Time-dependent seismic hazard maps for the New Madrid

seismic zoneand Charleston, South

Carolina areas

James Hebden Seth Stein

Department of Earth and Planetary

Sciences

Northwestern University

Underestimated uncertainty and

biasin measured

speed of light 1875-1960

Uncertainties are hard to assess and generally

underestimated

IS NEW MADRID IS AS HAZARDOUS AS CALIFORNIA?

Frankel et al., 1996

U.S. Geological Survey How robust is this model result?

HIGH MODELED HAZARD RESULTS FROM ASSUMPTIONS

- Redefined from maximum acceleration predicted at10% probability in 50 yr to 2% in 50 yr (1/ 500 yr to 1/2500 yr)

- Large magnitude of 1811-12 and thus future large earthquakes

-High ground motion in large events

- Time-independent recurrence of large events

Arbitrary choice on policy/economic grounds

Don’t understand how to best model recurrence

Lack of data

Uncertainty in interpreting intensity data

New Madrid hazard higher than California

results largely from redefining hazard as

largest shaking expected every

2500 yr:Not so for 500 yr

500 yr 2500 yr

Searer & Freeman, 2002

500 yr

2500 yr

Newman et al., 2001

PREDICTED HAZARD

DEPENDS ON ASSUMED MAXIMUM

MAGNITUDE OF LARGEST

EVENTS AND ASSUMED GROUND

MOTION MODEL

EFFECTS OF ASSUMED GROUND

MOTION MODEL

Effect as large as one magnitude unit

Frankel model, developed for maps, predicts significantly greater shaking for M >7

Frankel M 7 similar to other models’ M 8

Frankel & Toro models averaged in 1996 maps; Atkinson & Boore not used

Newman et al., 2001

ASSUMED HAZARD DEPENDS ON EARTHQUAKE PROBABILITY ASSUMPTION

Constant since last event: time independent

Small after last event, then grows: time dependent

Time dependent lower until ~2/3 mean recurrence

Results depend on model & parameters

Time dependent lower until ~2/3 mean recurrence

Charleston & New Madrid "early" in their cycles so time dependent predicts lower hazard

RELATIVE PREDICTED HAZARD DEPENDS ON POSITION IN EARTHQUAKE CYCLE

CHARLESTON

2% in 50 yr (1/2500 yr)

NEW MADRID

2% in 50 yr (1/2500 yr)

Time dependent model for

eastern US predicts lower New Madrid &

Charleston hazard

Effect larger than lowering

Mmax and thus ground motion

model

Mw 7.7 (NMSZ)Mw 7.3 (Charleston)

? ?9k 7k 6k 4k12k 3k 1k Today

Portageville Cycle Reelfoot Cycle New Madrid Cycle

SlipCluster

SlipCluster

SlipCluster

Quiescent Quiescent Quiescent

Holocene Punctuated SlipEarthquake history inferred from Mississippi river channels

Holbrook et al., 2006

NEW MADRID HAZARD WOULD BE EVEN LOWER IF RECENT EARTHQUAKE CLUSTER IS ENDING

Stein & Newman, 1994

GPS sites show little or no interseismic motion

Present seismicity may be 1811-12 aftershocks

Predicted seismic hazard depends on whether large earthquake recurrence modeled as time independent or time dependent

Time dependent model predicts lower New Madrid & Charleston hazards

Magnitude of effect depends on model type & parameters

Effect can be larger than Mmax or ground motion model

Significant contribution to uncertainty

SUMMARY