The California Earthquake Advisory Plan: A History

Abstract

Since 1985, the California Office of Emergency Services (Cal OES) has issued advisory statements to local jurisdictions and the public following seismic activity that scientists on the California Earthquake Prediction Evaluation Council view as indicating elevated probability of a larger earthquake in the same area during the next several days. These advisory statements are motivated by statistical studies showing that about 5% of moderate earthquakes in California are followed by larger events within a 10-km five-day space-time window (Jones, 1985; Agnew and Jones, 1991; Reasenberg and Jones, 1994). Cal OES issued four earthquake advisories from 1985 to 1989. In October 1990, the California Earthquake Advisory Plan formalized this practice, and six Cal OES Advisories have been issued since then. This article describes that protocol's scientific basis and evolution.

Full text

○
E
The California Earthquake Advisory Plan:
A History
by Evelyn Roeloffs and James Goltz
ABSTRACT
Since 1985, the California Office of Emergency Services (Cal
OES) has issued advisory statements to local jurisdictions and
the public following seismic activity that scientists on the Cal-
ifornia Earthquake Prediction Evaluation Council view as in-
dicating elevated probability of a larger earthquake in the same
area during the next several days. These advisory statements are
motivated by statistical studies showing that about 5% of mod-
erate earthquakes in California are followed by larger events
within a 10-km five-day space–time window (Jones, 1985;Ag-
new and Jones, 1991;Reasenberg and Jones, 1994). Cal OES
issued four earthquake advisories from 1985 to 1989. In Oc-
tober 1990, the California Earthquake Advisory Plan formal-
ized this practice, and six Cal OES Advisories have been issued
since then. This article describes that protocol’s scientific basis
and evolution.
Electronic Supplement: Table listing earthquakes of magnitude ∼5
and larger in and near California between 1 January 1985 and
31 December 2015.
INTRODUCTION
California faces enormous hazard from the strong earthquakes
that will inevitably strike its population centers. Such events
have low daily probabilities and remain unpredictable, all but
precluding advance warning. Yet statistical studies since 1985
have shown that a moderate earthquake in California has about
a 5% chance of being followed by a larger eventoccurring nearby
within several days (Jones, 1985;Agnew and Jones, 1991;Re-
asenberg and Jones, 1994). Emergency managers can exploit this
brief period of elevated hazard to take advance actions that
enhance preparedness, and to craft public announcements that
could lessen the impact of a major seismic event (Jordan et al.,
2014;Goltz, 2015).
Temporary earthquake probability gains following po-
tential foreshocks have also been demonstrated statistically
in Nevada (Savage and dePolo, 1993), Japan (Maeda, 1996),
and, for large earthquakes, worldwide (Reasenberg, 1999).
Only California, however, has a protocol for using these find-
ings to promote public safety. The purpose of such protocols
became clear in 2009, when three months of vigorous seismic-
ity near L’Aquila, Italy, culminated in an M5.9 earthquake on
6 April that killed 297 people (Jordan, 2013). Although seis-
mologists on a government-appointed commission knew that
the elevated seismicity indicated a larger earthquake was tem-
porarily more likely, nonscientist officials had announced
publicly on March 31 that the seismic activity posed no dan-
ger (Jordan, 2013). Seismologists were accused of criminal
negligence and manslaughter for providing incomplete, im-
precise, and contradictory information (Hall, 2011) that led
residents to remain in seismically vulnerable buildings.
Almost 25 years before the L’Aquila earthquake, the state
of California had implemented protocols that enable
emergency managers to take advance actions when a potential
foreshock to a damaging earthquake occurs. This protocol,
formalized as the California Earthquake Advisory Plan, contin-
ues to be used today but seems little known among earthquake
scientists. The history presented below draws on records main-
tained by the California Geological Survey, California Office
of Emergency Services (Cal OES), and the personal files of one
of us (J. G.) who served as Executive Secretary of the California
Earthquake Prediction Evaluation Council (CEPEC) from 1986
to 1993 and as Cal OES ex-officio representative to CEPEC
from 2007 to 2011. We hope this account of California’s
approach to advising the public of temporarily heightened
earthquake probability can inform future efforts to utilize
earthquake science for public safety.
SCIENTIFIC BASIS AND INCEPTION OF THE
ADVISORY PLAN
In the 1970s, optimism about earthquake prediction capability
led to concerns of detrimental societal responses to such pre-
dictions (Geschwind, 2001). Following public responses to sev-
eral earthquake forecasts, officials at Cal OES and at California
Division of Mines and Geolog y (CDMG; now California
Geological Survey) sought expert earthquake-science advice.
In particular, Cal OES, responsible for advising the California
Governor whether conditions warrant proclaiming a State of
Emergency, wanted rapid access to experts who could judge
784 Seismological Research Letters Volume 88, Number 3 May/June 2017 doi: 10.1785/0220160183
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

whether smaller earthquakes on faults that had generated
damaging events in the past might be foreshocks.
The California Earthquake Prediction Evaluation Council
(CEPEC)
Cal OES established CEPEC as an informal advisory commit-
tee in 1974, chaired by California State Geologist James E.
Slosson. Although its name implies that CEPEC reviews only
predictions, the council also reviews research or data that sug-
gest an increase in seismic or volcanic potential affecting
California. Currently, California remains the only state with
a formal body of this type. (The National Earthquake Predic-
tion Evaluation Council [NEPEC] is a Federal Advisory Com-
mittee that advises the Director of the U.S. Geological Survey
[USGS] on similar matters outside California, but NEPEC has
rarely met in urgent sessions to evaluate potential foreshocks.)
By 1976, CEPEC had evolved into a formally appointed
body of nine scientists with expertise in geology, volcanology,
seismology, and geophysics (CEPEC, 1977). The California
State Geologist, as Chair of CEPEC, interfaces between these
scientists and the California state emergency management
agency, screening information that may have public policy im-
plications. CEPEC members are recommended by the Director
of Cal OES, in consultation with the State Geologist, and ap-
pointed to four-year terms by the Governor. CEPEC always
invites an executive-level representative from Cal OES to mon-
itor CEPEC meetings for policy issues, including closed rapid
assessment scientific discussions, but some meetings have taken
place without such a representative present.
The first prediction that CEPEC evaluated was in 1976,
when seismologist James Whitcomb hypothesized that anoma-
lous seismic velocity changes near Palmdale, California, might
be precursors to an M5.5–6.5 earthquake within the next year.
Chaired by Acting California State Geologist Thomas E. Gay,
CEPEC did not endorse this prediction, but judged the data
sufficiently suggestive to warrant further scientific investigation
(Ingram, 1983;Geschwind, 2001). Prior to CEPEC’s evalu-
ation, OES Director Charles Manfred had already written local
governments and state agencies urging them to review and up-
date their emergency plans (CEPEC, 1976), and OES took no
further action.
The California State Geologist, with approval from the
Director of the California state agency responsible for emer-
gency management, has authority to convene CEPEC for a
rapid assessment conference call when increased earthquake
activity in or near California raises the probability of a larger
earthquake that could affect population centers in California.
As of 2015, the criteria for convening CEPEC are met by
earthquakes larger than M5, or for three or more closely-
spaced closely-timed events larger than M3.5 (J. Parrish, writ-
ten comm., 2015). The State Geologist will also consider con-
vening CEPEC in rapid-assessment mode if recognized
earthquake experts so urge, even if these criteria are not met.
There is no automatic mechanism to trigger a CEPEC tele-
conference. From 1986 through 2016, CEPEC has convened
for 17 rapid assessments.
CEPEC’s rapid assessment conference calls are confiden-
tial, as is the written summary statement that CEPEC sends to
Cal OES, recommending whether to issue an Advisory. There-
fore, we cannot describe the scientific arguments on which
individual CEPEC recommendations were based. The Direc-
tor of Cal OES has discretion as to whether to act upon
CEPEC’s recommendation, and is responsible for transmitting
Advisories to local jurisdictions.
CEPEC members are immune from legal liability in their
deliberations regarding specific earthquake predictions or
assessments of related research. At the time of this writing,
CEPEC has never invoked this immunity.
Foreshock Probability Research
USGS seismologist Lucy Jones (1985) first established the
statistical foundation for earthquake advisory statements in
California. Based on an analysis of all southern California
earthquakes with M≥3between 1932 and July 1983 (4811
events), Jones (1985) estimated the probability that an M≥3
earthquake in southern California will be followed by a larger
magnitude event within 5 days and 10 km to be ∼6%. Further,
Jones showed that the probability that an earthquake would be
followed by an M≥5mainshock increases with the magnitude
of the foreshock from less than 1% at M3 to 6.5% for M≥5.
The probability of a larger follow-on earthquake is the highest
immediately after the candidate foreshock, with a significant
percentage of the increased probability concentrated in the first
30 min.
Another way to view the Jones (1985) result is that during
the five-day 10-km time–space window, the most probable out-
come is that no larger shock will occur: 94% of advisory state-
ments based on these statistics would be false alarms. Despite
these odds, California state officials decided that advisories mo-
tivated by potential foreshocks were still worthwhile. From an
emergency management viewpoint, even the low probability of
a significant earthquake is actionable if the potential conse-
quences include damage and casualties. Preparedness and mit-
igation actions need not be extreme, and can be scaled to a low
probability of occurrence, as discussed below and described by
Reitherman (1986) and Goltz (2015).
No reliable identifying characteristics of foreshocks are
known in advance of the mainshock; Cal OES relies on
CEPEC’s judgment in deciding whether to issue an Advisory.
Table 1includes all of the earthquakes and earthquake swarms
for which CEPEC has been convened for rapid assessment.
Figure 1is a map of earthquakes M>5that have affected
California since 1 January 1985, indicating which were assessed
by CEPEC; these events are also listed in ⒺTable S1, available
in the electronic supplement to this article. We selected the
case histories described here to illustrate the advisory protocol’s
use and evolution since 1985.
PREPLAN ADVISORIES 1985–1990
California’s first four earthquake advisories, which predated the
formal Advisory Plan, revealed the difficulties of coordinating
Seismological Research Letters Volume 88, Number 3 May/June 2017 785
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

