Reply to comment by Arthur Frankel on “Why Earthquake Hazard Maps Often Fail and What to do About It”

Abstract

Earthquake hazard maps are produced to tell the public that some areas are more at risk than others, and to help governments prepare appropriately. To the extent a hazard map proves accurate, it helps society protect itself against earthquake damage. On the other hand, an inaccurate hazard map may lead to either inadequate preparation or suboptimal use of limited resources (Stein and Stein, in press). As we noted in our original paper, Stein et al. (2012), cited as SGL2012 below, the March 11, 2011 Tohoku, Japan, earthquake is a classic and unfortunate example of what can go wrong. Japan's national hazard map said (and still says) that the Tokai region (along the Pacific coast from Shizuoka to Nagoya) and the Tonankai and Nankai regions (farther to the west) had the highest hazard, with the hazard in Tohoku being considerably smaller. This indirectly led to a relatively low level of countermeasures in Tohoku generally and at the Fukushima Dai-ichi nuclear power plant in particular, which contributed to an unnecessarily large number of casualties and property damage, including the level-7 nuclear accident at Fukushima (Geller, 2011; Noggerath et al., 2011). The 2011 Tohoku earthquake is, alas, not an isolated example. As pointed out by SGL2012, the 2003 Algeria earthquake, 2004 Morocco earthquake, 2008 Wenchuan earthquake, and 2010 Haiti earthquake all also occurred in regions mapped as relatively low hazard. SGL2012 also cited work (Reyners, 2011) that showed that the February 22, 2011 Mw6.3 earthquake, which did considerable damage in Christchurch, New Zealand, produced much stronger ground motion than the hazard map predicted would occur in the