4:15 PM - 4:30 PM
[SCG52-04] Instrument- and angle-dependent errors in GNSS-A seafloor crustal deformation observations clarified by water tank experiments
Keywords:GNSS-A, Seafloor geodesy
Japan Coast Guard has been directly observing seafloor crustal deformation for more than 20 years using a method called the GNSS-Acoustic ranging combination technique (GNSS-A). In GNSS-A observations, various observation error factors have been pointed out, not only errors due to GNSS but also errors due to acoustic ranging. Here, we briefly summarize the research conducted by the Japan Coast Guard to reduce the errors in acoustic ranging.
First, a transducer was installed on the bottom of S/V Meiyo in order to shorten the observation time and make the survey line geometrically well-balanced (Kawai et al., 2009). Secondly, analysis with a rigid array constraint was introduced, in which the relative positions of the four seafloor stations were fixed for the entire observation period, to reduce analysis errors (Matsumoto et al., 2008). Thirdly, multi-acoustic ranging method that enables shortening of observation time by continuously transmitting signals to multiple seafloor stations in a single acoustic ranging sequence (Yokota & Okumura, 2015; Yokota et al., 2017) was installed. Recently, the horizontal gradient of the underwater sound velocity structure has been incorporated into the analysis model (Yokota et al., 2019), and effects of observation plans such as survey lines and the array size of seafloor stations have been studied (Nakamura et al., 2021). Therefore, many error factors have been investigated.
Now, new error factors have been pointed out due to the improvement of such various error factors and the replacement of equipment in every few years. For example, errors depending on observation equipment such as sea-surface transducers and seafloor transponders, and the incident angle of sound waves have become apparent from the observations. Honsho et al. (2021) pointed out that the waveform changed depending on the incident angle of the acoustic signal.
In this study, in order to qualitatively and quantitatively examine the observation error of acoustic ranging that depends on the equipment and the incident angle, we conducted an experiment using a marine engineering water tank at the Institute of Industrial Science, the University of Tokyo. In this experiment, we prepared multiple types of sea-surface stations and seafloor stations, and performed 10 acoustic ranging each from various incident angles while changing their combinations. During the experiment, hydrophones were installed on both stations, and each station was fixed during the acoustic ranging.
By this experiment, we were able to reproduce the angular dependence that has been pointed out so far. On the other hand, it became clear that the angle-dependent error distribution differs depending on the device. In addition, similar to the actual observation, even if the acoustic ranging was performed at the same angle and the same observation equipment, errors equivalent to 1 to 2 wavelengths (equivalent to approximately 15-30 cm) were confirmed during the round trip. In this presentation, we will describe the outline of the water tank experiment and the results and problems obtained from the experiment.
First, a transducer was installed on the bottom of S/V Meiyo in order to shorten the observation time and make the survey line geometrically well-balanced (Kawai et al., 2009). Secondly, analysis with a rigid array constraint was introduced, in which the relative positions of the four seafloor stations were fixed for the entire observation period, to reduce analysis errors (Matsumoto et al., 2008). Thirdly, multi-acoustic ranging method that enables shortening of observation time by continuously transmitting signals to multiple seafloor stations in a single acoustic ranging sequence (Yokota & Okumura, 2015; Yokota et al., 2017) was installed. Recently, the horizontal gradient of the underwater sound velocity structure has been incorporated into the analysis model (Yokota et al., 2019), and effects of observation plans such as survey lines and the array size of seafloor stations have been studied (Nakamura et al., 2021). Therefore, many error factors have been investigated.
Now, new error factors have been pointed out due to the improvement of such various error factors and the replacement of equipment in every few years. For example, errors depending on observation equipment such as sea-surface transducers and seafloor transponders, and the incident angle of sound waves have become apparent from the observations. Honsho et al. (2021) pointed out that the waveform changed depending on the incident angle of the acoustic signal.
In this study, in order to qualitatively and quantitatively examine the observation error of acoustic ranging that depends on the equipment and the incident angle, we conducted an experiment using a marine engineering water tank at the Institute of Industrial Science, the University of Tokyo. In this experiment, we prepared multiple types of sea-surface stations and seafloor stations, and performed 10 acoustic ranging each from various incident angles while changing their combinations. During the experiment, hydrophones were installed on both stations, and each station was fixed during the acoustic ranging.
By this experiment, we were able to reproduce the angular dependence that has been pointed out so far. On the other hand, it became clear that the angle-dependent error distribution differs depending on the device. In addition, similar to the actual observation, even if the acoustic ranging was performed at the same angle and the same observation equipment, errors equivalent to 1 to 2 wavelengths (equivalent to approximately 15-30 cm) were confirmed during the round trip. In this presentation, we will describe the outline of the water tank experiment and the results and problems obtained from the experiment.