U31B-01 INVITED
Extreme Earthquakes: Implications of the Current State of Earthquake Risk Assessment and Forecasting
Major earthquakes are a type of extreme event that can produce great damage and loss of life. As world populations become more heavily concentrated in seismically active regions, it is only a matter of time before a major earthquake destroys a large metropolitan area. Until recently, it was thought that little or no information could be detected about future major earthquake locations or occurrence times using only seismic catalogs. The state-of-the-art in spatial information about future earthquakes has been the national and world hazard maps, which are actually forecasts of ground shaking probabilities over a 50-year time window. Similarly, the temporal dependence of major earthquakes has been thought to be governed by Poisson statistics, implying that future earthquakes have no memory of past earthquake locations or occurrence times. However, very recent research is leading to a substantial reappraisal of the information content of standard seismic catalogs. Using new ensemble classifier technology, together with signal detection theory, both of which grow out of modern pattern analysis methods, we are now finding that much tighter limits can be placed upon the space and time windows in which future major earthquakes will occur. Using these ideas, it may be possible to assess earthquake risk by computing actual space-time probabilities relating to the occurrence of major earthquakes in a region. In this talk, I discuss current research in this field, as well as the difficulties that may occur in practical use of these ideas in the public policy and economic arenas.
U31B-02
Extreme Subduction Earthquake Scenarios and their Economical Consequences for Mexico City and Guadalajara, Jalisco, Mexico
The destructive effects of large magnitude, thrust subduction superficial (TSS) earthquakes on Mexico City (MC) and Guadalajara (G) has been shown in the recent centuries. For example, the 7/04/1845 a TSS earthquake with Ms 7+ and epicentral distance of about 250 km from MC occurred on the coast of the state of Guerrero, a Maximum Mercalli Modified Intensity (MMI) of IX-X was reported in MC. Furthermore, the 19/09/1985 a Ms 8.1, Mw 8.01, TSS earthquake with epicentral distance of about 340 km from MC occurred on the coast of the state of Michoacan, a maximum MMI of IX-X was reported in MC. Also, the largest, Ms 8.2, instrumentally observed TSS earthquake in Mexico, occurred in the Colima-Jalisco region the 3/06/1932, with epicentral distance of the order of 200 km from G in northwestern Mexico. The 9/10/1995 another similar event, Ms 7.4, Mw 8, with an epicentral distance of about 240 km from G, occurred in the same region and produced MMI IX in the epicentral zone and MMI up to VI in G. The frequency of occurrence of large TSS earthquakes in Mexico is poorly known, but it might vary from decades to centuries [1]. On the other hand, the first recordings of strong ground motions in MC dates from the early 1960´s and most of them were recorded after the 19/09/1985 earthquake. In G there is only one recording of the later event, and 13 for the one occurred the 9/10/1995 [2]. In order to fulfill the lack of strong ground motions records for large damaging TSS earthquakes, which could have an important economical impact on MC [3] and G, in this work we have modeled broadband synthetics (obtained with a hybrid model that has already been satisfactorily compared with observations of the 9/10/1995 Colima-Jalisco Mw 8 earthquake, [4]) expected in MC and G, associated to extreme magnitude Mw 8.5, TSS scenario earthquakes with epicenters in the so-called Guerrero gap and in the Colima-Jalisco zone, respectively. The proposed scenarios are based on the seismic history and up to date seismotectonic, seismological, geophysical, and geotechnical information for the mentioned subduction zones and for MC and G. The economical impacts of the proposed extreme TSS earthquake scenarios for MC and G are fully discussed. We acknowledge the support of DGSCA, UNAM, for using its supercomputer facilities. ----------------------- [1] Nishenko S.P. and Singh SK, BSSA 77, 6, 1987 [2] Chavez M. and Ramirez R., 12th World Conf. Earthq. Eng., 2000 [3] Cabrera E., Chavez M., Madariaga R., Mai M, Frisenda M., Perea N., AGU, Fall Meeting, 2005 [4] Chavez M., Olsen K.B., Cabrera E., 13th World Conf. Earthq. Eng., 2004
U31B-03
A regional index approach to probabilistic modeling of extreme earthquakes
We discuss the application of extreme value theory to earthquake hazards. Recent innovations in flood frequency analysis provide a new perspective on the analysis of large earthquakes. Probability plot correlation coefficient hypothesis tests and L-moment goodness-of-fit evaluations reveal that the Gumbel (GUM) distribution provides a good approximation to the probability distribution function (pdf) of series of annual maximum (AM) earthquake magnitudes. Homogeneity tests based on the theory of L-moments further reveal that broad regions of the globe are homogeneous in the sense that the AM observations of earthquake magnitudes are well approximated by a GUM pdf with fixed upper moments. The homogeneity of global earthquake data across broad tectonic environments enables us to pool data into a regional pdf of earthquake magnitudes, termed an index earthquake distribution. Research in hydrology has shown that frequency analysis based on pooling of data using an analogous index-flood method are much more accurate than frequency analysis based on site or region specific data. The index earthquake distribution is a dimensionless GUM distribution with fixed scale parameter so only the mean earthquake magnitude must be estimated for a region to define the frequency distribution of large earthquakes. We show how the degree of spatial homogeneity of earthquake magnitudes across broad tectonic environments can be exploited to yield improved estimates of the risk posed by extreme earthquake magnitudes.
