Earthquakes: Risk, Detection, Warning, and Research Peter Folger Specialist in Energy and Natural Resources Policy April 19, 2010 Congressional Research Service 7-5700 RL33861 CRS Report for Congress Prepared for Members and Committees of Congress Earthquakes: Risk, Detection, Warning, and Research Summary The 1994 Northridge (CA) earthquake caused as much as $26 billion (in 2005 dollars) in damage and was one of the costliest natural disasters to strike the United States. The Federal Emergency Management Agency has estimated that earthquakes cost the United States over $5 billion per year. A hypothetical scenario for a magnitude 7.8 earthquake in southern California estimated a possibility of 1,800 fatalities and over $200 billion in economic losses. The January 12, 2010, magnitude 7.0 earthquake that struck Haiti only 15 miles from Port-au-Prince, the capital city, has caused an estimated 230,000 fatalities and 300,000 injuries. On February 27, 2010, a magnitude 8.8 earthquake occurred offshore of Chile and has caused massive damage, although far fewer fatalities than the Haiti earthquake. The Cascadia Subduction Zone megathrust fault, which stretches from southern British Columbia to northern California, has the potential to generate a massive earthquake and tsunami, similar to the Chilean earthquake, that could threaten Oregon, Washington, and northern California. The United States faces the possibility of large economic losses from earthquake-damaged buildings and infrastructure. California alone accounts for most of the estimated annualized earthquake losses for the nation, and with Oregon and Washington the three states account for nearly $4.1 billion (77%) of the U.S. total estimated annualized loss. A single large earthquake, however, can cause far more damage than the average annual estimate. An ongoing issue for Congress is whether the federally supported programs aimed at reducing U.S. vulnerability to earthquakes are an adequate response to the earthquake hazard. Under the National Earthquake Hazards Reduction Program (NEHRP), four federal agencies have responsibility for long-term earthquake risk reduction: the U.S. Geological Survey (USGS), the National Science Foundation (NSF), the Federal Emergency Management Agency (FEMA), and the National Institute of Standards and Technology (NIST). They variously assess U.S. earthquake hazards, send notifications of seismic events, develop measures to reduce earthquake hazards, and conduct research to help reduce overall U.S. vulnerability to earthquakes. Congress reauthorized NEHRP in 2004 (P.L. 108-360) through FY2009. Appropriations for NEHRP from FY2005 to FY2009 did not meet authorized levels; the total funding enacted was $613.2 million, approximately 68% of the total amount of $902.4 million authorized by P.L. 108360. The American Recovery and Reinvestment Act (ARRA; P.L. 111-5) provided some additional funding for earthquake activities under NEHRP through FY2010. In the 111th Congress, H.R. 3820 would reauthorize NEHRP through FY2014, authorizing total appropriations of $906 million over five years for the program, with 90% of the funding designated for the USGS and NSF, and the remainder for FEMA and NIST. The total authorized amounts are slightly greater than what was authorized by P.L. 108-360 over five years. If future appropriations match the funding levels authorized under H.R. 3820, these funds would exceed the total cumulative amounts actually appropriated between FY2005 and FY2009. What effect funding at the levels enacted through FY2009 under NEHRP has had on the U.S. capability to detect earthquakes and minimize losses after an earthquake occurs is not clear. It is also difficult to predict precisely how NEHRP reauthorized under H.R. 3820 would achieve a major goal of the bill: to reduce the loss of life and damage to communities and infrastructure through increasing the adoption of hazard mitigation measures. A perennial issue for Congress is whether activities under NEHRP can reduce the potential for catastrophic loss in the next giant earthquake to strike the United States. Congressional Research Service Earthquakes: Risk, Detection, Warning, and Research Contents Introduction ................................................................................................................................1 National Earthquake Hazards Reduction Program (NEHRP) .......................................................1 A Shift in Program Emphasis to Hazard Reduction................................................................2 NEHRP Reauthorization in the 111th Congress ......................................................................4 Authorization of Appropriations in H.R. 3820 .................................................................5 Other Changes Proposed in H.R. 3820 ............................................................................6 Earthquake Hazards and Risk......................................................................................................6 Potential Losses from Earthquakes ...................................................................................... 12 A Decrease in Estimated Loss?...................................................................................... 15 The New Madrid Seismic Zone ........................................................................................... 16 Monitoring................................................................................................................................ 17 Advanced National Seismic System (ANSS) ....................................................................... 17 ANSS Funding.............................................................................................................. 17 Dense Urban Networks ................................................................................................. 18 Regional Networks........................................................................................................ 18 Backbone Stations......................................................................................................... 18 National Strong-Motion Project (NSMP)....................................................................... 