Cumulative Coulomb Stress Triggering as an Explanation for the Canterbury (New Zealand) Aftershock Sequence: Initial Conditions Are Everything?

Using 2 years of aftershock data and three fault-plane solutions for each of the initial M7.1 Darfield earthquake and the larger (M>6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M >6$$\end{document}) aftershocks, we conduct a detailed examination of Coulomb stress transfer in the Canterbury 2010–2011 earthquake sequence. Moment tensor solutions exist for 283 of the events with M≥3.6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M \ge 3.6$$\end{document}, while 713 other events of M≥3.6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M \ge 3.6$$\end{document} have only hypocentre and magnitude information available. We look at various methods for deciding between the two possible mechanisms for the 283 events with moment tensor solutions, including conformation to observed surface faulting, and maximum Δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta$$\end{document}CFF transfer from the Darfield main shock. For the remaining events, imputation methods for the mechanism including nearest-neighbour, kernel smoothing, and optimal plane methods are considered. Fault length, width, and depth are arrived at via a suite of scaling relations. A large (50–70 %) proportion of the faults considered were calculated to have initial loading in excess of the final stress drop. The majority of faults that accumulated positive Δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta$$\end{document}CFF during the sequence were ‘encouraged’ by the main shock failure, but, on the other hand, of the faults that failed during the sequence, more than 50 % of faults appeared to have accumulated a negativeΔ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta$$\end{document}CFF from all preceding failures during the sequence. These results were qualitatively insensitive to any of the factors considered. We conclude that there is much unknown about how Coulomb stress triggering works in practice.