Table 1
Earthquakes and Earthquake Sequences That Have Prompted Review by the California Earthquake Prediction Evaluation
Council (CEPEC), or That Are Described in the Text for Other Reasons
Date and Time
(yyyy/mm/dd hh:mm:ss.ss)
Latitude
Longitude (°)
Depth
(km) Magnitude Location Action
1985/06/18 00:12:55.23 UTC 32.691 5.34 ML3.7 San Diego Advisory 1
1985/06/17 05:12 p.m. PDT −117.15
1985/06/18 03:22:28.67 UTC 32.679 5.06 ML3.9 San Diego
1985/06/17 08:22 p.m. PDT −117.151
1985/06/18 04:28:15.28 UTC 32.684 5.34 ML3.7 San Diego
1985/06/17 09:28 p.m. PDT −117.152
1986/07/08 09:20:44.56 UTC 33.999 9.47 ML5.6 North Palm Springs CEPEC review, no advisory
recommended
1986/07/08 02:20 a.m. PDT −116.608
1986/07/20 14:29:45.44 UTC 37.5667 5.04 ML5.9 Chalfant Valley
foreshock
1986/07/20 07:29 a.m. PDT −118.4382
1986/07/21 14:42:26.00 UTC 37.538 8.8 ML6.4 Chalfant Valley Advisory 2
1986/07/21 07:42 a.m. PDT −118.4428
1987/10/01 14:42:20.02 UTC 34.061 8.88 Mw5.9 Whittier Narrows CEPEC review, no advisory
recommended
1987/10/01 07:42 a.m. PDT −118.079
1987/10/04 10:59:38.19 UTC 34.074 7.72 ML5.3 Whittier Narrows
aftershock
1987/10/04 03:59 a.m. PDT −118.098
1987/11/24 01:54:14.66 UTC 33.09 10.85 Mw6.2 Elmore Ranch
1987/11/23 05:54 p.m. PST −115.792
1987/11/24 13:15:56.71 UTC 33.015 11.18 Mw6.6 Superstition Hills CEPEC review, no advisory
recommended
1987/11/24 05:15 a.m. PST −115.852
1988/06/10 23:06:43.05 UTC 34.943 5.9 ML5.4 Tejon Ranch CEPEC review, no advisory
recommended
1988/06/10 04:06 p.m. PDT −118.743
1988/06/27 18:43:22.33 UTC 37.1283 12.63 ML5.3 Lake Elsman Advisory 3
1988/06/27 11:43 a.m. PDT −121.895
1989/08/08 08:13:27.39 UTC 37.1482 13.41 ML5.4 Lake Elsman Advisory 4
1989/08/08 01:13 a.m. PDT −121.9268
1989/10/18 00:04:15.19 UTC 37.0362 17.21 ML7.0 Loma Prieta
1989/10/17 05:04 p.m. PDT −121.8798
1992/04/23 04:50:23.23 UTC 33.96 11.63 Mw6.1 Joshua Tree Advisory 5
1992/04/22 09:50 p.m. PDT −116.317
1992/04/25 18:06:05.18 UTC 40.3353 9.86 Mw6.7 Cape Mendocino
1992/04/25 11:06 a.m. PDT −124.2287
1992/04/26 07:41:40.09 UTC 40.4325 18.82 Mw6.5 Cape Mendocino
aftershock
1992/04/26 00:41 a.m. PDT −124.566
1992/04/26 11:18:25.98 UTC 40.3828 21.71 Mw6.6 Cape Mendocino
aftershock
CEPEC review; nonadvisory
statement
1992/04/26 04:18 a.m. PDT −124.555
1992/06/28 11:57:34.12 UTC 34.201 1.1 Mw7.3* Landers Advisory 6
1992/06/28 04:57 a.m. PDT −116.436
1992/06/28 15:05:30.73 UTC 34.203 3.63 Mw6.3 Big Bear
1992/06/28 08:05 a.m. PDT −116.827
1994/01/17 12:30:55.39 UTC 34.213 18.2 Mw6.7 Northridge
1994/01/17 04:30 a.m. PST −118.537
1998/08/12 14:10:25.14 UTC 36.7545 8.82 Mw5.1 Near San Juan
Bautista
First aftershock forecast; U.S.
Geology Survey-Caltech
statement
1998/08/12 07:10 a.m. PDT −121.4615
(Continued next page.)
786 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

Table 1 (continued)
Earthquakes and Earthquake Sequences That Have Prompted Review by the California Earthquake Prediction Evaluation
Council (CEPEC), or That Are Described in the Text for Other Reasons
Date and Time
(yyyy/mm/dd hh:mm:ss.ss)
Latitude
Longitude (°)
Depth
(km) Magnitude Location Action
2001/11/13 20:43:14.95 UTC 33.3172 5.55 ML4.1 Bombay Beach
swarm
Advisory 7
2001/11/13 12:43 p.m. PST −115.7002
2003/12/22 19:15:56.24 UTC 35.7005 8.38 Mw6.5 San Simeon CEPEC review of M8
prediction
2003/12/22 11:15 a.m. PST −121.1005
2004/09/28 17:15:24.25 UTC 35.8182 8.14 Mw6.0 Parkfield
2004/09/28 10:15 a.m. PDT −120.366
2007/10/31 03:04:54.81 UTC 37.4335 9.74 Mw5.5 Alum Rock Advisory 8
2007/10/30 08:04 p.m. PDT −121.7743
2009/03/24 11:55:43.93 UTC 33.3172 5.96 ML4.8 Bombay Beach
swarm
CEPEC recommends advisory,
but not issued
2009/03/24 04:55 a.m. PDT −115.7282
2009/03/25 07:51:23.01 UTC 33.2900 4.40 ML3.6 Bombay Beach
swarm
2009/03/25 00:51 a.m. PDT −115.7213
2009/03/25 19:59:44.10 UTC 33.2920 6.50 ML3.7 Bombay Beach
swarm
2009/03/25 12:59 a.m. PDT −115.7197
2009/03/26 03:25:21.98 UTC 33.2930 7.38 ML4.0 Bombay Beach
swarm
2009/03/25 08:25 p.m. PDT −115.7215
2009/04/08 22:04: 2.36 UTC 33.3180 4.11 ML3.5 Bombay Beach
swarm
2009/04/08 03:04 p.m. PDT −115.7227
2010/06/15 04:26:58.24 UTC 32.705 8.99 Mw5.7 El Mayor Cucapah
aftershock
Advisory 9
2010/06/14 09:26 p.m. PDT −115.9113
2012/08/26 19:31:23.04 UTC 33.0172 8.26 Mw5.3 Brawley CEPEC review
2012/08/26 12:31 a.m. PDT −115.5537
2012/08/26 20:57:58.22 UTC 33.0185 8.22 Mw5.4 Brawley No advisory recommended
2012/08/26 01:57 p.m. PDT −115.5403
2014/08/24 10:20:44.06 UTC 38.2155 11.21 Mw6.0 South Napa CEPEC review; statement
2014/08/24 03:20 a.m. PDT −122.3117
2016/09/26 14:31:08.8 UTC 33.2976 2.2 Mw4.3 Bombay Beach
swarm†
Advisory 10
2016/09/26 7:31 a.m. PDT −115.7137
2016/09/27 03:23:58.58 UTC 33.2998 4.84 Mw4.3
2016/09/26 8:23 p.m. PDT −115.7123
2016/09/27 03:36:15.15 UTC 33.3058 2.52 Mw4.1
2016/09/26 8:36 p.m. PDT −115.7010
2016/09/27 03:46:30.30 UTC 33.3080 3.1 Mw3.5
2016/09/26 8:46 p.m. PDT −115.6973
2016/09/28 01:05:51.51 UTC 33.2987 7.03 Mw3.6
2016/09/27 6:05 p.m. PDT −115.7118
2016/09/28 01:13:04.4 UTC 33.2988 6.85 Mw3.5
2016/09/27 6:13 p.m. PDT −115.7053
Information is from the Advanced National Seismic System (ANSS) catalog (NCEDC, 2014) except as noted. Magnitudes are not
necessarily the same as the magnitudes upon which advisory-related decisions were based. PDT, Pacific Daylight Time; PST,
Pacific Standard Time.
*see Data and Resources.
†ANSS Comprehensive Earthquake Catalog (ComCat); search performed 20 December 2016.
Seismological Research Letters Volume 88, Number 3 May/June 2017 787
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

