9:45 AM - 10:00 AM
[SGD02-04] New Effective Method to Detect Coseismic Gravity Changes from Short-term GRACE Level-2 Data
Keywords:satellite, coseismic gravity change, short-term data, gaussian filtering
To detect coseismic gravity changes from satellite gravimetry level-2 data, low-pass or band-pass filters are often used to reduce various types of noises. However, some gravity signals in wavelengths at high accuracies can also be weakened in this process. This is particularly problematic in the first several years following a huge earthquake because sufficient data are not accumulated to cancel out random noises in those short-term data.
In this research, we developed a new method to obtain the signals of coseismic gravity changes rapidly and accurately from satellite gravimetry level-2 data and applied the method to the 2011 Tohoku-oki earthquake (Mw 9.0). In this new method, time series of each Stokes coefficient (level-2 data) are directly analyzed by least squares before they are converted to the spatial domain and important coefficients are detected. In this process, we also attempt to avoid misdetection by considering the accuracy of each coefficient provided in the level-2 data. The resultant figure of coseismic gravity changes based only on the data 14 months before and 4 months after the earthquake is very similar to that on the 10-or-more-year data before and after the earthquake. This method can obtain high S/N results of coseismic gravity changes by future huge earthquakes.
In this research, we developed a new method to obtain the signals of coseismic gravity changes rapidly and accurately from satellite gravimetry level-2 data and applied the method to the 2011 Tohoku-oki earthquake (Mw 9.0). In this new method, time series of each Stokes coefficient (level-2 data) are directly analyzed by least squares before they are converted to the spatial domain and important coefficients are detected. In this process, we also attempt to avoid misdetection by considering the accuracy of each coefficient provided in the level-2 data. The resultant figure of coseismic gravity changes based only on the data 14 months before and 4 months after the earthquake is very similar to that on the 10-or-more-year data before and after the earthquake. This method can obtain high S/N results of coseismic gravity changes by future huge earthquakes.