18 Global Seismic Network (GSN) .......................................................................................... 19 Detection, Notification, and Warning......................................................................................... 19 National Earthquake Information Center (NEIC)................................................................. 20 ShakeMap........................................................................................................................... 20 Prompt Assessment of Global Earthquakes for Response (PAGER) ..................................... 22 Pre-disaster Planning: HAZUS-MH .................................................................................... 24 Research—Understanding Earthquakes ..................................................................................... 24 U.S. Geological Survey....................................................................................................... 24 National Science Foundation............................................................................................... 25 Conclusion................................................................................................................................ 26 Figures Figure 1. NEHRP Agency Responsibilities and End Users of NEHRP Outcomes .........................3 Figure 2. Earthquake Hazard in the United States ........................................................................7 Figure 3. Histogram of the Number of U.S. Earthquakes from 2000 to 2008 by Magnitude (1.0 to 6.9)..............................................................................................................9 Figure 4. Example of a ShakeMap............................................................................................. 21 Figure 5. Example of PAGER Output for the January 12, 2010, Magnitude 7.0 Haiti Earthquake............................................................................................. 23 Tables Table 1. Authorized and Enacted Funding for NEHRP.................................................................4 Congressional Research Service Earthquakes: Risk, Detection, Warning, and Research Table 2. NEHRP Authorization for Appropriations Under H.R. 3820 ...........................................5 Table 3. Urban Areas Facing Significant Seismic Risk.................................................................9 Table 4. Earthquakes Responsible for Most U.S. Fatalities Since 1970....................................... 10 Table 5. The 10 Most Damaging Earthquakes in the United States............................................. 11 Table 6. U.S. Metropolitan Areas with Estimated Annualized Earthquake Losses of More Than $10 Million ................................................................................................................... 13 Contacts Author Contact Information ...................................................................................................... 27 Congressional Research Service Earthquakes: Risk, Detection, Warning, and Research Introduction Close to 75 million people in 39 states face some risk from earthquakes. Earthquake hazards are greatest in the western United States, particularly California, but also Alaska, Washington, Oregon, and Hawaii. Earthquake hazards are also prominent in the Rocky Mountain region and the New Madrid Seismic Zone (a portion of the central United States), as well as portions of the eastern seaboard, particularly South Carolina. Given the potentially huge costs associated with a large, damaging earthquake in the United States, an ongoing issue for Congress is whether the federally supported earthquake programs are appropriate for the earthquake risk. Under the National Earthquake Hazards Reduction Program (NEHRP), the federal government supports efforts to assess and monitor earthquake hazards and risk in the United States. Four federal agencies, responsible for long-term earthquake risk reduction, coordinate their activities under NEHRP: the U.S. Geological Survey (USGS), the National Science Foundation (NSF), the Federal Emergency Management Agency (FEMA), and the National Institute of Standards and Technology (NIST). Congress reauthorized NEHRP in 2004 (P.L. 108-360), and authorized appropriations through FY2009 for a total of $902.4 million over five years. In the 111th Congress, H.R. 3820 (Title I) would reauthorize NEHRP through FY2014. This report discusses: • NEHRP—the multi-agency federal program to reduce the nation’s risk from earthquakes; • earthquake hazards and risk in the United States; • federal programs that support earthquake monitoring; • the U.S. capability to detect earthquakes and issue notifications and warnings; and • federally supported research to improve the fundamental scientific understanding of earthquakes with a goal of reducing U.S. vulnerability. National Earthquake Hazards Reduction Program (NEHRP) In 1977 Congress passed the Earthquake Hazards Reduction Act (P.L. 95-124) establishing NEHRP as a long-term earthquake risk reduction program for the United States. The program initially focused on research, led by USGS and NSF, toward understanding and ultimately predicting earthquakes. Earthquake prediction has proved intractable thus far, and the NEHRP program shifted its focus to minimizing losses from earthquakes after they occur. FEMA was created in 1979 and President Carter designated it as the lead agency for NEHRP. In 1980, Congress reauthorized the Earthquake Hazards Reduction Act (P.L. 96-472), defining FEMA as the lead agency and authorizing additional funding for earthquake hazard preparedness and mitigation to FEMA and the National Bureau of Standards (now NIST). Congressional Research Service 1 Earthquakes: Risk, Detection, Warning, and Research A Shift in Program Emphasis to Hazard Reduction In 1990, Congress reauthorized NEHRP (P.L. 101-614) and made substantive changes, to decrease the emphasis on earthquake prediction, clarify the role of FEMA, clarify and expand the program objectives, and require federal agencies to adopt seismic safety standards for new and existing federal buildings. In 2004, Congress