scientific assessments with emergency management actions.
During this period, CEPEC was chaired by James F. Davis, who
served as California State Geologist from 1978 to 2003.
Advisory 1: 17 June 1985 San Diego Bay Sequence
San Diego, California, residents felt three earthquakes on the
evening of 17 June 1985 (ML3.7 at 5:12 p.m. Pacific Daylight
Time [PDT], ML3.9 at 8:22 p.m. PDT, and ML3.7 at 9:28
p.m. PDT). This swarm and an ML3.3 earthquake on 21 June
1985 are ascribed to the Rose Canyon fault zone, which passes
beneath the city of San Diego (Grant and Shearer, 2004).
An M7 earthquake on the Rose Canyon fault was one of
seven scenario earthquakes for which probable effects on pop-
ulation centers had been evaluated by the USGS in 1981
(USGS, 1981). That report assigned an earthquake of this mag-
nitude a very low probability of occurring before 2001 (0. 01%
chance per year). Nevertheless, USGS scientists were concerned
that a larger earthquake (M≥5), endangering San Diego, might
follow the 17 June 1985 events(Goltz, 1985). The evening of 17
June 1985, the USGS communicated an advisory to Cal OES,
labeled “For Internal Use Only: Not for Release to Press or Pub-
lic.”The account below is condensed from Goltz (1985).
As a state agency, Cal OES had no plan for responding to
this unprecedented message from the federal USGS, and San
Diego County officials only learned of the advisory informally
upon contacting Cal OES for information about the earth-
quakes. After several conference calls, San Diego County Office
of Disaster Preparedness (ODP) was given responsibility for
deciding how to proceed.
San Diego County officials weighed several factors. The
USGS request that the information not be released and the
reluctance of Cal OES to issue a warning exposed the County
to possible legal liability if it issued a warning that led to ad-
verse outcomes. Public information needed to be crafted care-
fully to avoid overreaction. After consultation with Cal OES,
San Diego County ODP issued an earthquake potential advi-
sory at 5 p.m. Pacific Standard Time (PST) on 18 June 1985,
24 hr after the swarm began: “According to the United States
Geological Survey, there is a slight increase in the probability of
a potentially damaging earthquake in San Diego. This state-
ment is based on historical data on the type of earthquake
swarms we had yesterday. One in twenty of these swarms
has been followed by a damaging earthquake of Richter mag-
nitude 5.0 or greater within a five day period.”
Numerous phone calls to San Diego County ODP soon
revealed that the public equated the earthquake potential ad-
visory to an earthquake prediction. Rumors arose that the
probability of a larger earthquake was 20%, that the expected
earthquake would be M8.3 or larger, and that the beaches and
city were being evacuated. San Diego County ODP responded
with press releases requesting the populace to check back with
them to verify information. Further confusion developed on
20 June 1985, when newspapers, citing a Caltech seismologist,
reported that the advisory had been downgraded, although San
Diego County ODP had reaffirmed that the advisory remained
in effect.
After conferring with the USGS, San Diego County is-
sued a public statement at 5 p.m. PDT on 21 June 1985 stating
that, barring increased seismic activity, the advisory would be
cancelled at midnight PDT on 22 June 1985. There were
no damaging follow-on earthquakes in the immediate San
Diego area.
The San Diego advisory abruptly exposed the challenges of
issuing a public earthquake warning: appropriately phrasing
the warning, providing sufficient public information, dealing
with rumors, and updating and cancelling the warning. Lack-
ing previously developed procedures, scientists and public of-
ficials had to devise a plan for evaluating and communicating
seismic observations and crafting an appropriate alert during
the hours when a larger earthquake was most likely to follow
the 17 June seismicity.
CEPEC Deliberations on How to Respond to Potential
Foreshock Activity
Possibly prompted by the 1985 San Diego advisory, CEPEC
formed a committee to advise OES on feasibility of preplanned
seismic response policies, chaired by Karen McNally of the
University of California at Santa Cruz. Reporting back in No-
vember 1986, the committee recommended collecting research
papers on faults with seismic gaps, and maintaining a database
of knowledgeable researchers who could be reached on short
notice by CEPEC.
Accompanying the minutes to a December 1987, CEPEC
meeting was a handwritten document by Karen McNally, re-
viewing foreshock activity prior to 11 earthquakes of M≥7in
California, beginning with the 1857 Mw7.9 Fort Tejon earth-
quake, and concluding: “It would be prudent for OES to be on
alert for ∼72 hours following moderate earthquakes M≥5:5
within 30-50 km of the ends of known seismic gaps. Local
emergency officials might be notified. (Public announcement
not necessarily needed.) M≥5.0-5.4 might also warrant alert in
these locations - if there aren’t too many false alarms for OES
practicality. The case histories are too few (for M≥7earth-
quakes) to use a purely statistical approach. The above advice
is based on a cautionary premise.”
The 1987 McNally document recommends an alert period
and magnitude threshold similar to those of Jones (1985).
McNally was presumably aware of the Jones (1985) paper, for
which the dataset included fewer than 10 M≥7earthquakes
after 1932.
Four Nonadvisory Earthquakes 1986–1988
Between 1986 and 1988, CEPEC evaluated four southern
California earthquake sequences and decided against recom-
mending advisories.
The first two—the 8 July 1986 ML5.6 North Palm
Springs earthquake (Jones et al., 1986) and the 1 October
1987 Mw5.9 Whittier Narrows earthquake (Hauksson and
Jones, 1989)—were themselves damaging, but in locations not
believed capable of generating much larger earthquakes. No
larger events followed. However, an Mw5.3 aftershock three
788 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

2003/12/22
M6.5
−124° ° −122 −120° −118° −116° −114°
32°
34°
36°
38°
40°
42°
0100
km
2007/10/31
M5.5
1985/06/18
M3.7,M3.9,M3.7
1986/7/13 M5.8
1985/6/21 M3.3
1998/8/12 M5.1
1988/6/27 M5.3
1989/8/8 M5.4
1989/10/18 M7.0
2014/8/24
M6.0
1990/4/18 M5.4
1988/6/13 M5
2004/9/28
M6
1987/2/14
M5.2
1988/6/10
M5.4
1992/7/11
M5.3
1995/8/17 M5.4
1995/9/20 M5.6
2012/2/13
M5.6
2001/8/10
M5.3
2013/5/24
M5.7
M 7
M 6
M 5
M 4
CEPEC
convened Advisory
not
considered
Advisory
issued
(d) (e)
(f)
2008/4/30 M 5.4
(c)
(b)
(a)
2016/8/10
M5.1
2016/12/14
M5.0 2016/12/28
M5.7, M5.7, M5.5
−119° −118°
37°
38°
050
km
1986/7/20 M5.9
1986/7/21 M6.4
1993/5/17 M6.1
2004/9/18
M5.6,M5.4
1999/5/15
M5.5
(c)
−119° −118°
33°30'
34°00'
34°30'
050
km
1994/1/17
M6.7
1987/10/4 M5.3
1987/10/1 M5.9
1991/6/28
M5.8
1990/2/28
M5.5
2008/7/29
M5.4
(d)
050
−125° −124°
40°
41°
1992/4/25
M6.7
1994/12/26 M5.4
2007/2/26
M5.4 2010/1/10 M6.5
2010/2/4
M5.9
1987/7/31
M5.6
1990/1/16 M5.4 1992/3/8
M5.3
1997/1/22
M5.6
(b) km
1992/4/26
M6.5, M6.6
1991/8/17
M6.0
2015/1/28
M5.7
−117° −116° −115°
32°
33°
0 50
km
United States
Mexico
2010/4/4
M7.2
2012/8/26
M5.3,M5.4
2010/7/7
M5.4
2010/6/15
M5.7
1987/11/24
M6.2,M6.6
(f) 2016/6/10
M5.2
−117° −116°
34°
35°
050
km
1986/07/08
M5.6 1992/04/23
M6.1
1992/06/28
M7.3
1992/06/28
M6.3
1999/10/16
M7.1
(e) (g)
−115°50' −115°45' −115°40'
33°15'
33°20'
05
km
(g) 2001 2016
2009
M 5
M 4.5
M 4
M 3.5
Salton Sea
San
Andreas
fault zone
Brawley Seismic zone
▴Figure 1. Map showing epicenters of earthquakes M5 or greater in and near California from 1 January 1985 through 31 December 2016.
Information is from the Advanced National Seismic System (ANSS) catalog (Northern California Earthquake Data Center [NCEDC], 2014)
except as noted in the footnotes to Table 1and ⒺTable S1 (available in the electronic supplement to this article): (a) California and
surrounding regions; (b) Cape Mendocino; (c) Chalfant Valley; (d) Greater Los Angeles; (e) eastern California shear zone; (f) southernmost
California; and (g) Salton Sea, events with magnitude greater than 3.5.
Seismological Research Letters Volume 88, Number 3 May/June 2017 789
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

