Visualization of Seismicity along Subduction Zones: Toward a Physical Basis

Abstract

It has been a common practice to plot epicenters and hypocenters of earthquakes as symbols of equal size on geologic maps and cross-sections. Such plots can be highly effective. For instance, in regions dominated by high-angle faults, one can readily identify traces of seismically active faults by visually connecting dots that represent epicenters in map view. However, the practice lacks a clear physical basis and opens the door to arbitrary decisions regarding symbol sizes that could obscure important features in seismicity or even create misleading visual effects. Figure 1. ▴ Cross-sections normal to the Tonga Trench (marked by an inverted triangle), comparing visual effects of four different representations of the same data set. Each hypocenter is plotted as a solid circle whose radius is (A)that of a circular earthquake rupture estimated according to Equation 1, (B) set to a uniform value of 15 km, (C) set to a uniform value of 2 km, and (D) scaled to moment magnitude ( e.g., Mw 6 = 9 km). Only case A has a physical basis. In B and C, different choices of symbol sizes result in vastly different, biased visual impressions. For instance, even though outboard earthquakes (OE) release a significant amount of seismic moment (A), they appear to be negligible in conventional plots (C). In principle, D could preserve the relative size of earthquakes but is not effective in practice because of the highly compressed magnitude scale and arbitrary choices of scale. See Figure 2 for location of the crosssection. Throughout this article, we used the database of large to moderate-sized ( mb ≥ 5) earthquakes relocated by Engdahl et al. (1998) (for years 1964-2001, including the latest updates) and those from the Preliminary Determination of Epicenters (for year 2002). These catalogs are supplemented by precisely determined focal depths of more than …