reauthorized NEHRP through FY2009 (P.L. 108360) and shifted primary responsibility for planning and coordinating NEHRP from FEMA to NIST. It also established a new interagency coordinating committee and a new advisory committee, both focused on earthquake hazards reduction. The current program activities are focused on four broad areas: • developing effective measures to reduce earthquake hazards; • promoting the adoption of earthquake hazards reduction measures by federal, state, and local governments, national building standards and model building code organizations, engineers, architects, building owners, and others who play a role in planning and constructing buildings, bridges, structures, and critical infrastructure or lifelines; • improving the basic understanding of earthquakes and their effects on people and infrastructure, through interdisciplinary research involving engineering, natural sciences, and social, economic, and decision sciences; and • developing and maintaining the Advanced National Seismic System (ANSS), the George E. Brown Jr. Network for Earthquake Engineering Simulation (NEES), and the Global Seismic Network (GSN).1 The House Science Committee report in the 108th Congress on H.R. 2608 (P.L. 108-360) noted that NEHRP has produced a wealth of useful information since 1977, but it also stated that the program’s potential has been limited by the inability of the NEHRP agencies to coordinate their efforts.2 The committee asserted that restructuring the program with NIST as the lead agency, directing funding towards appropriate priorities, and implementing it as a true interagency program would lead to improvement. The 2004 reauthorization directed the Director of NIST to chair the Interagency Coordinating Committee. Other members of the committee include the directors of FEMA, USGS, NSF, the Office of Science and Technology Policy, and the Office of Management and Budget. The Interagency Coordinating Committee is charged with overseeing the planning, management, and coordination of the program. Primary responsibilities for the NEHRP agencies break down as follows (see also Figure 1): • NIST supports the development of performance-based seismic engineering tools and works with other groups to promote the commercial application of the tools through building codes, standards, and construction practices. 1 ANSS is a nationwide network of seismographic stations operated by the USGS. GSN is a global network of stations coordinated by the Incorporated Institutions for Seismology. NEES is an NSF-funded project that consists of 15 experimental facilities and an IT infrastructure with a goal of mitigating earthquake damage by the use of improved materials, designs, construction techniques, and monitoring tools. 2 U.S. House, Committee on Science, National Earthquake Hazards Reduction Program Reauthorization Act of 2003, H.Rept. 108-246 (Aug. 14, 2003), p. 13. Congressional Research Service 2 Earthquakes: Risk, Detection, Warning, and Research • FEMA assists other agencies and private-sector groups to prepare and disseminate building codes and practices for structures and “lifelines”,3 and aids development of performance-based codes for buildings and other structures. • USGS conducts research and other activities to characterize and assess earthquake risks, and (1) operates a forum, using the National Earthquake Information Center (NEIC), for the international exchange of earthquake information, (2) works with other NEHRP agencies to coordinate activities with earthquake reduction efforts in other countries, and (3) maintains seismic hazard maps in support of building codes for structures and lifelines, and other maps needed for performance-based design approaches. • NSF supports research to improve safety and performance of buildings, structures, and lifelines using the large-scale experimental and computational facilities of NEES and other institutions engaged in research and implementation of NEHRP. Figure 1. NEHRP Agency Responsibilities and End Users of NEHRP Outcomes Source: NEHRP program office at (modified by CRS). Table 1 shows the authorized and enacted appropriations for NEHRP from FY2005 through FY2010. The total enacted amount for FY2005-FY2009 was $613.2 million, or 68% of the $902.4 million total amount authorized in P.L. 108-360 over the five-year span. President Obama requested a total of $129.3 million for NEHRP in FY2011, even though authorization for the program under P.L. 108-360 expired at the end of FY2009. 3 Lifelines are essential utility and transportation systems. Congressional Research Service 3 Earthquakes: Risk, Detection, Warning, and Research Table 1. Authorized and Enacted Funding for NEHRP ($ millions) USGS NSF FEMA NIST Total Authorized 77.0 58.0 21.0 10.0 166.0 Enacted 58.4 53.1 14.7 0.9 127.1 Authorized 84.4 59.5 21.6 11.0 176.5 Enacted 54.5 53.8 9.5 0.9 118.7 FY2007 Authorized 85.9 61.2 22.3 12.1 181.5 Enacted 55.4 54.2 7.2 1.7 118.5 FY2008 Authorized 87.4 62.9 23.0 13.3 186.6 Enacted 58.1 53.6 6.1 1.7 119.5 Authorized 88.9 64.7 23.6 14.6 191.8 Enacted 61.2 55.0 9.1 4.1 129.4 FY2010 Enacted 62.8 55.3 8.9 4.1 131.1 FY2011 Requested 62.3 53.8 9.1 4.1 129.3 FY2005 FY2006 FY2009 Source: NEHRP program office, at Notes: According to the NEHRP program office, ARRA funds are not included; and the FY2011 requested budget is the estimated portion of the President’s budget request that would be allocated for NEHRP activities. Also, the FY2010 enacted amounts are estimates. NEHRP Reauthorization in the 111th Congress Title I of H.R. 3820, the Natural Hazards Risk Reduction Act of 2009, would reauthorize NEHRP through FY2014, retain NIST as the lead NEHRP agency, and authorize total appropriations of about $906 million over five years. Title II of H.R. 3820 would reauthorize the National Windstorm Impact Reduction Act (first enacted in 2004 as Title II of P.L. 108-360 and modeled after NEHRP), and Title III would create an interagency coordinating committee, chaired by the Director of NIST, that would oversee the planning and coordination of both the earthquake and wind hazards programs. The single interagency coordinating committee would replace the two separate interagency committees overseeing