days after the Whittier Narrows earthquake caused additional
damage, injuries, and one fatality (Stover and Coffman, 1993).
In the third of these sequences, the Mw6.2 Elmore
Ranch earthquake at 5:54 p.m. PST on 23 November 1987
was followed 11.4 hr later by the Mw6.6 Superstition Hills
earthquake, 9 km to the southwest (Wa ld et al., 1990). Had
CEPEC been convened before the second of these two events,
their ability to identify a potential foreshock would have been
tested. As it happened, CEPEC met after the mainshock, and
their judgment that no larger earthquake would follow proved
correct. Their decision may have been based on an absence of
known faults in that area extensive enough to host a larger
earthquake.
The fourth of these earthquakes was the ML5.4 Tejon
Ranch earthquake on 10 June 1988, only a few kilometers
from the Mojave segment of the San Andreas fault, which
ruptured in the great 1857 Mw7.9 earthquake. This fault
segment was considered unlikely to re-rupture after only
131 years of quiescence—an assumption since revisited, for
example, by Scharer et al. (2010). Though one may wonder
whether CEPEC would have decided differently in light of
later research, their 1988 judgment was correct: no larger
earthquake occurred.
Advisory 2: 21 July 1986 Chalfant Valley Earthquake
Sequence
This vigorous earthquake sequence began with an ML5.9 fore-
shock on 20 July 1986, 24.2 hr prior to the ML6.4 mainshock
(Savage and Cockerham, 1987). CEPEC convened on 21 July
1986, after this mainshock, and recommended that an earth-
quake advisory be issued. This advisory, dated 22 July 1986, was
a letter to William M. Medigovich, Director of Cal OES, from
Dallas L. Peck, Director of the USGS, stating: “It is our assess-
ment that the region of Chalfant Valley, from Bishop north to
the Nevada border, may experience additional earthquakes sim-
ilar to the July 21 event during the next few days.”The letter
described likely damaging effects of such an earthquake, and
stated that an update would follow on 24 July 1986. No addi-
tional earthquakes M≥6occurred.
Some alarmed speculation and rumors arose in the town of
Bishop, California, prompted by the letter’s statement that the
Chalfant Valley had been recognized to have the potential for
generating a magnitude-7 event (Ramos and Reich, 1986). The
public may also have been reacting to Cal OES placing two
Army National Guard helicopters, a medical evacuation heli-
copter, a communications van, and a fuel tanker on standby in
Bishop.
Bishop residents in 1986 reacted much like those in San
Diego in 1985 in seeing the advisory as a prediction, with ensu-
ing rumors of a much higher probability of a larger (M∼7)
earthquake than scientists had actually warned of. And again,
CEPEC’s recommendation had come after the mainshock,
missing a 24.2-hr window between the ML5.9 foreshock
and the ML6.4 mainshock, during which an advisory, if issued,
would have been fulfilled.
Advisories 3 and 4: The 1988 and 1989 Lake Elsman
(Lexington Reservoir) Earthquakes
Scientific consensus is that the 1988 and 1989 Lake Elsman
earthquakes in Santa Clara County, California, were fore-
shocks of the damaging 18 October 1989 ML7.0 Loma Prieta
earthquake (Sykes and Jaumé, 1990;Perfettini et al., 1999).
Harris (1998a,b) provides thorough accounts of the advisories
prompted by these two events.
The first Lake Elsman earthquake (ML5.3) occurred
during the morning of 27 June 1988 (Olson, 1990). That after-
noon, USGS seismologist Alan Lindh informed California
State Geologist Jim Davis that the earthquake was the largest
since 1906 to have occurred at the northern end of the south-
ern Santa Cruz Mountains segment of the San Andreas fault.
Noting that the Working Group on California Earthquake
Probabilities (WGCEP, 1988) had assigned this segment a
30% probability of an M6.5–7 earthquake in the 30-year
period from 1988 to 2018, USGS seismologist Lindh told
California State Geologist Davis that this ML5.3 event could
be a foreshock to an M6–6.5 earthquake.
CEPEC convened, concluded that the probability of a
larger earthquake on this segment of the San Andreas fault was
temporarily elevated, and recommended that Cal OES send an
earthquake advisory to local governments, but not announce
this to the public via a news release (for details, see appendix
2ofHarris, 1998a). The following day, however, California
state officials did issue a news release.
The earthquake advisory was for five days, consistent with
the time window used by Jones (1985) for southern California
foreshocks, although no northern California earthquakes had
been part of the Jones (1985) foreshock study. The aftershock
sequence of the ML5.3 Lake Elsman earthquake was not very
vigorous, and the advisory expired on 2 July 1988.
The second Lake Elsman earthquake (ML5.4) occurred
on 8 August 1989 (Olson, 1990), within 10 km of the 27 June
1988 ML5.3 earthquake, prompting Cal OES to issue another
CEPEC-recommended five-day advisory. No unusual activity
followed in those five days, so this advisory also expired.
The next notable seismic activity in the area was the Loma
Prieta mainshock itself (ML7.0) on 18 October 1989 (Olson,
1990). It ruptured a 35-km-long patch on a steeply dipping
fault subparallel to and just southwest of the San Andreas fault,
extending northward to within 15 km of the Lake Elsman epi-
centers. The two Lake Elsman earthquakes, 16 months and 2.5
months prior to the mainshock, were retrospectively recog-
nized as foreshocks.
FURTHER STATISTICAL STUDIES OF
EARTHQUAKE SEQUENCES
Reasenberg and Jones (1989) presented a method to estimate
probabilities of not only a larger follow-on earthquake but also
of aftershocks. They introduced a conceptual framework in
which foreshocks, mainshocks, and aftershocks are not intrinsi-
cally different, so that it is only a matter of chance whether or
790 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

not the mainshock (the largest event) is the first earthquake in
the sequence.
Next, Agnew and Jones (1991) analyzed foreshock prob-
abilities for the entire San Andreas system, assigning higher
probabilities of being a foreshock to events on or near faults
where a mainshock is regarded as about due. (Time-dependent
probabilities of specific California mainshocks had been devel-
oped by WGCEP, 1988.) Agnew and Jones (1991, p. 11,960)
hypothesized that “the large characteristic earthquakes on a
fault zone are not simply the largest members of the total pop-
ulation of earthquakes there, but are somehow derived from a
different population. Foreshocks to such events can thus rea-
sonably be regarded as also being a separate class of events from
the background earthquakes.”However, they acknowledged
that foreshocks lack features distinguishing them from back-
ground events.
Agnew and Jones (1991) pointed out that fault segments
known to have produced great earthquakes historically now
have little background seismicity, and they argued that the
probability that an event is a foreshock is higher when back-
ground activity is lower. Michael (2012) pointed out that this
proposal implied earthquakes should be more capable of trig-
gering other earthquakes in locations with lower background
activity, but in southern California he found no evidence for
that behavior.
DEVELOPMENT OF THE CALIFORNIA
EARTHQUAKE ADVISORY PLAN
By 1990, California state emergency management officials and
CEPEC scientists had learned to collaborate effectively, despite
their distinct professional cultures. Scientists prefer to speak
accurately and to withhold judgment when large uncertainties
exist, whereas emergency managers, more concerned with not
being blindsided, prefer uncertain information to no informa-
tion at all. Earthquake scientists, who lack expertise in devising
appropriate ways to manage preparedness and public informa-
tion, may view a chaotic large-scale evacuation as the only
possible response to an earthquake advisory (e.g., Wang and
Rogers, 2014). In contrast, emergency managers are schooled
in a broad spectrum of mitigation measures. For example, in
the case of low-probability events, they may recommend only
that their own agencies prepare internally, facilitating a possible
event response without elevating public concern (for a list
of potential actions, see Reitherman, 1986). The California
Earthquake Advisory Plan was therefore based on a principle
that has now been articulated in the post-L’Aquila earthquake
report by the International Commission on Earthquake Fore-
casting for Civil Protection (2011): that of “transmitting sci-
entific information about future earthquake occurrence to
decision-makers in a way that appropriately separates hazard
estimation by scientists from the public protection role of civil
authorities.”
In October 1990, the formal Advisory Plan was published
(State of California, Governor’s Office of Emergency Services,
1990b), both as an appendix to the California Short-Term
Earthquake Prediction Response Plan (State of California,
Governor’s Office of Emergency Services, 1990a) and as a
stand-alone document, developed by emergency management
consultants Terrence Haney and James Powers. The Short-
Term Earthquake Prediction Response Plan recommends
graduated response procedures tied to potential impacts and
probabilities of occurrence should a scientific earthquake
prediction be validated by CEPEC. The Advisory Plan ad-
dresses anomalous seismic activity (e.g., an M≥5earthquake
on a fault known to have generated large damaging earthquakes
in the past) that could be precursory within a short time span
to a stronger potentially damaging event.
Earthquake Advisories are “statements by Cal OES regard-
ing scientific assessments that, within a specified period [usually
3–5 days], there is an enhanced likelihood for damaging earth-
quakes to occur in areas designated in the Advisory”(State of
California, Governor’s Office of Emergency Services, 1990b,
p. 3). The California Earthquake Advisory Plan, which recom-
mends actions appropriate if the estimated probability within
72 hr of a damaging earthquake of M>5does not exceed 25%,
has been exercised frequently. Were circumstances to indicate
that the probability of such an event were greater than 25%,
it is likely that the more extensive actions prescribed in the
Short-Term Prediction Response Plan would instead be imple-
mented, but that has never happened.
In addition to formalizing the previously developed
protocol for convening CEPEC to assess potential foreshocks,
the Advisory Plan specifies procedures to be followed by Cal
OES: sending checklists of actions to local jurisdictions in the
affected region, notifying state and federal agencies with emer-
gency response functions, and formulating a release to the
media and public. The Advisory Plan includes standard mes-
sage content for communicating the Advisory, and for contin-
uing or cancelling the Advisory (with CEPEC’s concurrence).
Also in 1990, the USGS was refining a separate protocol
for assessing possible increases in mainshock probabilities on
the southern San Andreas fault (Jones et al., 1991;Jordan and
Jones, 2010). This USGS “short-term earthquake hazard assess-
ment”defines alert levels D, C, and B, corresponding to pro-
gressively larger magnitude potential foreshocks in this region,
which are assigned higher probabilities of being followed by a
mainshock than events of the same magnitude elsewhere in
California. In keeping with the principle of separating the roles
of scientists from those of emergency management officials,
CEPEC reviewed a draft of this USGS geographically targeted
plan in 1990, and requested revisions to “avoid any confusion
regarding the distinction between actions taken at certain alert
levels by the USGS of a scientific nature and those of the State
of a public safety nature.”
OPERATING UNDER THE ADVISORY PLAN 1990–
2016
After October 1990, a procedure was in place whereby CEPEC
would rapidly evaluate anomalous seismic activity, and
Cal OES would carry out predefined actions upon accepting
Seismological Research Letters Volume 88, Number 3 May/June 2017 791
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

