The ability of the non-tidal oceanographic model for the ocean bottom pressure time series in the dense observation networks around Japan: towad understanding the transient crustal deformation

Abstract:

Ocean Bottom Pressure-gauges (OBPs) are very useful to understand the vertical crustal deformation in offshore region. The OBP time series usually contain tidal and non-tidal components, specific sensor drift, and vertical crustal deformation signal. In particular, the contribution of the non-tidal component is one of the obstacles in interpreting OBP time series. Inazu et al. (2012) developed a global barotropic ocean model forced by atmospheric disturbances for the detection of seafloor vertical displacements from in situ OBP data. In recent years, a very dense cabled OBP network has been developed around Japan (e.g., S-net and DONET). In this study, we assess the performance of the model of Inazu et al. (2012) for those OBP network and discuss the ability to monitor the crustal deformation after removing non-tidal component.

We analyzed the DONET OBP time series for the whole year of 2017. The obtained OBP time series were averaged hourly and “tide killer” filter (Hanawa and Mitsudera, 1985) were applied to eliminate the tidal components. The model by Inazu et al. (2012) were applied to the tide-free time series. We used the JRA-55 for the input wind stress as the driving force of the model. To evaluate the performance of the model, the standard deviation (SD) was calculated for each of the 30 time windows (1, 2, ...30 days) before and after the model was applied. The next, we calculated the reducing value of the SD before and after the model applying.

In the DONET network, almost all stations showed the SD reduction longer than 10-day time window. The improvement of the SD was, however, limited with shorter than 20-day time window. In contrast, SD values increased in the 3-9 day time window. This result indicates that the model does not adequately explain the observation in a short time window.

In the presentation, we will show the performance of the model for more longer time series and discuss in more detail the ability of DONET to detect the transient crustal deformation in different time windows.