the current earthquake and wind hazards programs. The bill was reported by the House Science and Technology Committee on February 26, 2010, and was passed by the House on March 2, 2010. No action on the bill has been taken in the Senate. The interagency coordinating committee also would be given authority to “make proposals for planning and coordination of any other federal research for natural hazards mitigation that the Committee considers appropriate.” The potentially broader mandate for the interagency coordinating committee—to embrace all natural hazards in its deliberations—could reflect an emphasis on natural hazard mitigation presented in the bill’s “Findings” section. The bill finds that research is needed to better understand institutional, social, behavioral, and economic factors that influence how risk mitigation is implemented, and that a major goal of federally supported natural hazards-related research should be to increase the adoption of hazard mitigation measures. This theme is an aspect of an “all-hazards” approach to reducing risk, and could reflect observations presented during a June 11, 2009, hearing of the House Science and Technology Congressional Research Service 4 Earthquakes: Risk, Detection, Warning, and Research Committee at which a witness noted that “there appear to be significant similarities in societal responses to different hazards.”4 In that testimony, the witness observed that there are limited opportunities to study earthquake emergency response and recovery because damaging earthquakes are relatively infrequent; however, there are important lessons to be learned from studying other, more frequent, hazards, such as tornadoes. 5 Authorization of Appropriations in H.R. 3820 H.R. 3820 would authorize total appropriations for NEHRP of approximately $906 million for a five-year period ending in FY2014, with 90% of the funding authorized for the USGS and NSF, and the remainder for FEMA and NIST. (See Table 2.) The total authorized amounts would be slightly greater than what was authorized by P.L. 108-360 over five years. Also, funding under H.R. 3820 would exceed the amounts actually appropriated between FY2005 and FY2009 by over 60% for NEHRP if Congress enacted appropriations over five years that matched the authorized amounts. Compared to the previously enacted authorized funding, total authorized amounts would be greater for USGS and NSF under H.R. 3820 and less for NIST and FEMA. Table 2. NEHRP Authorization for Appropriations Under H.R. 3820 ($ millions) FY2010 FY2011 FY2012 FY2013 FY2014 Total FY2010-FY2014 auth. Total FY2005-FY2009 auth. Total FY2005-FY2009 enact. USGS 90.0 92.1 94.3 96.5 98.8 471.7 423.6 287.6 NSF 64.1 66.1 68.0 70.1 72.2 340.5 306.3 272.0 FEMA 10.2 10.6 10.9 11.2 11.5 54.4 111.5 46.6 NIST 7.0 7.7 7.9 8.2 8.4 39.2 61.0 9.3 Total 171.3 176.5 181.1 190.9 905.8 902.4 615.5 186 Source: U.S. House of Representatives, H.R. 3820; and NEHRP program office, at ppt_sdr.pdf. Note: Totals may not sum due to rounding. The USGS would receive the largest share—about 52%—of total authorized appropriations under H.R. 3820, as under the previous reauthorization of NEHRP, and the total amount would be approximately $48 million more than the amount authorized for FY2005 through FY2009. As with the previous reauthorization, H.R. 3820 singles out the Advanced National Seismic System (ANSS) to receive a subset of authorized appropriations within the total USGS-authorized amount. Specifically, ANSS would be authorized to receive $36 million in FY2010, $37 million in FY2011, $38 million in FY2012, $39 million in FY2013, and $40 million in FY2014. That would total $190 million over five years, compared to a total of $174 million over five years in the previous reauthorization. 4 U.S. Congress, House Committee on Science and Technology, Subcommittee on Technology and Innovation, Reauthorization of the National Earthquake Hazards Reduction Program: R&D for Resilient Communities, testimony of Dr. Michael Lindell, 111th Cong., 1st sess., June 11, 2009. 5 Ibid. Tornadoes could be considered similar to earthquakes as “rapid onset disasters” that provide little or no warning, but which elicit similar societal responses, according to Dr. Lindell. Congressional Research Service 5 Earthquakes: Risk, Detection, Warning, and Research Other Changes Proposed in H.R. 3820 Section 105 of H.R. 3820 would shift a post-earthquake investigations program from the USGS to be led instead by NIST, in consultation with the other NEHRP agencies, and the program would be organized to study the implications of earthquakes in the areas of responsibility of each NEHRP agency. The interagency coordinating committee that would be created under Title III of H.R. 3820 would be largely similar to the current interagency committee except that it would also include the National Oceanic and Atmospheric Administration (NOAA), and the committee would have the discretion to include the head of any other federal agency it considers appropriate. In addition to acting as a single coordinating committee for both the earthquake and wind hazards program, the committee is charged with developing a strategic plan for both programs,6 providing progress reports to Congress, and developing coordinated budgets for each program. Also, the Director of NIST is required to establish advisory committees for both programs, similar to the current advisory committees established under P.L. 108-360. Title III of H.R. 3820 also requires the Subcommittee on Disaster Reduction of the Committee on Environment and Natural Resources of the National Science and Technology Council to submit a report, within two years of enactment, identifying federal research, development, and technology transfer activities for natural disasters, common areas of research among the natural hazards, and opportunities to “create synergies between the research activities for the hazards.” Earthquake Hazards and Risk Figure 2 indicates that detailed information exists on where earthquakes are likely to occur in the United States and how severe the earthquake magnitude and resulting ground shaking are likely to be. The map in Figure 2 depicts the potential shaking hazard from future