a CEPEC-recommended Advisory. Yet the Advisory Plan did
not eliminate the need for CEPEC to evaluate, and Cal OES to
respond to, situations that had not specifically been covered by
published statistical analyses.
25 April 1992 Mw6.7 Cape Mendocino Earthquake
The tectonically complex, seismically active Mendocino Triple
Junction (MTJ) is the northern terminus of the San Andreas
fault, where it meets the Cascadia subduction zone to the north
and the Mendocino oceanic transform fault to the west (Ba-
kun, 2000). The statistical analyses of Jones (1985) and Agnew
and Jones (1991) did not include earthquakes from the MTJ
that were not on the San Andreas fault system.
On 25 April 1992, the Mw6.7 Cape Mendocino earth-
quake occurred at 11:06 a.m. PDT, reaching Mercalli intensity
VIII and causing damage valued at $48–$66 million (Oppen-
heimer et al., 1993). CEPEC did not convene, however, until
two aftershocks of Mw6.5 and 6.6, 13 and 17 hr later, respec-
tively, had caused fires and additional damage (Oppenheimer
et al., 1993). CEPEC crafted a statement (not an Advisory)
that Cal OES issued on 26 April 1992, warning that more
strong aftershocks could hit the Humboldt County area.
No further damaging events occurred after this warning.
Detailed analysis (Oppenheimer et al.,1993)revealed
that the 1992 Cape Mendocino earthquake was a thrust event
on the Cascadia subduction interface. CEPEC’s decision not
to issue an Advisory for this earthquake predates consensus
that Cascadia has potential to host an earthquake M8or
larger (Atwater et al., 1995).
Advisory 5: The 23 April 1992 Mw6.1 Joshua Tree
Earthquake and Advisory 6: The 28 June 1992 Mw7.3
Landers and Mw6.3 Big Bear Earthquakes
According to Jones (1994, p. 892), the Joshua Tree earthquake
“had a probability of 21% of being a foreshock to a San An-
dreas mainshock within the next 3 days. On this basis, the state
of California issued an advisory warning of the potential of
a damaging earthquake on the San Andreas within 3 days.”
(p. 892). No such earthquake occurred, however, and the Ad-
visory expired.
But two months later, at 04:57 a.m. PDT on 28 June
1992, the Mw7.3 Landers earthquake ruptured a series of
faults of the Eastern California shear zone, north-northeast of
the JoshuaTree epicenter (Hauksson et al., 1993). The Landers
earthquake was preceded by a swarm of foreshocks over 7 hr,
but the largest was of ML3.4 (Dodge et al., 1995), too small to
convene CEPEC. The Landers mainshock was followed after
3hrbytheMw6.3 Big Bear earthquake at 8:05 a.m. PDT,
which, though unquestionably an aftershock, ruptured a sep-
arate fault 30 km west of the Landers rupture zone (Hauksson
et al., 1993).
CEPEC convened shortly after the Big Bear earthquake,
and Cal OES issued an Advisory mid-morning of June 28.
This Advisory made the unprecedented recommendation of
“no unnecessary travel until further notice,”due to road dam-
age and landslides in the epicentral areas. The travel ban was
reportedly unpopular and Cal OES rescinded it at 5 p.m. PDT
the same day.
Concern rose considerably on 29 June, when an Mw5.5
Landers aftershock occurred on the San Andreas fault zone
near Indio at 9:01 a.m. PDT. That evening, CEPEC reviewed
strain changes recorded at Piñon Flat Observatory (Wyatt et al.,
1994), considered stress changes imposed by the Landers and
Big Bear events (Harris and Simpson, 1992;Stein et al., 1992),
and concluded that 193 km of the San Andreas fault between
Cajon Pass and Bombay Beach could have been significantly
destabilized. From Jones (1994, p. 893): “Concern about the
possibility of a major San Andreas earthquake was so high that
…(CEPEC) recommended that the state of California prepare
for an imminent earthquake alert if an M≥6earthquake were
to occur ‘on or near’the southern San Andreas fault.”
On 1 July 1992, CEPEC and Cal OES adopted a tempo-
rary protocol informally called the go-to-war scenario (Jordan
and Jones, 2010), whereby certain events such as an M6.0
event within 3 km of the southern San Andreas fault would
prompt Cal OES to implement plans for a 25% chance of a
major earthquake. Although no Advisory was officially extended,
Caltech and USGS scientists agreed to remain in a 24/7 height-
ened mode of monitoring.
At least one spate of rumors arose during the aftershock
sequence: when the USGS announced that it was upgrading
the Landers earthquake from M7.3 to 7.5, some people be-
lieved scientists and officials were concealing the full extent
of the hazard (Reich, 1992).
The go-to-war scenario officially ended in February 1997,
when CEPEC members’consensus was that the probability of a
large earthquakeon the southern San Andreas faulthad returned
to pre-1992-Landers–Big Bear sequence levels (Minutes, 28 Feb-
ruary 1997 CEPEC meeting). At the time of this writing, the
feared southern San Andreas event has not occurred.
The decision to enact the go-to-war scenario was not based
on rigorous statistics: the Landers aftershocks had increased the
background seismicity rate, violating an assumption of the Ag-
new and Jones (1991) analysis. Jones (1994) extended that
methodology, arguing that the elevated background activity
lowered the probability that an M6 aftershock would be a
foreshock. However, she also noted that the Landers sequence
imposed static stress changes that presumably brought the
southern San Andreas fault closer to failure, a principle that
is currently being evaluated for use in forecasting the evolution
of earthquake sequences (e.g., Segou et al., 2013;Nic Bhlo-
scaidh et al., 2014).
It became apparent retrospectively that the Joshua Tree
earthquake had been a Landers foreshock: although the two
earthquakes’epicenters were 35 km apart, seismic swarms had
occurred between the Joshua Tree epicenter and the southern
end of the future Landers rupture in the nine weeks between
those events (Hauksson et al., 1993). Be cause the Joshua Tree–
Landers–Big Bear sequence occurred off the San Andreas sys-
tem, where no authoritative probability estimates existed, its
implications for the probability of a major earthquake could be
assessed only qualitatively. Nevertheless, CEPEC exercised its
792 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

judgment by recommending a heightened state of alert, which
Cal OES accepted and followed.
The 1994 Northridge Earthquake: No Foreshocks
The devastating 17 January 1994 Mw6.7 Northridge earth-
quake, on a subsurface thrust fault, had no foreshocks within
a 30-km radius that justified convening CEPEC (Hauksson
et al., 1995). Jones (1984) had previously found that, of the
four M≥5California mainshocks before 1980 with reverse-
faulting mechanisms, none had foreshocks. The Northridge
earthquake underscores that even perfect identification of fore-
shocks would provide advisories for fewer than half of damag-
ing earthquakes.
Implementation of Aftershock Forecasts
In 1996, the Northern California Seismic Network (NCSN,
operated by the USGS and the University of California
at Berkeley) began planning to augment its reports for earth-
quakes M≥4with seven-day aftershock probabilities, calcu-
lated automatically following the Reasenberg and Jones (1989,
1994) methodology. NCSN’s draft wording alluded to the pos-
sibility of a larger follow-on event with the statement “Usually,
aftershocks are smaller than the mainshock….”After CEPEC
discussed the algorithms, protocol, and text between December
1996 and October 1997, the threshold for automated after-
shock reports was raised to M≥5,andNCSN’s template,
while retaining the probability of a larger mainshock, was
revised to emphasize that a larger mainshock would probably
not occur.
The CEPEC-approved protocol was exercised on 12 Au-
gust 1998, when an automated aftershock warning was linked
on the NCSN’s public website to the Preliminary Earthquake
Report for an Mw5.1 earthquake near San Juan Bautista, Cal-
ifornia. CEPEC recommended that USGS and Cal OES issue
a“Joint Summary Statement,”specifically stating that it was
not an earthquake Advisory. The Joint Summary Statement
gave a 5%–10% seven-day probability of another earthquake
of M≥5:1. That estimate, based on the Reasenberg and Jones
(1994) aftershock statistics, was at least as high as the fore-
shock probability implied by the Agnew and Jones (1991)
methodology.
Advisory 7: 13 November 2001 Bombay Beach Swarm
On 13 November 2001, a microearthquake swarm near Bombay
Beach, California, began with an ML2.4 event at 5:54 a.m. PST,
continuing with several events ML3 and larger. Bombay Beach
is at the southern terminus of the Coachella segment of the San
Andreas fault, for which WGCEP (1988) had estimated a 40%
chance of an M≥7:5earthquake before 2018. Consequently,
events M≥3:1here are assigned up to a 1% 72-hr probability
of being followed by an M7.5 earthquake, rising to as much as
5% for an M4.2 event (Jones et al., 1991). The swarm prompted
CEPEC conference calls at 9:30 and 11 a.m. PSTthat morning.
At 2 p.m. PST, following an ML4.1 event at 12:43 p.m. PST,
Cal OES issued an Advisory for a possible major southern San
Andreas earthquake. The ML4.1 earthquake proved to be the
largest of the swarm, however.
According to Jordan and Jones (2010, p. 572), “The pub-
lic scarcely noticed this Level C alert, and there was little media
interest in the situation.”Nevertheless, by taking the prepar-
atory actions specified in the Advisory plan, Cal OES had read-
ied itself to respond quickly had a major earthquake occurred.
22 December 2003 Mw6.5 San Simeon Earthquake
The San Simeon earthquake occurred on a previously un-
known thrust fault subparallel to the San Andreas fault in cen-
tral California (Hardebeck et al., 2004). Despite the event’s
magnitude, CEPEC was not convened, possibly because no
larger earthquake was known to have occurred on this fault.
On 6 January 2004, newspapers reported that University
of California, Los Angeles, scientist Vladimir Keilis-Borok
claimed to have predicted the San Simeon earthquake and was
citing that success to support his new prediction that an earth-
quake of M≥6:4would occur by 5 September 2004 in a region
including the southeast Mojave Desert as well as an area south
of that (Wolpert, 2004). CEPEC, chaired by Acting California
State Geologist Michael S. Reichle, met 20 February 2004, to
discuss this new prediction, and declined to either validate or
invalidate it (California Department of Conservation, 2004).
The predicted southern California earthquake did not occur.
However, a summary of CEPEC’s deliberations states that the
2003 San Simeon earthquake had satisfied a June 2003 predic-
tion by the Keilis-Borok team (California Earthquake Predic-
tion Evaluation Council, 2004).
28 September 2004 Mw6.0 Parkfield Earthquake
The Parkfield segment of the San Andreas fault had been
densely instrumented since 1985 in anticipation that an M∼
6event would occur before 1993, and a separate protocol for
the USGS to communicate directly with Cal OES in the event
of possible foreshocks and/or anomalous crustal deformation
was in place for Parkfield (Bakun et al., 2005). The Mw6.0
Parkfield earthquake did not occur until 28 September 2004,
however, and was not preceded by any foreshocks or other pre-
cursory phenomena. The potential for a Parkfield rupture to
extend further southwest along the San Andreas fault had been
noted (Harris and Archuleta, 1988), and a Parkfield earth-
quake was a foreshock to the 1857 Mw7.9 Fort Tejon earth-
quake (Meltzner and Wald, 1999). Nevertheless, with a
detailed response plan in place, a sparse and earthquake-aware
local population, and intense scientific attention to a long-
awaited Parkfield M∼6event, convening CEPEC was judged
unnecessary (J. Parrish, written comm., 2015).
Advisory 8: 31 October 2007 Mw5.5 Alum Rock
Earthquake
The 2007 Alum Rock earthquake occurred 15 km northeast of
San Jose on the Calaveras fault (Oppenheimer et al., 2010). In
response to this event, CEPEC was convened by current Cal-
ifornia State Geologist John G. Parrish, who has served since
2005. As recommended by CEPEC, Cal OES issued an Advi-
Seismological Research Letters Volume 88, Number 3 May/June 2017 793
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

