GC44A-01
Scientific, Engineering, and Financial Factors of the 1989 Human-Triggered Newcastle Earthquake in Australia
This presentation emphasizes the dualism of natural resources exploitation and economic growth versus geomechanical pollution and risks of human-triggered earthquakes. Large-scale geoengineering activities, e.g., mining, reservoir impoundment, oil/gas production, water exploitation or fluid injection, alter pre-existing lithostatic stress states in the earth's crust and are anticipated to trigger earthquakes. Such processes of in-situ stress alteration are termed geomechanical pollution. Moreover, since the 19th century more than 200 earthquakes have been documented worldwide with a seismic moment magnitude of 4.5 < M$_{w}$<7.3, which are triggered by geoengineering activities. Particularly since the second half of the 20th century, the number earthquakes increased rapidly. An example of a human-triggered earthquake is the 1989 Newcastle event in Australia that was a result of almost 200 years of coal mining and water over-exploitation, respectively. This earthquake, an M$_{w}$=5.6 event, caused more than 3.5 billion U.S. dollars in damage (1989 value) and was responsible for Australia's first and only to date earthquake fatalities. It is therefore thought that, the Newcastle region tends to develop unsustainably if comparing economic growth due to mining and financial losses of triggered earthquakes. An hazard assessment, based on a geomechanical crust model, shows that only four deep coal mines were responsible for triggering this severe earthquake. A small-scale economic risk assessment identifies that the financial loss due to earthquake damage has reduced mining profits that have been re-invested in the Newcastle region for over two centuries beginning in 1801. Furthermore, large-scale economic risk assessment reveals that the financial loss is equivalent to 26% of the Australian Gross Domestic Product (GDP) growth in 1988/89. These costs account for 13% of the total costs of all natural disasters (e.g., flooding, drought, wild fires) and 94% of the costs of all earthquakes recorded in Australia between 1967 and 1999. In conclusion, the increasing number and size of geoengineering activities, such as coal mining near Newcastle or planned carbon dioxide Geosequestration initiatives, represent a growing hazard potential, which can negatively affect socio-economic growth and sustainable development. Finally, hazard and risk degrees, based on geomechanical-mathematical models, can be forecasted in space and over time for urban planning in order to prevent economic losses of human-triggered earthquakes in the future.
GC44A-02
Seaside, Oregon, Tsunami Vulnerability Assessment Pilot Study
The results of a pilot study to assess the risk from tsunamis for the Seaside-Gearhart, Oregon region will be presented. To determine the risk from tsunamis, it is first necessary to establish the hazard or probability that a tsunami of a particular magnitude will occur within a certain period of time. Tsunami inundation maps that provide 100-year and 500-year probabilistic tsunami wave height contours for the Seaside-Gearhart, Oregon, region were developed as part of an interagency Tsunami Pilot Study(1). These maps provided the probability of the tsunami hazard. The next step in determining risk is to determine the vulnerability or degree of loss resulting from the occurrence of tsunamis due to exposure and fragility. The tsunami vulnerability assessment methodology used in this study was developed by M. Papathoma and others(2). This model incorporates multiple factors (e.g. parameters related to the natural and built environments and socio-demographics) that contribute to tsunami vulnerability. Data provided with FEMA's HAZUS loss estimation software and Clatsop County, Oregon, tax assessment data were used as input to the model. The results, presented within a geographic information system, reveal the percentage of buildings in need of reinforcement and the population density in different inundation depth zones. These results can be used for tsunami mitigation, local planning, and for determining post-tsunami disaster response by emergency services. (1)Tsunami Pilot Study Working Group, Seaside, Oregon Tsunami Pilot Study--Modernization of FEMA Flood Hazard Maps, Joint NOAA/USGS/FEMA Special Report, U.S. National Oceanic and Atmospheric Administration, U.S. Geological Survey, U.S. Federal Emergency Management Agency, 2006, Final Draft. (2)Papathoma, M., D. Dominey-Howes, D.,Y. Zong, D. Smith, Assessing Tsunami Vulnerability, an example from Herakleio, Crete, Natural Hazards and Earth System Sciences, Vol. 3, 2003, p. 377-389.
GC44A-03
Quantification and visualization of the human impacts of anticipated precipitation extremes in South America
The research described here quantifies and visualizes the human impacts of extreme events, which in turn can lead to enhanced disaster readiness levels as well as response or mitigation strategies. Specifically, we investigate the space-time impact of anticipated precipitation extremes on human population in South America. The research attempts to integrate two recent and ongoing lines of research. In the first study (Sabesan et al., 2006; Abercrombie et al, 2006) LandScan� high-resolution population data sets were used to develop threat metrics in space and time. In the second study (Khan et al, 2006; Kuhn and Ganguly, 2006), grid-based observations of precipitation time series in South America were utilized to quantify the probability of precipitation extremes in space and time and define a geo-referenced �extremes volatility ratio� (EVR) for unanticipated, or the "truly unusual", extremes. Here we define an �extremes volatility index� (EVI) which scales from zero to unity and provides an anticipated measure of surprise corresponding to the truly unusual extremes. An EVI of zero indicates no possibility of surprise with the truly unusual extremes statistically identical to the "typical extremes", or the extremes considered, for example, in engineering design. We investigate the EVI in conjunction with maps for ambient population in South America obtained from a high-resolution global population database called LandScan� to produce a �human risk index� (HRI) in space and time. The EVI is roughly interpreted as a probability number which is multiplied with the population at each grid in space and time to obtain a measure of risk. Future research needs to explore measures of risk that consider other costs of disasters, for example impacts on critical infrastructures. A geo-referenced index, the �disaster impact index� (DII) is proposed. The DII at each grid is computed by dividing the HRI with the Gross Domestic Product (GDP) for each country. The GDP is utilized for each country as a proxy for the ability of a country to respond to disasters. Future research needs to develop more robust measures for disaster response which include the availability of disaster warning and management systems, economic development and other geo-political considerations. The research methodologies developed here can be generalized to develop threat profiles for extreme events in other disciplines.