earthquakes. It is based on the frequency at which earthquakes occur in different areas and how far the strong shaking extends from the source of the earthquake. In Figure 2, the hazard levels indicate the potential ground motion—expressed as a percentage of the acceleration due to gravity (g). In a sense, the map shows the likelihood of where earthquakes could occur, and where the strongest shaking could take place. All 50 states and the District of Columbia are vulnerable to earthquake hazards, although risks vary greatly across the country. Seismic hazards are greatest in the western United States, particularly California, Washington, Oregon, and Alaska and Hawaii. Alaska is the most earthquake-prone state, experiencing a magnitude 7 earthquake almost every year and a magnitude 8 earthquake every 14 years on average. (See box below for a description of earthquake magnitude.) Because of its low population and infrastructure density, Alaska has a relatively low risk for large economic losses from an earthquake. In contrast, California has more citizens and infrastructure at risk than any other state because of the state’s frequent seismic activity combined with its high population. 6 The current interagency coordinating committee for NEHRP submitted a strategic plan to Congress in October 2008. It is available at Congressional Research Service 6 Earthquakes: Risk, Detection, Warning, and Research Figure 2. Earthquake Hazard in the United States Source: USGS Fact Sheet 2008-3018 (April 2008), at products_data/images/nshm_us02.gif. Modified by CRS. Note: The bar in the upper right shows the potential ground motion—expressed as a percentage of the acceleration due to gravity (g)—with up to a 1 in 10 chance of being exceeded over a 50-year period. Figure 2 also shows relatively high earthquake hazard in the Rocky Mountain region, portions of the eastern seaboard—particularly South Carolina—and a part of the central United States known as the New Madrid Seismic Zone (see “The New Madrid Seismic Zone,” below). Other portions of the eastern and northeastern United States are also vulnerable to moderate seismic hazard. According to the USGS, 75 million people in 39 states are subject to “significant risk.”7 Earthquake Magnitude and Intensity Earthquake magnitude is a number that characterizes the relative size of an earthquake. It is often reported using the Richter scale (magnitudes in this report are generally consistent with the Richter scale). Richter magnitude is calculated from the strongest seismic wave recorded from the earthquake, and is based on a logarithmic (base 10) scale: for each whole number increase in the Richter scale, the ground motion increases by 10 times. The amount of energy released per whole number increase, however, goes up by a factor of 32. The moment magnitude scale is another expression of earthquake size, or energy released during an earthquake, that roughly corresponds to the Richter magnitude and is used by most seismologists because it more accurately describes the size of very large earthquakes. Intensity is a measure of how much shaking occurred at a site based on observations and amount of damage. Intensity is usually reported on the Modified Mercalli Intensity Scale as a Roman numeral ranging from I (not felt) to XII (total destruction). 7 U.S. Geological Survey, Dept. of the Interior, Earthquake Hazards—A National Threat, Fact Sheet 2006-3016, March 2006, During the period 1975-1995, only four states did not experience detectable earthquakes: Florida, Iowa, North Dakota, and Wisconsin. See USGS Earthquake Hazards Program, Earthquake Facts, at Congressional Research Service 7 Earthquakes: Risk, Detection, Warning, and Research Shaking hazards maps, such as the one in Figure 2, are often combined with other data, such as the strength of existing buildings, to estimate possible damage in an area due to an earthquake. The combination of seismic risk, population, and vulnerable infrastructure can help improve the understanding of which urban areas across the United States face risks from earthquake hazards that may not be immediately obvious from the probability maps of shaking hazards alone. The USGS has identified 26 urban areas that face a significant seismic risk from the combination of population and severity of shaking. Table 3 lists those areas at greatest risk. The USGS estimates that several million earthquakes occur worldwide each year, but the majority are of small magnitude or occur in remote areas, and are not detectable. More earthquakes are detected each year as more seismometers8 are installed in the world, but the number of large earthquakes (magnitude greater than 6.0)9 has remained relatively constant. Between 2000 and 2008 there were between 2,261 and 3,876 earthquakes per year in the United States, according to the National Earthquake Information Center (NEIC). (See Figure 3.) National Seismic Hazards Maps and Earthquake Forecast for California On April 21, 2008, the USGS released National Seismic Hazards Maps that updated the version published in 2002. Compared to the 2002 version, the new maps indicate lower ground motions (by 10% to 25%) for the central and eastern United States, based on modifications to the ground-motion models used for earthquakes. The new maps indicate that estimates of ground motion for the western United States are as much as 30% lower for certain types of ground motion, called long-period seismic waves, which affect taller, multi-story buildings. Ground motion that affects shorter buildings of a few stories, called short-period seismic waves, is roughly similar to the 2002 maps. The new maps show higher estimates for ground motion for western Oregon and Washington compared to the 2002 maps, due to new ground motion models for the offshore Cascadia subduction zone. In formulating the 2008 maps, the USGS gave more weight to the probability of a catastrophic magnitude 9 earthquake occurring along the Cascadia subduction zone. The Cascadia subduction zone fault ruptures, on average, every 500 years, and has the potential to generate destructive earthquakes and tsunamis along the coasts of Washington, Oregon, and northern California. According to a report released on