sory of a heightened chance of a damaging earthquake along
the Hayward and/or Calaveras faults. No such earthquake oc-
curred before the Advisory expired.
Online commentary during this Advisory, the first since
broadband Internet service had become widely available, offers
some insight into public reaction. Web searches did not uncover
any undue overreaction to the Advisory, but a reposting of the
Cal OES Advisory statement on FreeRepublic.com (2007) elic-
ited the following comment: “I’m probably missing something
but how can the probability be significantly increased for (sic)
above the normal level of a damaging quake while at the same
time have an overall low likelihood?”This question, which
emergency managers and scientists often struggle to answer,
received three helpful replies:
•“Because the basic probability of a large quake at a moment
in time is very, very low. Several times greater riskcan still be
a low probability. It is (sic) not as goofy as it sounds.: -)”
•“The normal probability level is even lower.”
•“If you buy one lottery ticket, your odds of winning are low.
If you buy a hundred lottery tickets, your odds of winning
have gone up a hundredfold—a significant increase—but
they’re still low.”
Some additional comments at this site were frivolous, but
several urged preparedness, and none indicated a lack of respect
for earthquake science. In fact, one states, referring to CEPEC:
This is an official advisory body of legitimate scientists. There
is a heightened risk of a far more serious quake on either Cal-
averas or Hayward for the next few days, if this is a foreshock.
A Restructured Emergency Management Agency Does
Not Issue a CEPEC-Recommended Advisory: March 2009
Bombay Beach Swarm
In March 2009, another microearthquake swarm began near
Bombay Beach, California, including three M>3events within
30 min of each other on 21 March. A subsequent ML4.8 event
on 24 March 2009 was the largest earthquake ever recorded
within 10 km of the southern half of the Coachella segment
of the San Andreas fault (Jordan and Jones, 2010). By this time,
the Uniform California Earthquake Rupture Forecast, v.2 (Field
et al., 2008), had estimated a 24% probability of an M≥7earth-
quake on the Coachella segment before 2038.
CEPEC met by teleconference three and a half hours after
the ML4.8 earthquake on 24 March 2009, and recommended
that an Advisory be issued, stating: “CEPEC believes that
stresses associated with this earthquake swarm may increase the
probability of a major earthquake on the San Andreas Fault to
values between 1 to 5 percent over the next several days.”
However, as of 1 January 2009, Cal OES had been reor-
ganized, merged into the California Governor’s Office of
Homeland Security, and renamed the California Emergency
Management Agency (Cal EMA). New managers at Cal EMA,
just 2.5 months into their assignments, had not been briefed on
the Earthquake Advisory Plan. Although prepared to respond
to a damaging earthquake, they had not anticipated receiving
information that the probability of a future damaging earth-
quake was elevated, and, as a result, did not issue an Advisory.
Instead, on 24 March 2009 they sent a Preparedness Advisory,
including CEPEC’s statement, to potentially affected counties
in southern California, recommending that they “reach out to
agencies and jurisdictions …to raise their awareness of today’s
events”and “ensure the readiness of systems essential to emer-
gency operations and remind the public about the importance
of being prepared for earthquakes.”This Preparedness Advi-
sory did not include the specific actions or wording specified
in the Advisory Plan.
In 2013, Cal EMA was renamed Cal OES and became an
agency reporting directly to the Governor. The protocol for
handling Advisories has been clarified, but the situation that
developed after the governmental restructuring illustrated the
challenge of keeping a protocol alive as organizations evolve.
Advisory 10: 26 September 2016 Bombay Beach Swarm
On Monday, 26 September 2016 at 4:03 a.m. PDT, a third
microearthquake swarm began beneath the Salton Sea near
Bombay Beach, California, just south of the 2001 swarm and
north of the southern edge of the 2009 swarm (SCSN, 2016;
Fig. 1g). Two M4.3 events and an M4.1 event occurred at
07:31 a.m. PDT, 8:23 p.m. PDT, and 8:36 p.m. PDT, respec-
tively. The Los Angeles Times (2016) reported that retired
USGS seismologist Lucy Jones tweeted that evening that mag-
nitude 4 earthquakes near the San Andreas “increase the chance”
of a big earthquake “a little bit. But we have swarms without big
[earthquakes]—most likely nothing more will happen.”
The following day, 27 September, at 8:30 a.m. PDT, CE-
PEC convened a conference call at the urging of Cal OES Di-
rector Mark Ghilarducci. More than 150 microearthquakes
had occurred, most of M<2, and television news broadcasts
had reported on the swarm an hour earlier (KTLA, 2016). As
in 2001 and 2009, CEPEC recommended that Cal OES issue
an Advisory, stating that, based on several models, “stresses as-
sociated with this earthquake swarm may increase the probability
of a major earthquake on the San Andreas Fault to values be-
tween 0.03 percent and 1.0 percent for a M7.0 or larger earth-
quake occurring over the next week (to 09:00 hrs PDT, Tuesday,
October 4, 2016)”(CEPEC, 2016).
Cal OES issued the recommended Advisory, and on the
afternoon of 27 September, held a conference call with emer-
gency managers in eight southern California counties to dis-
cuss the earthquake swarm. Cal OES also posted a notice on
their public website at 10:45 p.m. that evening, but that no-
tice rephrased CEPEC’s statement as having concluded “that
stresses associated with this recent earthquake swarm may in-
crease the probability an additional earthquake as large, or
larger, than the September 26 magnitude 4.3 occurring over
the next week.”CalOEScorrectedthatnoticeon30Septem-
ber to state the following, consistent with reflecting CEPEC’s
actual conclusion verbiage: “This advisory was updated to
clarify that CEPEC specifically evaluated the potential for the
earthquake swarms to trigger a larger earthquake (M7.0+) on
the San Andreas Fault. Scientists estimated values between
0.03 percent and 1.0 percent possibility of that happening.”
(California Office of Emergency Services, 2016).
794 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

Major news outlets reported CEPEC’s conclusion after
this update, with one article posted that morning, 30 Septem-
ber, quoting Lucy Jones’tweet that the swarm “is over and risk
mostly gone”(CBS Los Angeles, 2016).
DISCUSSION
When a potential foreshock has temporarily elevated the prob-
ability of a damaging earthquake, California has been able to
alert the public without significant adverse reaction by exercis-
ing its Earthquake Advisory Plan. As of December 2016, 10
advisories have been issued, but no mainshock has occurred
during any of them. This outcome is consistent with the
approximate 5% probability that an earthquake is a foreshock
to a mainshock within the time–space window identified by
Jones (1985). Although such peer-reviewed statistical findings
on foreshock occurrence constitute the scientific basis for the
Advisory Plan, CEPEC has often exercised scientific judgment
when deciding whether to recommend an earthquake Advi-
sory. In particular, from the potential foreshocks they have
evaluated, CEPEC scientists have winnowed out several events
that they decided were probably not foreshocks. The vigor of
the aftershock sequence and possible observations of aseismic
deformation are among the factors that CEPEC has considered
(J. Parrish, personal comm., 2012), although these features have
not been shown to reliably diagnose foreshocks. CEPEC has
also assessed swarms consisting of events smaller than those
included in statistical analyses. As of this writing, no damaging
earthquake has occurred before which CEPEC had met and
recommended against an Advisory.
Advisories need to be issued much more rapidly. Each of
the 10 advisories recommended by CEPEC was publicly issued
hours after potential foreshocks occurred. Unfortunately, these
time lags are long compared with the less than 30 min interval
between foreshock and mainshock for about half the events
studied by Jones (1985). Also, the two largest foreshocks of
the 1857 Mw7.9 Fort Tejon earthquake preceded the main-
shock by 2 and 1 hr, respectively (Meltzner and Wald, 1999).
More recently, foreshock–mainshock intervals for the 1986
ML6.4 Chalfant Valley and 1987 Mw6.6 Superstition Hills
earthquakes were 24.2 and 11.4 hr, respectively, yet advisories
could not be issued before those mainshocks because CEPEC
did not convene until after they had occurred. The protocol
whereby CEPEC’s summary statement is transmitted to Cal
OES for issuance could be made more efficient, though must
continue to honor the key principle of separating scientists’
responsibility (using their expertise to judge how seismic activ-
ity is likely to evolve) from emergency managers’responsibility
(deciding the best way to use that information to promote pub-
lic safety). In our current Internet and social media environ-
ment, the public wants instantaneous information following
earthquakes, and information voids can be filled with unscien-
tific speculation and misinformation.
Damaging mainshocks have also occurred more than a few
days after foreshocks. CEPEC addressed this reality following
the 1992 Landers earthquake by recommending the go-to-war
scenario, in which seismologists and emergency managers
maintained heightened vigilance after public advisories ex-
pired. Deployment of additional seismic or geodetic monitor-
ing stations could exploit these long foreshock–mainshock
intervals not only to benefit emergency preparedness but also
to advance scientific understanding of how the events in an
earthquake sequence are linked.
Little effort has been made to determine whether issuing
advisories has reduced the risk of losses in earthquakes. Do ad-
visories promote preparedness or stimulate constructive actions
such as voluntary seismic retrofitting? Do negative outcomes
such as rumors have significant consequences? Do successive
unfulfilled advisories reduce public confidence in earthquake
science or lead to a cry wolf syndrome? Future earthquake
advisories should be better documented and their benefits
(or lack thereof ) evaluated. Even if CEPEC deliberations must
remain confidential to encourage frank exchange of scientific
ideas, summary reports should be prepared and archived. Social
science research is needed to facilitate continual improvement
in the usefulness of advisories to the public at risk.
The Advisory Plan could be adapted for use outside of
California. Where seismicity rates are lower, historic and in-
strumental earthquake catalogs generally include too few events
for robust earthquake statistics. However, CEPEC has success-
fully assessed several situations outside the scope of statistical
studies. People everywhere inevitably crave earthquake forecast
information whenever seismic activity begins, and it is in every-
one’s interest that earthquake scientists and public officials pro-
vide authoritative information.
The appropriate division of responsibilities between earth-
quake scientists and emergency managers is a key feature of the
Advisory Plan that can be implemented anywhere. The Plan
specifies that forecasting the behavior of seismic sequences
must be done by earthquake scientists, while designing appro-
priate responses to such forecasts is the job of experts in emer-
gency management, informed by social science research. This
separation of scientific forecasting from decision making by
stakeholders has also been recommended by the International
Commission on Earthquake Forecasting for Civil Protection
(2011) in the wake of the 2009 L’Aquila earthquake.
The California Earthquake Advisory Plan is a pioneering
effort to derive public safety benefits from uncertain earthquake
science, supplemented by expert judgment. Implementing,
maintaining, and improving such a protocol for collaboration
between earthquake scientists and emergency managers provides
a necessary framework for advising the public during situations
of elevated earthquake probability.
DATA AND RESOURCES
All data used in this article came from published sources listed
in the references. Some plots were made using the Generic
Mapping Tools v.4.2.1 (www.soest.hawaii.edu/gmt, last accessed
February 2014; Wessel et al., 2013). The data on the Landers
earthquake are obtained from scedc.caltech.edu/significant/
landers1992.html (last accessed May 2016).
Seismological Research Letters Volume 88, Number 3 May/June 2017 795
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