U31B-04 INVITED
Observations and Impact Assessments of Extreme Space Weather Events
"Space weather" refers to conditions on the Sun, in the solar wind, and in Earth`s magnetosphere, ionosphere, and thermosphere. Activity on the Sun such as solar flares and coronal mass ejections can lead to high levels of radiation in space and can cause major magnetic storms at the Earth. Space radiation can come as energetic particles or as electromagnetic emissions. Adverse conditions in the near-Earth space environment can cause disruption of satellite operations, communications, navigation, and electric power distribution grids. This can lead to a variety of socioeconomic losses. Astronauts and airline passengers exposed to high levels of radiation are also at risk. Society`s vulnerability to space weather effects is an issue of increasing concern. We are dependent on technological systems that are becoming more susceptible to space weather disturbances. We also have a permanent human presence in space with the International Space Station and the President and NASA have expressed a desire to expand our human space activities with missions to the moon and Mars. This will make space weather of even greater concern in the future. In this talk I will describe many space weather effects and will describe some of the societal and economic impacts that extreme events have had.
U31B-05
Analysis of the Risk and Vulnerability of the Cancun Beach System-Wilma Hurricane Case
In the last decade, many researchers have been focused on the growth in risk associated with global warming and its implications; such as rising sea levels, increasing cyclone frequency and intensity, among others. However, in some cases, for an adequate understanding of the processes, it is also important to incorporate short time analysis of anthropogenic modifications that induce increased vulnerability, for example the effects of Hurricane Wilma (2005) at Cancun, Mexico. Cancun is located on the Mexican Caribbean Sea (latitude 21º05' N, longitude 86º46' W) and is the most important tourist destination in Mexico. For this research several studies have been carried out integrating previous reports, historical photo analysis, field work and the application of several numerical models (wave, currents, storm surge, sediment transport, etc.) for the characterization of the system for normal and extreme conditions. The measurements of wave conditions during the passing of Hurricane Wilma in front of Cancun show maximum wave heights of around 18 m, mean wave periods of 16 s, surface and bottom currents of 2 m/s. Incredibly, more than 7 million cubic meters of sand were moved from the Cancun beach system to other coast cells thus leaving the resort with no beach. The data presented concerning modifications on the barrier island demonstrates that these extreme meteorological events were responsible for the littoral changes, due to the loss of system flexibility in the biological dynamics and physical equilibrium of the systems, with social, environmental and economic implications. The main conclusion of this work is that local anthropogenic modifications have induced more vulnerability and risk to Cancun beach than those associated with global warming.
U31B-06
Flood Risk in Megadelta Coastal Cities: Lessons From New Orleans
The flood hazard posed by storm surges to the city of New Orleans is increasing for three reasons. First, as a
result of the city's location on thick deposits of recent delta sediments along the edge of an oceanic basin, it is
sinking at geologically rapid rates. Second, over the last decade, global sea level has increased as a result of
climate change and is projected to continue to rise in the future. And third, the level of Atlantic basin hurricane
activity has also increased, with the biggest increase for the strongest storms (with the largest surges),
particularly in and around the Gulf of Mexico.
The future flood risk in New Orleans has been explored using stochastic hurricane and storm surge generation
models. Levels of risk have been determined at different geographical locations within the city today, and have
been estimated for the future based on likely changes in the flood hazard. The study has also incorporated
vulnerability and breaching models to determine the probabilities of failure and likely breach size, relative to the
height of the storm surge outside, for each section of the flood defences that protect the city.
Flood maps of relative risk throughout the region were developed from high resolution digital elevation maps,
showing expected flood depths at various return periods. For example, the map for the 100-year return flood
shows the extent of flooding that is expected on average once every 100 years, corresponding to an annual
exceedance probability of one per cent. Additionally, four locations within the city of New Orleans were chosen for
a more detailed study of predicted flood depth return periods, using the simple metric of the modeled return
period of first flooding from a storm surge.
The study shows that while the risk of flooding from storm surges can be reduced by the repair and enhancement
of flood defences, the risk will begin to rise again as soon as improvements stop. The situation in New Orleans
raises wider questions about the viability of megadelta coastal cities that are faced with the impacts of climate
change.
http:www.rms.com
U31B-07
Is Hurricane Mitigation in the United States Cost-Effective?
Costs associated with hurricanes have been rising steeply since well before the record-breaking Atlantic
hurricane season of 2005; the rise has been especially dramatic since 1980. Many researchers (van der Vink et
al., 2005) have demonstrated that these costs are driven by a small number of extreme events. Such events
generally overwhelm most or all protection mechanisms in place. When these mechanisms are overwhelmed,
many areas which were "protected" suffer damages tantamount to the damages which would have occurred
without any protection at all. When this occurs, the best proxy for the cost of the hurricane is simply the amount of
people and property in the way of the hurricane. There are several benefits to hurricane mitigation. Any mitigation
effort will decrease the expected damages in an area affected by a hurricane provided that the mitigation
mechanisms are not overwhelmed. Mitigation efforts will also lessen the probability that these mechanisms are
overwhelmed at all. However, mitigation efforts also decrease the perceived risk of moving people and property
into the area which the efforts are designed to protect. If mitigation results in more people and property being
moved into a vulnerable area, this increases the expected loss for an event which overwhelms mitigation
mechanisms. Even if this is a low-probability event, this still increases the overall expected loss in a given year
with the mitigation in place. As overall expected losses increase, the net benefit of the mitigation efforts will
decrease. This study finds that the above effects of mitigation are not currently taken into account in cost-benefit
analyses of mitigation. It examines quantitative relationships between mitigation efforts, population increases,
and the resulting changes in expected losses. While it cannot be proven that mitigation efforts in the United
States are not cost-effective, the decrease in the net benefits represents a substantial portion of the estimated
benefits of mitigation. These impacts should be taken into account whenever new mitigation projects are
undertaken.
http:www.princeton.edu/~abrett/Thesis