April 14, 2008, California has a 99% chance of experiencing a magnitude 6.7 or larger earthquake in the next 30 years. The likelihood of an even larger earthquake, magnitude 7.5 or greater, is 46% and will likely occur in the southern part of the state. The fault with the highest probability of generating at least one earthquake of magnitude 6.7 or greater over the next 30 years is the San Andreas in southern California (59% probability); for northern California it is the Hayward-Rodgers Creek fault (31%). The earthquake forecasts are not predictions (i.e., they do not give a specific date or time), but represent probabilities over a given time period. In addition, the probabilities have variability associated with them. The earthquake forecasts are known as the “Uniform California Earthquake Rupture Forecast (UCERF)” and are produced by a working group composed of the USGS, the California Geological Survey, and the Southern California Earthquake Center. Sources: USGS Fact Sheet 2008-3018, “2008 United States National Seismic Hazard Maps” (April 2008), at; USGS Fact Sheet 2008-3027, “Forecasting California’s Earthquakes—What Can We Expect in the Next 30 Years?” (2008), at 8 Seismometers are instruments that measure and record the size and force of seismic waves, essentially sound waves radiated from the earthquake as it ruptures. Seismometers generally consist of a mass attached to a fixed base. During an earthquake, the base moves and the mass does not, and the relative motion is commonly transformed into an electrical voltage that is recorded. A seismograph usually refers to the seismometer and the recording device, but the two terms are often used interchangeably. 9 See USGS “Earthquakes Facts and Statistics” at Congressional Research Service 8 Earthquakes: Risk, Detection, Warning, and Research Table 3. Urban Areas Facing Significant Seismic Risk (alphabetically by state for cities with at least 300,000 people) State City State City Alaska Anchorage Nevada Las Vegas California Fresno Nevada Reno California Los Angeles New Mexico Albuquerque California Sacramento New York New York California Salinas Oregon Eugene-Springfield California San Diego Oregon Portland California San Francisco-Oakland Puerto Rico San Juan California Santa Barbara South Carolina Charleston California Stockton-Lodi Tennessee Chattanooga-Knoxville Idaho Boise Tennessee Memphis Indiana Evansville Utah Provo-Orem Massachusetts Boston Utah Salt Lake City Missouri St. Louis Washington Seattle Sources: USGS Fact Sheet 2006-3016 (March 2006); USGS Circular 1188, Table 3. Note: These areas are identified using a population-based risk factor based on 1999 population data. (William Leith, ANSS Coordinator, USGS, Reston, VA, telephone conversation, Nov. 15, 2006). Number of Earthquakes Figure 3. Histogram of the Number of U.S. Earthquakes from 2000 to 2008 by Magnitude (1.0 to 6.9) 2000 1800 1600 1400 1200 1000 800 600 400 200 0 1.0 to 1.9 2.0 to 2.9 3.0 to 3.9 4.0 to 4.9 5.0 to 5.9 6.0 to 6.9 Magnitude 2000 2001 2002 2003 2004 2005 2006 2007 2008 Source: USGS, “Earthquake Facts and Statistics,” at; data as of December 7, 2009. Note: Earthquakes greater than magnitude 7.0 and less than 1.0 are not shown. According to the USGS, 6 earthquakes of magnitude 7.0 or greater occurred in the United States between 2001 and 2007. Congressional Research Service 9 Earthquakes: Risk, Detection, Warning, and Research As Figure 3 shows, about 98% of earthquakes detected each year by the NEIC are smaller than magnitude 5.0; only 59 earthquakes exceeded magnitude 6.0 for the nine-year period (less than 0.3% of the total earthquakes detected) for an average of less than seven earthquakes per year of at least 6.0 magnitude. Large earthquakes, although infrequent, cause the most damage and are responsible for most earthquake-related deaths. The great San Francisco earthquake of 1906 claimed an estimated 3,000 lives, as a result of both the earthquake and subsequent fires. Over the past 100 years, relatively few Americans have died as a result of earthquakes, compared to citizens in some other countries. 10 Since 1970, three major earthquakes in the United States were responsible for 188 of the 212 total earthquake-related fatalities (see Table 4). Table 4. Earthquakes Responsible for Most U.S. Fatalities Since 1970 Date Location Magnitude Deaths February 9, 1971 San Fernando Valley, CA 6.6 65 October 18, 1989 Loma Prieta, CA 6.9 63 January 17, 1994 Northridge, CA 6.7 60 Source: USGS, Note: Other sources report different numbers of fatalities associated with the Northridge earthquake. Since 2000, only two deaths directly caused by earthquakes have occurred in the United States, both associated with falling debris in Paso Robles (CA) during the December 22, 2003, San Simeon earthquake of magnitude 6.5. In contrast, earthquakes have been directly or indirectly responsible for more than 685,000 fatalities in other countries since 2000.11 Approximately 65% of those estimated deaths resulted from the December 2004 Indonesian earthquake (and resulting tsunami) of magnitude 9.1, and the January 2010 magnitude 7.0 earthquake in Haiti. Since the devastating Indonesian earthquake occurred, an October 2005 magnitude 7.6 earthquake in Pakistan killed approximately 86,000 people, and a May 2008 magnitude 7.9 earthquake struck Eastern Sichuan, China, causing as many as 87,000 fatalities (see box). The 1994 Northridge earthquake was the nation’s most damaging earthquake in the past 100 years, preceded five years earlier by the second most costly earthquake—Loma Prieta. Table 5 shows the 10 costliest U.S. earthquakes in terms of insured and uninsured losses. Comparing losses between different earthquakes, and between earthquakes and other disasters such as hurricanes, can be difficult because of the different ways losses are calculated. Calculations may include a combination of insured losses, uninsured losses, and estimates of lost economic activity. For example, insured losses from Hurricane Katrina in 2005—mainly property—may be $41 billion, according to one estimate. 