ACKNOWLEDGMENTS
We thank John Parrish, California State Geologist, for access to
unpublished papers documenting some California Earthquake
Prediction Evaluation Council (CEPEC) proceedings. We
thank Ruth Harris, Tom Brocher, Steve Hickman, Suzanne
Hecker, Kate Long, and two anonymous reviewers for their
comments and suggestions.
REFERENCES
Agnew, D. C., and L. M. Jones (1991). Prediction probabilities from fore-
shocks, J. Geophys. Res. 96, 11,959–11,971.
Atwater, B. F., A. Nelson, J. Clague, G. Carver, D. Yamaguchi, P. Bobrow-
sky, J. Bourgeois, M. Darienzo, W. Grant, E. Hemphill-Haley, et al.
(1995). Summary of coastal geologic evidence for past great earth-
quakes at the Cascadia subduction zone, Earthq. Spectra 11, 1–18.
Bakun,W. H. (2000). Seismicity of California’s north coast, Bull. Seismol.
Soc. Am. 90, 797–812.
Bakun, W. H., B. Aagaard, B. Dost, W. L. Ellsworth, J. L. Hardebeck, R.
A. Harris, C. Ji, M. J. S. Johnston, J. Langbein, J. J. Lienkaemper,
et al. (2005). Implications for prediction and hazard assessment
from the 2004 Parkfield earthquake, Nature 437, 969–974.
California Department of Conservation (2004). California Earthquake
Evaluation Prediction Council Assesses UCLA Prediction Research,
http://www.conservation.ca.gov/index/news/2004%20News%20Releases/
Pages/NR2004_Special_CEPEC_Prediction_Analy sis.aspx (last accessed
December 2016).
California Earthquake Prediction Evaluation Council (CEPEC) (1976).
Palmdale bulge evaluation, Calif. Geol. 29, no. 7, 153.
California Earthquake Prediction Evaluation Council (CEPEC) (1977).
Earthquake prediction evaluation guidelines, Calif. Geol. 30, no. 7,
158–160.
California Earthquake Prediction Evaluation Council (CEPEC) (2004).
Report to the Director, Governor’s Office of Emergency Services By the
California Earthquake Prediction Evaluation Council March 2,
2004,http://www.seismic.ca.gov/pdf.files/CEPEC_KB‑Report_3‑2‑4.pdf
(last accessed February 2017).
California Earthquake Prediction Evaluation Council (CEPEC) (2016).
Memorandum Re: The Salton Sea earthquake swarm of September,
2016,http://www.oesnews.com/wp‑content/uploads/2016/09/CEPEC
‑MEMORANDUM.pdf (last accessed December 2016).
California Office of Emergency Services (2016). UPDATED: Governor’s
Director of Emergency Services Reminds Californians to be Prepared for
Earthquakes,http://www.oesnews.com/governors‑director‑emergency‑
services‑reminds‑californians‑prepared‑earthquakes/ (last accessed De-
cember 2016).
CBS Los Angeles (2016). Earthquake Advisory Issued for Southern Cal-
ifornia after Salton Sea Quake Swarm,http://losangeles.cbslocal.
com/2016/09/30/earthquake‑advisory‑issued‑for‑southern‑california‑
after‑salton‑sea‑quake‑swarm/ (last accessed January 2017).
Dodge, D. A., G. C. Beroza, and W. L. Ellsworth (1995). Foreshock
sequence of the 1992 Landers, California, earthquake and its
implications for earthquake nucleation, J. Geophys. Res. 100,
9865–9880.
Field, E. H., T. E. Dawson, K. R. Felzer, A. D. Frankel, V. Gupta, T. H.
Jordan, T. Parsons, M. D. Petersen, R. S. Stein, R. J. Weldon II, et al.
(2008). The Uniform California Earthquake Rupture Forecast,
Version 2 (UCERF 2), U.S. Geol. Surv. Open-File Rept. 2007-1437,
104 pp., pubs.usgs.gov/of/2007/1437/ (last accessed June 2016).
FreeRepublic.com (2007). CEPEC Issues “Earthquake Advisory”For Bay
Area Counties in Response to Magnitude 5.6 Earthquake,http://www.
freerepublic.com/focus/f‑news/1919305/posts (last accessed May
2016).
Geschwind, C.-H. (2001). California Earthquakes: Science, Risk, and the
Politics of Hazard Mitigation, Johns Hopkins University Press, Bal-
timore, Maryland, 337 pp.
Goltz, J. (1985). The Parkfield and San Diego earthquake predictions: A
chronology, Special Report by the Southern California Earthquake
Preparedness Project, Los Angeles, California.
Goltz, J. D. (2015). A further note on operational earthquake forecasting:
An emergency management perspective, Seismol. Res. Lett. 86,
1231–1233, doi: 10.1785/0220150080.
Grant, L. B., and P. M. Shearer (2004). Activity of the offshore Newport–
Inglewood Rose Canyon fault zone, coastal southern California,
from relocated microseismicity, Bull. Seismol. Soc. Am. 94, 747–752.
Hall, S. S. (2011). At fault? Nature 477, 264–269.
Hardebeck, J. L., J. Boatwright, D. Dreger, R. Goel, V. Graizer, K. Hud-
nut, C. Ji, L. Jones, J. Langbein, J. Lin, et al. (2004). Preliminary
report on the 22 December 2003, M 6.5 San Simeon, California
earthquake, Seismol. Res. Lett. 75, 155–172.
Harris, R. A. (1998a). The Loma Prieta, California, earthquake of
October 17, 1989—Forecasts, in The Loma Prieta, California, Earth-
quake of October 17, 1989: Earthquake Occurrence,W. H. Bakun and
W. H. Prescott (Coordinators), U.S. Geol. Surv. Profess. Pap. 1550-B,
28 pp.
Harris, R. A. (1998b). Forecasts of the 1989 Loma Prieta, California,
earthquake, Bull. Seismol. Soc. Am. 88, 898–916.
Harris, R. A., and R. J. Archuleta (1988). Slip budget and potential for a M 7
earthquake in central California, Geophys. Res. Lett. 15, 1215–1218.
Harris, R. A., and R. W. Simpson (1992). Changes in static stress on
southern California faults after the 1992 Landers earthquake,
Nature 360, 251–254, doi: 10.1038/360251a0.
Hauksson, E., and L. M. Jones (1989). The 1987 Whittier Narrows earth-
quake sequence in Los Angeles, southern California: Seismological
and tectonic analysis, J. Geophys. Res. 94, 9569–9589.
Hauksson, E., L. M. Jones, and K. Hutton (1995). The 1994 Northridge
earthquake sequence in California: Seismological and tectonic as-
pects, J. Geophys. Res. 100, 12,335–12,355.
Hauksson, E., L. M. Jones, K. Hutton, and D. Eberhart-Phillips (1993).
The 1992 Landers earthquake sequence: Seismological observations,
J. Geophys. Res. 98, 19,835–19,858.
Ingram, M. (1983). Earthquake prediction response: Legal authorities
and liabilities—Final draft, Southern California Earthquake Prepar-
edness Project, October 1983 report.
International Commission on Earthquake Forecasting for Civil Protection
(2011). Operational earthquake forecasting: State of knowledge and
guidelines for utilization, Ann. Geophys. 54, doi: 10.4401/ag-5350.
Jones, L. M. (1984). Foreshocks (1966–1980) in the San Andreas system,
California, Bull. Seismol. Soc. Am. 74, 1361–1380.
Jones, L. M. (1985). Foreshocks and time-dependent earthquake hazard
assessment in southern California, Bull. Seismol. Soc. Am. 75,
1669–1679.
Jones, L. M. (1994). Foreshocks, aftershocks, and earthquake probabil-
ities: Accounting for the Landers earthquake, Bull. Seismol. Soc.
Am. 84, 892–899.
Jones, L. M., L. K. Hutton, D. D. Given, and C. R. Allen (1986). The
North Palm Springs, California, earthquake sequence of July 1986,
Bull. Seismol. Soc. Am. 76, 1830–1837.
Jones, L. M., K. E. Sieh, D. C. Agnew, C. R. Allen, R. Bilham, M. Ghil-
darducci, B. H. Hager, E. Hauksson, K. W. Hudnut, D. D. Jackson,
et al. (1991). Short-term earthquake hazard assessment for the San
Andreas fault in southern California, U.S. Geol. Surv. Open-File
Rept. 91-32, 42 pp.
Jordan, T., and L. M. Jones (2010). Operational earthquake forecasting:
Some thoughts on why and how, Seismol. Res. Lett. 81, 571–584.
Jordan, T. H. (2013). Lessons of L’Aquila for operational earthquake
forecasting, Seismol. Res. Lett. 84, 4–7.
Jordan, T. H., W. Marzocchi, A. J. Michael, and M. C. Gerstenberger
(2014). Operational earthquake forecasting can enhance earthquake
preparedness, Seismol. Res. Lett. 85, 955–959.
796 Seismological Research Letters Volume 88, Number 3 May/June 2017
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021

