Comparative assessment of NIF and PCAIM for inverting equivalent fault slip in far-field scenarios

The study utilized GPS deformation time series, network inversion filter (NIF), and principal component analysis based inversion method (PCAIM) to invert for postseismic equivalent fault slip, obtaining the spatiotemporal evolution characteristics of regional fault slip, and investigating the performance of the two methods in a far-field postseismic inversion scenario. Firstly, the NIF inverted equivalent slip is influenced by the different baseline motion scaling parameter . When mm/a 1/2 , the influence on the slip amount basically stabilizes, with the change value tending to be within 5 mm. Secondly, PCAIM exhibits a stronger dependence on the fault model compared to NIF, and the inverted slip distribution is more sensitive to changes in the number of discretized subfaults. Finally, taking the 2011 regional equivalent slip in the Tokai of Japan as an example, the PCAIM inversion time (0.14h) is far shorter than the NIF inversion time (7.79h). The inverted postseismic slip propagation area (34.0°N–35.6°N,136.8°E–138.0°E) is significantly larger than the NIF inverted propagation area (34.8°N–35.8°N,137.2°E–138.4°E). The maximum slip amount is 1.81cm larger than that of NIF, and the propagation range is wider. Concurrently, the MAE and RMSE for the surface displacement vectors predicted by PCAIM are both lower than those predicted by NIF, indicating that the PCAIM method exhibits superior surface displacement prediction performance in this scenario.