12 Total property damage would rise if uninsured property were included; and including interrupted economic activity in the calculation could bring the total loss for Hurricane Katrina to $100 billion, according to one estimate.13 10 Estimates of earthquake-related fatalities vary, and an exact tally of deaths and injuries is rare. For more information on the difficulties of counting earthquake-related deaths and injuries, see casualty_totals.php. 11 U.S. Geological Survey, Earthquakes with 1,000 or More Deaths Since 1900, at earthquakes/world/world_deaths.php. This estimate does not include fatalities from the February 27, 2010, magnitude 8.8 Chilean earthquake, which has resulted in widespread destruction but few fatalities compared to the Indonesian, Pakistan, and Haiti earthquakes. 12 Insurance Information Institute, Loss estimates are in 2005 dollars. 13 Risk Management Solutions (RMS), Newark, CA, press release (Sept. 2, 2005), at PR_090205_HUKatrina.asp. Congressional Research Service 10 Earthquakes: Risk, Detection, Warning, and Research The May 12, 2008, Sichuan Earthquake in China and Implications for the United States On May 12, 2008, a catastrophic earthquake of magnitude 7.9 struck Eastern Sichuan, China. The epicenter was located approximately 960 miles southwest of Beijing, and the earthquake was triggered approximately 12 miles below the earth’s surface. Approximately 87,000 fatalities have been reported. The earthquake was felt in parts of eastern, southern, and central China, and as far away as Bangladesh, Taiwan, Thailand, and Vietnam. Several large aftershocks occurred after the main seismic event. The May 12 earthquake resulted from movement along a northeast-trending reverse or thrust fault, reflecting stresses from the convergence of rocks of the Tibetan Plateau, to the west, against the crust underlying the Sichuan Basin and southeastern China. The region has experienced large earthquakes in the past; on August 25, 1933, a magnitude 7.5 earthquake struck the northwestern margin of the Sichuan Basin, resulting in approximately 9,300 fatalities. Some concerns have been raised about the possibility of an earthquake of similar magnitude occurring in a seismically active region of the United States, such as southern California, where fault movement similar to the Eastern Sichuan earthquake may occur. On May 22, 2008, the USGS released a hypothetical scenario for a magnitude 7.8 southern California earthquake, called the ShakeOut Scenario. In the scenario, scientists hypothetically simulated the ground shaking and fault rupture associated with a magnitude 7.8 earthquake, and estimated the resulting damage to buildings and infrastructure. The scenario estimated approximately 1,800 fatalities and $213 billion in economic losses as a result of the earthquake. The report points to aggressive retrofitting programs that have increased the seismic resistance of buildings, highways, and other critical infrastructure in southern California as one reason why the number of possible fatalities is relatively low. Some scientists have raised the possibility that earthquakes, such as the May 12, 2008, Sichuan event, may sometimes exhibit cascading behavior, where bursts of seismic energy are released along different places in a single fault, or jump between connected faults. Earthquakes that occur along the Sierra Madre fault in southern California, for example, could trigger a series of cascading seismic events along other faults, such as the San Andreas. Seismic hazard estimates may not fully account for the damage that could be caused by cascading earthquakes along a connected fault system. Scientists are hoping to examine the Sichuan earthquake in more detail to better understand the nature of cascading seismic events and how they affect the U.S. seismic hazard estimates. Sources: Ken Hudnut, geophysicist, USGS, Pasadena, CA, phone conversation, June 11, 2008; USGS Earthquake Hazards Program, at; USGS Newsroom, Earthquake Fatalities High in 2008, at; and USGS, The ShakeOut Scenario, Open-File Report 2008-1150 (2008), at Table 5.The 10 Most Damaging Earthquakes in the United States Year 1994 Location Northridge, Magnitude $2005 (billions) CAa 6.7 $26.0 1989 Loma Prieta, CA 6.9 $11.0 1964 Anchorage, AK 9.2 $3.1 1971 San Fernando, CA 6.5 $2.7 2001 Nisqually, WA 6.8 $2.5 1987 Whittier Narrows, CA 5.9 $0.62 1933 Long Beach, CA 6.3 $0.60 1953 Kern County, CA 7.5 $0.44 1992 Landers, CA 7.6 $0.13 1992 Cape Mendocino, CA 7.1 $0.092 Source: Insurance Information Institute, at Note: Includes insured and uninsured losses. a. Estimates for total losses resulting from the Northridge earthquake vary; the Congressional Budget Office estimated $43 billion in total losses ($50 million in 2005 dollars). See Federal Reinsurance for Disasters, Congressional Budget Office (September 2002), p. 19. Congressional Research Service 11 Earthquakes: Risk, Detection, Warning, and Research January 12, 2010, Magnitude 7.0 Earthquake Strikes Haiti On Tuesday, January 12, 2010, a magnitude 7.0 earthquake struck Haiti at 4:53 p.m. The epicenter was located approximately 15 miles southwest of Port-au-Prince, and the earthquake occurred at a depth of about 8 miles, according to initial USGS reports. The relatively shallow earthquake, and its close proximity to the capital city, exposed millions of Haitians to severe to violent ground shaking. The earthquake occurred along the EnriquilloPlantain Garden fault system, a major east-west trending strike-slip fault system that lies between the Caribbean tectonic plate and the North American tectonic plate; the Caribbean plate actively moves against the North American plate and shear stresses are created at the boundary. At a strike-slip fault, the rocks move past each other horizontally along the fault line (in contrast to a thrust fault, where rocks on one side of the fault move on top of the rocks on the other side). Earthquakes comparable to or stronger than the January 12 event have struck Haiti at least four times over the past three centuries. Earthquakes in 1751 and 1770—which probably occurred along the same fault system—destroyed Port-au-Prince. Other examples of strike-slip faults are the San Andreas fault in California and the Red River fault in China. The January 12 earthquake caused widespread damage in the Port-au-Prince area, causing