish them. However, various survey styles have been proposed to resolve them. In this presentation, we will summarize the proposed survey styles, evaluate the precisions both in observed data and analytical simulation, and discuss the contribution of marine geodetic data to physical oceanography.
GNSS-A technique was first contrived by Spiess (1985), that detects horizontal displacement through point survey just above the center of a regular triangle, which consists of three precision transponders (PXPs) called an array. The concept is ingenious with quite a simple formulation that change in sound speed does not affect the estimate of horizontal movement as far as the sound speed is laterally stratified. Furthermore, only a meter-order estimate of individual PXP positions is needed if the ship can keep its position within a 10 m radius. In practice, horizontal movement is dominant rather than vertical in most tectonic phenomena of interest. Kido et al. (2006) remove the restriction of point keeping ability by representing the effect of sound speed change as vertically normalized traveltime delay (called NTD: Nadir Total Delay) because they employed a poorly controllable moored buoy. On the contrary, Fujita et al. (2006) and Ikuta et al. (2008) separately developed a method to determine 3-D position of individual PXPs based on moving surveys, in which the vertical motion was still poorly resolved than horizontal was.
By introducing multi-campaign analysis and other refinements, today’s precision in vertical motion has been much improved. Although vertical motion cannot intrinsically be resolved by traditional point survey, Honsho et al. (2017) estimated vertical motions by utilizing variation in the incident angles for the site of 6-PXPs of small and large triangles. We recently constructed a couple of 4-PXP sites and started surveys consisting of a traditional triangle and an additional centered PXP to also make incident angle variation with less cost. In our combined moving and point surveys, a significant improvement in vertical precision was observed for the 4-PXP site compared to the triangle site, which is also confirmed numerically by Tomita et al. (2019) and analytically by our Dilution of Precision (DOP) based evaluation.
Further quantitative discussions will be presented on the computation of sound speed with the so-called UNESCO’s equation, on the observed horizontal gradient in sound speed, and on the temporal variation in NTDs.
GNSS-A technique was first contrived by Spiess (1985), that detects horizontal displacement through point survey just above the center of a regular triangle, which consists of three precision transponders (PXPs) called an array. The concept is ingenious with quite a simple formulation that change in sound speed does not affect the estimate of horizontal movement as far as the sound speed is laterally stratified. Furthermore, only a meter-order estimate of individual PXP positions is needed if the ship can keep its position within a 10 m radius. In practice, horizontal movement is dominant rather than vertical in most tectonic phenomena of interest. Kido et al. (2006) remove the restriction of point keeping ability by representing the effect of sound speed change as vertically normalized traveltime delay (called NTD: Nadir Total Delay) because they employed a poorly controllable moored buoy. On the contrary, Fujita et al. (2006) and Ikuta et al. (2008) separately developed a method to determine 3-D position of individual PXPs based on moving surveys, in which the vertical motion was still poorly resolved than horizontal was.
By introducing multi-campaign analysis and other refinements, today’s precision in vertical motion has been much improved. Although vertical motion cannot intrinsically be resolved by traditional point survey, Honsho et al. (2017) estimated vertical motions by utilizing variation in the incident angles for the site of 6-PXPs of small and large triangles. We recently constructed a couple of 4-PXP sites and started surveys consisting of a traditional triangle and an additional centered PXP to also make incident angle variation with less cost. In our combined moving and point surveys, a significant improvement in vertical precision was observed for the 4-PXP site compared to the triangle site, which is also confirmed numerically by Tomita et al. (2019) and analytically by our Dilution of Precision (DOP) based evaluation.
Further quantitative discussions will be presented on the computation of sound speed with the so-called UNESCO’s equation, on the observed horizontal gradient in sound speed, and on the temporal variation in NTDs.