Study of Typhoon Morakot Loading Signal in Taiwan GPS Time Series

Thursday, 18 December 2014
Maxime Mouyen1, Linguo Yuan2, Anthony MEMIN3 and Benjamin Fong Chao1, (1)Academia Sinica, Taipei, Taiwan, (2)Southwest Jiaotong University, Faculty of Geosciences and Environmental Engineering, Chengdu, China, (3)University of Tasmania, Hobart, TAS, Australia
Redistribution of air masses deforms the Earth's crust and results in vertical and horizontal displacements of the ground. We investigated whether such displacements were recorded in the time series of the Taiwan GPS network while typhoon Morakot impacted Taiwan, early August 2009. As Taiwan concentrates numerous GPS stations (presently nearly 400) over a rather small area (400x200 km), we stacked the time series of 165 GPS stations to increase the signal to noise ratio and compared it to the displacements computed at the same sites. We assumed this stacking method is adequate due to the wave length of the deformation induced by typhoon, which is about 4 times wider than Taiwan. However, stacking also increases the amplitude of the common mode errors in the GPS time series, and classical methods to remove them removed the typhoon signal as well. Thus, comparing the observations with estimated displacements is also a way to identify common mode errors. Estimated displacements were computed by convolving load Green functions of displacement with surface pressure data released in the frame of the Year Of Tropical Convection (YOTC) project. The best agreement between the observation and the model is obtained for the vertical component of the displacements, which reach 1+/-0.3 cm on August 6th, the day where Morakot made landfall in Taiwan. The horizontal displacements, both the East-West and the North-South components, are more difficult to identify. For the East-West component, we may observe the arrival of Morakot, from the East, with an eastward displacement of the GPS stations of about 1+/-1 mm. For the North-South component, as Morakot crossed Taiwan in its middle, we divided the stacked time series in two groups depending on the location of the GPS stations, North or South of Morakot path. These two groups of GPS diverged by less than 1 mm at most when Morakot was over Taiwan, yet the uncertainty on the data prevented from firmly concluding on this displacement. Therefore, thanks to a high concentration of GPS stations, typhoon-induced displacements were observed in Taiwan, at least in the vertical component.