Contents

## How is the hazard defined Where do we expect earthquakes When do we expect earthquakes What will happen in those earthquakes

Hadley, Graham, Managing Editor has reference to this Academic Journal, PHwiki organized this Journal TOPIC 3: HOW WELL CAN WE PREDICT EARTHQUAKE HAZARDS Predictions are important as long as hazard mitigation policy How much should we believe them HAZARD ASSESSMENT IS HARD It has been described as “a game of chance of which we still don’t know all the rules Lomnitz, 1989 AND WHAT GOES INTO A MAP IS OFTEN NOT EXPLAINED OR EXPLAINED BADLY Its “simplicity is deeply veiled by user-hostile notation, antonymous jargon, in addition to proprietary software” Hanks in addition to Cornell, 1994 How is the hazard defined Where do we expect earthquakes When do we expect earthquakes What will happen in those earthquakes

This Particular University is Related to this Particular Journal

How is the hazard defined Hazard isnt a physical thing we measure. Its something mapmakers define on policy grounds. How they define hazard is the largest factor in determining the hazard. Different choices lead to different predicted hazards in addition to thus favor different policies. Frankel et al., 1996 Algermissen et al., 1982 Hazard redefined from maximum acceleration predicted at 10% probability in 50 yr (1/ 500 yr ) to much higher 2% in 50 yr (1/2500 yr) New Madrid hazard higher than Cali as long as nia results largely from redefining hazard as largest shaking expected every 2500 yr: Not so as long as 500 yr 500 yr 2500 yr Searer & Freeman, 2002 500 yr 2500 yr

ASSUMED HAZARD DEPENDS ON DEFINITION TIME WINDOW Over 100 years, Cali as long as nia site much more likely to be shaken strongly than NMSZ one Over 1000 years, more NMSZ sites shaken strongly once; many in Cali as long as nia shaken many times Short time relevant as long as buildings with 50-100 yr life Shaken areas MMI > VII R in addition to om seismicity simulation including seismicity & ground motion differences Where do we expect earthquakes Can use Earthquake history Plate motions Geology GPS On plate boundaries, these agree. In other places, we have to chose which to use Different choices lead to different predicted hazards Long record needed to see real hazard Swaf as long as d & Stein, 2007 1933 M 7.3 1929 M 7.2

Our glacial loading model suggests that earthquakes may occur anywhere along the rifted margin which has been glaciated. Stein et al., 1979 1985 Concentrated hazard bull’s-eyes at historic earthquake sites 2005 Diffuse hazard along margin GSC Map depends greatly on assumptions & thus has large uncertainty Peak Ground Acceleration 10% probability of exceedance in 50 years (once in 500 yr) GSHAP (1999) Present Study HUNGARY: ALTERNATIVE HAZARD MAPS Concentrated hazard inferred from historic seismicity alone Diffuse hazard inferred incorporating geology Toth et al., 2004 When do we expect earthquakes When we have a long history, we can estimate the average recurrence time – but theres a lot of scatter When we have a short history, we estimate the recurrence time of large earthquakes from small ones, but this can be biased In either case, we have to assume either that the probability of large earthquakes stays constant with time, or that it changes Different choices lead to different predicted hazards

EARTHQUAKE RECURRENCE IS HIGHLY VARIABLE M>7 mean 132 yr s 105 yr Estimated probability in 30 yrs 7-51% Sieh et al., 1989 Extend earthquake history with paleoseismology When we have a long history, we can estimate the average recurrence time – but theres a lot of scatter Mean 132 105 Mean 180 72 We can describe these using various distributions – Gaussian, log-normal, Poisson but its not clear that one is better than another Gutenberg-Richter relationship log10 N = a -b M N = number of earthquakes occurring M a = activity rate (y-intercept) b = slope M = Magnitude When we have a short history, we estimate the recurrence time of large earthquakes from small ones, but this can be biased

POSSIBLE BIASES IN ESTIMATING THE MAGNITUDE AND RECURRENCE TIME OF LARGE EARTHQUAKES FROM THE RATE OF SMALL ONES Undersampling: record comparable to or shorter than mean recurrence – Usually find too-short recurrence time. Can also miss largest events Direct paleoseismic study: Magnitude overestimated, recurrence underestimated Events missed, recurrence overestimated Earthquake Rate Stein & Newman, 2004 CHARACTERISTIC UNCHARACTERISTIC SIMULATIONS Short history: often miss largest earthquake or find a too-short recurrence time 10,000 synthetic earthquake histories as long as G-R relation with slope b=1 Gaussian recurrence times as long as M> 5, 6, 7 Various history lengths given in terms of Tav, mean recurrence as long as M>7 Stein & Newman, 2004 Long history: Can still find too-short or too-long recurrence time Stein & Newman, 2004

RESULTS VARY WITH AREA SAMPLED Stein et al., 2005 Increasing area around main fault adds more small earthquakes ASSUMED HAZARD DEPENDS ON EARTHQUAKE PROBABILITY ASSUMPTION Constant since last event: time independent (cant be overdue) Small after last event, then grows: time dependent Time dependent lower until ~2/3 mean recurrence Details depend on model & parameters Hebden & Stein, 2008 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 Hebden & Stein, 2008 Southern San Andreas broke in 1857 M 7.7 Fort Tejon, late in cycle so time-dependent predicts higher hazard (overdue)

Cali as long as nia Time-dependant probabilities Increased on southern San Andreas CHARLESTON 2% in 50 yr (1/2500 yr) Hebden & Stein, 2008 At present, time dependent predicts ~50% lower hazard Still less in 2250 What will happen in large earthquakes Major unknowns are magnitude of the earthquake in addition to the ground shaking it will produce Tradeoff between these two parameters Different choices lead to different predicted hazards

EFFECTS OF ASSUMED GROUND MOTION MODEL Effect as large as one magnitude unit Frankel model predicts significantly greater shaking as long as M >7 Frankel M 7 similar to other models M 8 Newman et al., 2001 Newman et al., 2001 Assumed maximum magnitude of largest events has largest effect near main fault Assumed ground motion model has regional effect because it also applies to small earthquakes off main fault When we look at a hazard map, remember that it is just one of a large number of quite different in addition to equally likely maps one could make, depending on model assumptions How is the hazard defined Where do we expect earthquakes When do we expect earthquakes What will happen in those earthquakes Often the last (Mmax, ground motion model) is discussed the most but the other assumptions are more important

Comparing maps made as long as different assumptions shows which features are best constrained (robust) We use these maps, but Its hard to say how good they are Wont know as long as 100s or 1000s of years, when we have enough experience to see how good their predictions were. Where the data are good, the assumptions in addition to thus predictions are probably pretty good. Where the data are poorer, the predictions are probably poorer. Our best bet is probably to look at any given map, ask whether the prediction makes sense, in addition to act accordingly. 2% in 50 yr (1/2500 yr) 154% %106 Large uncertainty in maps 54% effect in Memphis New Madrid: 200 years into hypothesized 500 year recurrence

## Hadley, Graham Managing Editor

Hadley, Graham is from United States and they belong to Daily Home, The and they are from Talladega, United States got related to this Particular Journal. and Hadley, Graham deal with the subjects like Local News; Regional News

## Journal Ratings by Ecole Franaise d’Electronique et d’Informatique

This Particular Journal got reviewed and rated by Ecole Franaise d’Electronique et d’Informatique and short form of this particular Institution is FR and gave this Journal an Excellent Rating.