Tables S1--S4. These give the parameters of the dislocation models and the resulting displacement values that we used to correct the campaign GPS time-series for coseismic offsets due to the 2003 Secretary Island (Fiordland), 2004 Macquarie and 2007 Fiordland (George Sound) earthquakes. These models provide a reasonable to good fit to observed data, but they are not necessarily the best dislocation model approximations obtainable for these earthquakes.

Tables S5--S7. These give the parameters of our variable-slip dislocation solutions CP1, CP2 and CP3 for the 2009 Dusky Sound earthquake. These are provided in separate tab-delimited text files: Beavan_DuskySound_TableS5_CP1.txt, Beavan_DuskySound_TableS6_CP2.txt and Beavan_DuskySound_TableS7_CP3.txt.

Figures S1--S2. These show the quadtree partitioning of the DInSAR images used in our analysis.

Figure S3. This shows the trade-off curve between solution misfit and solution roughness that is used to select the value of the κ2 hyper-parameter.

Figures S4--S9. These show variations in the CP3 solution as the κ2 and λ2 hyper-parameters are varied, the dimensions of the fault surface are increased and the DInSAR data are down-weighted compared to the GPS data.

Figures S10--S13. These contain plots of the quadtree-sampled DInSAR images, model fits and residuals used in our analysis but not included in the main paper.

GJI_4798_sm_Supplement.zip4220KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.