GC44A-04
Spatial and temporal variations of the geohazard situation in Norway under a changing climate
Various types of slides (snow, rock, clay, and debris) pose the main geohazards in Norway. Such events have caused more than 2000 deaths and considerable damage to infrastructure and the natural environment over the last 150 years. An integrated research project, GEOEXTREME, is set up to investigate the coupling between climatic parameters and slides, extrapolate this into the near future with a potentially changing climate, and estimate socioeconomic implications. Slides are preceded by a complex interaction of processes acting at short and long time scales and are further complicated by local and regional variations in the snow cover and geology. This makes it difficult to predict deterministically the time and location of any slide event. Short- and long-term meteorological variables such as precipitation exhibit a strong control on the timing of slide release and can be used to predict the probability of slides, given a set of meteorological parameters. We have coupled a database consisting of more than 20000 recorded slide events with a climate database to assess the predictability of slides caused by meteorological conditions. The slide database mainly contains events that have interfered with humans or infrastructure and are from all parts of Norway spanning the period from 1961 to 2005. The climate database contains daily maps of precipitation, temperature and wind for Norway. Meteorological parameters have been extracted for each slide event based on its location and date. These meteorological parameters are used as independent variables to predict slide days on local, regional and national scales using classification trees and logistic regression. The analyses show a high degree of predictability of days with slide events at the local scale, but a decrease in predictability with increasing scales. The most important meteorological triggering parameters for each slide type show spatial variations, reflecting the varying climate in Norway. As part of the project, climate scenarios are produced for various regions of Norway. Based on these scenarios and the analyses of the historical data, the project aims at assessing the expected slide activity in the various regions of the country, as well as evaluating the effects of this on the Norwegian society over the coming 50 years.
http://www.geoextreme.no
GC44A-05
Dynamic Natural Disaster Risk Assessment: A case study for Jamaica
Since 2000, natural disasters have caused 6,500 casualties, affected 10.3 million people, and created an estimated $9.3 billion in damages in the Caribbean region. Given the particularly active recent hurricane seasons, rapid development, and the influence of global warming a key concern is whether changes in intensity and frequency of climate-related events will increase risks substantially in the future. Previous studies have examined disaster risk on regional or global scales; however, these studies are limited in their deterministic and static approaches to risk assessment. This study evaluates risk assessment using a dynamic and probability based methodology for evaluating the impact of natural disasters. Specifically it provides a framework to assess present and future disaster risk while also considering the different types of uncertainty associated with these estimates. Using Jamaica as a case study, the study looks at tropical cyclones to evaluate direct economic losses as well as estimate the repercussions on GDP and fiscal impacts. The study then looks into the future to present a first order approximation of how and whether predicted climate change impacts could substantially increase disaster risk. A Catastrophe Simulation Model is used to compare how the GDP of Jamaica is influenced by changes in the intensity of tropical cyclone events. These impacts are compared with projections of economic development to determine the relative scales at which future losses could occur. In order to derive a better understanding of the impact of natural disaster losses on a broader scale, the study considers the uncertainties surrounding these assumptions and projections to provide a guideline for future research.
GC44A-06
Evaluating the Potential Usefulness of new Hurricane Indices for Emergency Management and Other Decision Makers
Over the past 35 years, the Saffir-Simpson scale has used wind speed as a means for categorizing damage and surge risks associated with hurricanes. Time has shown, however, that hurricanes with the same wind speed do not necessarily cause equal damage values and storm-surge heights. Therefore, it is prudent to now consider a different method for categorizing storms so that emergency management officials in a coastal location can have a better idea as to the potential hazards posed by a particular hurricane. Recognizing this need, three new indices were developed by Lakshmi Kantha in 2005 for evaluating hurricane intensity, hurricane damage potential, and hurricane surge potential. This paper applies these indices to a twenty-year database (1986-2005) of Atlantic, U.S.-landfalling hurricanes and compares the relative indices to known damage estimates and surge heights. Some general conclusions will be made regarding the possible usefulness of these indices for emergency management officials in areas prone to landfalling tropical cyclones.
GC44A-07
An Adaptive Regional Input-Output Model and its Application to the Assessment of the Economic Cost of Katrina
Recent events, especially natural disasters, have raised concerns about the response of local economies to large exogenous shocks. Clearly, our ability to assess the total economic costs of large-scale events like the Katrina's landfall is very poor, and numerous questions needs to be investigated. Among them, the taking into account of complex interactions within the economic system --- between economic sectors or regions --- is the topic of intense research. Several studies have approached these issues using Input- Output models, which are powerful tools to assess how a shock, on one or several sectors, propagate into the economy through intermediate consumption and demand. These models, however, have two main short- comings. First, since they do not take into account productive capacity, they cannot assess the consequence of a shock on the supply-side. Second, they do not allow for flexibility in the economic system in that producers and consumers cannot respond to a lack in input. The presentation will describe an adaptive regional input-output model, which aims at answering the previous concerns through the introduction of an hybrid modelling methodology. This model takes into account economic propagations through both intermediate demands and production shortages, and also represents the response of economic agents facing production shortages. An application of this model to the evaluation of the economic cost of Katrina will be demonstrated, highlighting both modelling and data issues. Also, beyond a simple ``best- guess'' assessment, systematic sensitivity analyses will be presented, allowing for an evaluation of the accuracy one can expect from such a model.