KTLA (2016). Nearly 200 earthquakes strike Salton Sea area, including 3
above magnitude 4.0, in quake swarm, http://ktla.com/2016/09/27/
salton‑sea‑area‑hit‑with‑more‑than‑100‑earthquakes‑including‑3‑above
‑magnitude‑4‑0‑in‑quake‑swarm/ (last accessed January 2017).
Los Angeles Times (2016). Nearly 200 earthquakes hit Salton Sea amid
quake swarm, http://www.latimes.com/local/lanow/la‑me‑quake‑swarm‑
20160926‑snap‑story.html (last accessed December 2016).
Maeda, K. (1996). The use of foreshocks in probabilistic prediction
along the Japan and Kuril trenches, Bull Seismol. Soc. Am. 86,
242–254.
Meltzner, A., and D. J. Wald (1999). Foreshocks and aftershocks of the
great 1857 California earthquake, Bull. Seismol. Soc. Am. 89, 1109–
1120.
Michael, A. J. (2012). Fundamental questions of earthquake statistics,
source behavior, and the estimation of earthquake probabilities
from possible foreshocks, Bull. Seismol. Soc. Am. 102, 2547–2562.
Nic Bhloscaidh, M., J. McCloskey, and C. J. Bean (2014). Response of the
San Jacinto fault zone to static stress changes from the 1992 Land-
ers earthquake, J. Geophys. Res. 119, 8914–8935, doi: 10.1002/
2014JB011164.
Northern California Earthquake Data Center (NCEDC) (2014).
Northern California Earthquake Data Center, UC Berkeley Seismo-
logical Laboratory, Dataset, doi: 10.7932/NCEDC.
Olson, J. A. (1990). Seismicity in the twenty years preceding the
Loma Prieta, California earthquake, Geophys. Res. Lett. 17,
1429–1432.
Oppenheimer, D., G. Beroza, G. Carver, L. Dengler, J. Eaton, L. Gee, F.
Gonzalez, A. Jayko, W. H. Li, M. Lisowski, et al. (1993). The Cape
Mendocino, California, earthquakes of April 1992: Subduction at
the triple junction, Science 261, 433–438.
Oppenheimer, D. H.,W. H. Bakun, T. Parsons, R. W. Simpson, J. Boat-
wright, and R. A. Uhrhammer (2010). The 2007 M 5.4 Alum Rock,
California, earthquake: Implications for future earthquakes on the
central and southern Calaveras fault, J. Geophys. Res. 115, no. B08305,
doi: 10.1029/2009JB006683.
Perfettini, H., R. S. Stein, R. Simpson, and M. Cocco (1999). Stress trans-
fer by the 1988–1989 M5:3and 5.4 Lake Elsman foreshocks to
the Loma Prieta fault: Unclamping at the site of peak mainshock
slip, J. Geophys. Res. 104, 20,169–20,182.
Ramos, G., and K. Reich (1986). More big temblors could rock Sierra in
next few days, Los Angeles Times,July23,articles.latimes.com/print/
1986‑07‑23/news/mn‑21554_1_geological‑survey (last accessed May
2012).
Reasenberg, P. A. (1999). Foreshock occurrence before large earthquakes,
J. Geophys. Res. 104, 4755–4768.
Reasenberg, P. A., and L. M. Jones (1989). Earthquake hazard after a
mainshock in California, Science 243, 1173–1176.
Reasenberg, P. A., and L. M. Jones (1994). Earthquake aftershocks: Up-
date, Science 265, 1251.
Reich, K. (1992). Emergency Officials Attempt to Squelch Rumors, Los
Angeles Times, July 16, http://articles.latimes.com/print/1992‑07‑16/
news/mn‑3643_1_state‑office (last accessed December 2012).
Reitherman,R.K.(1986).Emergencyplanningforearthquakepredic-
tions, Proc. of Conf. XXXII: Workshop on Future Directions in
Evaluating Earthquake Hazards of Southern California,W. M.
Brown III, W. J. Kockelman, and J. I. Ziony (Editors), 12–13 No-
vember, Los Angeles, California, U.S. Geol. Surv. Open-File Rept.
86-401,257–269 pp.
Savage, J. C., and R. S. Cockerham (1987). Quasi-periodic occurrence
of earthquakes in the 1978–1986 Bishop-Mammoth Lakes
sequence, eastern California, Bull. Seismol. Soc. Am. 77,
1347–1358.
Savage, M. K., and D. M. dePolo (1993). Foreshock probabilities in the
western Great-Basin eastern Sierra Nevada, Bull. Seismol. Soc. Am.
83, 1910–1938.
Scharer, K. M., G. P. Biasi, R. J. Weldon II, and T. E. Fumal (2010).
Quasi-periodic recurrence of large earthquakes on the southern
San Andreas fault, Geology 38, 555–558, doi: 10.1130/G30746.1.
Segou, M., T. Parsons, and W. Ellsworth (2013). Comparative e valuation
of physics-based and statistical forecasts in northern California, J.
Geophys. Res. 118, 6219–6240, doi: 10.1002/2013JB010313.
Southern California Seismic Network (SCSN) (2016). 09/26/2016, M
4.3 Event near Bombay Beach,http://www.scsn.org/index.php/2016/
09/26/09262016‑m4‑3‑event‑near‑bombay‑beach/ (last accessed De-
cember 2016).
State of California, Governor’s Office of Emergency Services (1990a).
California Short-Term Earthquake Prediction Response Plan, State
of California, Governor’s Office of Emergency Services, 100 p.
State of California, Governor’s Office of Emergency Services (1990b).
California Earthquake Advisory Plan, State of California, Governor’s
Office of Emergency Services, 38 p.
Stein, R. S., G. C. P. King, and J. Lin (1992). Change in failure stress on
the southern San Andreas fault system caused by the 1992 magni-
tude = 7.4 Landers earthquake, Science 258, 1328–1332, doi:
10.1126/science.258.5086.1328.
Stover, C. W., and J. L. Coffman (1993). Seismicity of the United States,
1568–1989 (Revised), U.S. Geol. Surv. Profess. Pap. 1527, 427 pp.
Sykes, L. R., and S. C. Jaumé (1990). Seismic activity on neighbouring faults
as a long-term precursor to large earthquakes in the San Francisco Bay
area, Nature 348, 595–599.
U.S. Geological Survey (USGS) (1981). Scenarios of possible earthquakes
affecting major California population centers, with estimates of in-
tensity and ground shaking, U.S. Geol. Surv. Open-File Rept. 81-115,
Vol. vi, 36 pp.
Wald, D. J., D. V. Helmberger, and S. H. Hartzell (1990). Rupture process
of the 1987 Superstition Hills earthquake from the inversion of
strong-motion data, Bull. Seismol. Soc. Am. 80, 1079–1098.
Wang, K., and G. C. Rogers (2014). Earthquake preparedness should not
fluctuate on a daily or weekly basis, Seismol. Res. Lett. 85, 569–571,
doi: 10.1785/0220130195.
Wessel, P.,W. H. F. Smith, R. Scharroo, J. F. Luis, and F. Wobbe (2013).
Generic Mapping Tools: Improved version released, Eos Trans.
AGU 94, 409–410.
Wolpert, S. (2004). Earthquakes can be predicted months in advance, re-
port UCLA scientists, who predicted San Simeon earthquake, UCLA
Newsroom, available at newsroom.ucla.edu/releases/Earthquakes‑Can‑Be
‑Predicted‑Months‑4835 (last accessed December 2015).
Working Group on California Earthquake Probabilities (WGCEP)
(1988). Probabilities of large earthquakes occurring in California
on the San Andreas fault, U.S. Geol. Surv. Open-File Rept. 88-398.
Wyatt, F. K., D. C. Agnew, and M. T. Gladwin (1994). Continuous mea-
surements of crustal deformation for the 1992 Landers earthquake
sequence, Bull. Seismol. Soc. Am. 84, 768–779.
Evelyn Roeloffs
U.S. Geological Survey
1300 SE Cardinal Court
Vancouver, Washington 98683 U.S.A.
evelynr@usgs.gov
James Goltz
Earthquake, Tsunami and Volcanic Hazards Program (retired)
California Office of Emergency Services
600 Garfield Avenue
South Pasadena, California 91030 U.S.A.
Published Online 8 March 2017
Seismological Research Letters Volume 88, Number 3 May/June 2017 797
Downloaded from http://pubs.geoscienceworld.org/ssa/srl/article-pdf/88/3/784/2745926/784.pdf
by james goltz
on 05 March 2021