odology to evaluate losses to the total performance of extended infrastructure systems, including transportation, lifelines, and acute health care. Challenging problems emerge from the interactions among these systems, for example, how damage to the regional transportation system affects local acute health care. Assessing these system-level effects will require much improved capabilities for regional-scale simulations of earthquake-induced ground motions.
Earthquake risk, measured in expected losses over a given period of time, depends on the seismic hazard through the exposure and on the vulnerability of the built environment (Figure 1.4). A sound strategy for the reduction of earthquake risk has four basic components: better characterization of seismic hazards; land-use policies to limit exposure to seismic hazards; preparation of the built environment to withstand future earthquakes; and rapid response to earthquake disasters.
Characterizing and mapping seismic hazards has progressed substantially in recent years, but much work remains to be done in collecting information on active faults and incorporating new results from earthquake research into national and global seismic hazard maps. In many regions of the world, there are few data on active faults, and hazard estimates (Figure 3.16) have too little resolution and accuracy to be useful. In many areas covered with thick vegetation, even the location of these faults is unknown. Earthquake forecasting information, such as fault slip rates and dates of past earthquakes, is badly needed. Regional deformation measurements made with GPS can provide useful constraints on the expected rate of large earthquakes across zones of deformation. The time dependence of seismic hazard can be evaluated based on the dating of past earthquakes and calculations of fault interaction. The synthesis of these research efforts will lead to a detailed global map of ground-motion forecasts.
Another key goal is the development of urban seismic hazard maps for populous centers in active seismic zones. These maps would integrate the latest models of time-dependent earthquake hazard with estimates of site response and the effects of sedimentary basins, which typically underlie major cities. Such urban seismic hazard maps would have much more spatial detail than national or regional seismic hazard maps. The urban maps would show variations of seismic hazard maps over a few city blocks. Maps depicting global ground-motion forecasts and detailed seismic hazard in high-risk urban areas are criti-