approximately 230,000 deaths and 300,000 injuries. Also, a series of aftershocks followed the main earthquake. There were 14 aftershocks greater than magnitude 5 and 36 greater than magnitude 4 within the first day following the magnitude 7.0 event. Aftershocks have the potential to cause further damage, especially to structures weakened by the initial large earthquake. In a February 23, 2010, statement, the USGS indicated that the frequency of aftershocks should diminish with time, but the threat of additional damaging earthquakes remains. The USGS noted that buildings in the Port-auPrince area will continue to be at risk from strong earthquake shaking, and that the fault responsible for the January 12 earthquake still stores sufficient strain to be released as a large, damaging earthquake during the lifetime of structures built during the reconstruction effort. The USGS based its probability estimates on techniques developed to assess earthquake hazards in the United States. Using these techniques, the USGS estimated that the probability of a magnitude 7 or greater earthquake occurring within the next 50 years along the Enriquillo fault near Port-au-Prince is between 5% and 15%. The range of probabilities reflects the current understanding of the seismicity and tectonics of the Haiti region. By comparison, the USGS has estimated that that the probability of a magnitude 7 or greater earthquake occurring within the next 50 years along the Hayward-Rodgers Creek fault east of San Francisco is about 15%. Sources: USGS Earthquake Hazards Program, Significant Earthquakes: Magnitude 7.0—Haiti Region, at; Michael Blanpied, Associate Coordinator for the USGS Earthquake Hazards Program, podcast, Jan. 13, 2010, at; email from David Applegate, Senior Science Advisor for Earthquake and Geologic Hazards, Jan. 13, 2010; USGS statement, “USGS Updates Assessment of Earthquake Hazard and Safety in Haiti and the Caribbean,” February 23, 2010, at See also CRS Report R41023, Haiti Earthquake: Crisis and Response, by Rhoda Margesson and Maureen Taft-Morales. Potential Losses from Earthquakes The United States faces potentially large total losses due to earthquake-caused damage to buildings and infrastructure and lost economic activity. As urban development continues in earthquake-prone regions in the United States, concerns are increasing about the exposure of the built environment, including utilities and transportation systems, to potential earthquake damage. 14 One estimate of loss from a severe earthquake in the Los Angeles area is over $500 billion. An even higher estimate—approximately $900 billion—includes damage to the heavily populated central New Jersey-Philadelphia corridor if a 6.5 magnitude earthquake occurred along a fault lying between New York City and Philadelphia.15 14 FEMA Publication 366, HAZUS MH Estimated Annualized Earthquake Losses for the United States (April 2008), at Hereafter referred to as FEMA 366. 15 A. M. Best Company Inc., 2006 Annual Earthquake Study: $100 Billion of Insured Loss in 40 Seconds (Oldwick, NJ: A.M. Best Company, 2006), p. 12. The A. M. Best report includes estimates from catastrophe-modeling companies of predicted damage from hypothetical earthquakes in Los Angeles, the Midwest, the Northeast, and Japan. The report (continued...) Congressional Research Service 12 Earthquakes: Risk, Detection, Warning, and Research Some studies and techniques combine seismic risk with the value of the building inventory16 and income losses (e.g., business interruption, wage, and rental income losses) in cities, counties, or regions across the country to provide estimations of economic losses from earthquakes. An April 2008 report from FEMA calculated that the annualized loss from earthquakes nationwide is $5.3 billion, with California, Oregon, and Washington accounting for nearly $4.1 billion (77%) of the U.S. total estimated annualized loss. 17 Table 6 shows metropolitan areas with estimated annualized U.S. earthquake losses over $10 million. Table 6. U.S. Metropolitan Areas with Estimated Annualized Earthquake Losses of More Than $10 Million (in $ millions) Rank 1 Metro area Los Angeles-Long Beach-Santa Ana, CA AEL Rank Metro area AEL $1,312 23 Reno-Sparks, NV $29 2 San Francisco-Oakland-Fremont, CA $781 24 Charleston-North Charleston, SC $22 3 Riverside-San Bernadino-Ontario, CA $397 25 Columbia, SC $22 4 San Jose-Sunnyvale-Santa Clara, CA $277 26 Stockton, CA $21 5 Seattle-Tacoma, WA $244 27 Atlanta-Sandy Springs-Marietta, GA $19 6 San Diego-Carlsbad-San Marcos, CA $155 28 Bremerton-Silverdale, WA $18 7 Portland-Vancouver-Carlsbad, OR $137 29 Ogden-Clearfield, UT $18 8 Oxnard-Thousand Oaks-Ventura, CA $111 30 Salem, OR $17 9 Santa Rosa-Petaluma, CA $69 31 Eugene-Springfield, OR $17 10 St. Louis, MO-IL $59 32 Napa, CA $16 11 Salt Lake City, UT $52 33 San Luis Obispo-Paso Robles, CA $16 12 Sacramento-Arden-Arcade-Roseville, CA $52 34 Nashville-Davidson-Murfreesboro, TN $15 13 Vallejo-Fairfield, CA $40 35 Albuquerque, NM $15 14 Memphis, TN $38 36 Olympia, WA $14 15 Santa Cruz-Watsonville, CA $36 37 Modesto, CA $13 16 Anchorage, AK $35 38 Fresno, CA $13 17 Santa Barbara-Santa Maria-Goleta, CA $34 39 Evansville, IN-KY $12 18 Las Vegas-Paradise, NV $33 40 Birmingham-Hoover, AL $11 19 Honolulu, HI $32 41 El Centro, CA $11 20 Bakersfield, CA $30 42 Little Rock-North Little Rock, AR $11 21 New York-Northern New JerseyLong Island, NY $30 43 Provo-Orem, UT $10 22 Salinas, CA $29 Source FEMA Publication 366, HAZUS MH Estimated Annualized Earthquake Losses for the United States (April 2008). Annualized earthquake losses (AEL) calculated in 2005 dollars. (...continued) cites an estimate by one such company, Risk Management Solutions (RMS), that a hypothetical 7.4 magnitude event along the Newport-Inglewood Fault near Los Angeles would cause $549 billion in total property damage. A hypothetical 6.5 magnitude earthquake along a fault between Philadelphia and New York City would produce $901 billion in total loss, according to an RMS estimate. 16 Building inventory refers to four main inventory groups: (1) general building stock, (2) essential and high potential loss facilities, (3) transportation systems, and (4) utility systems (FEMA 366). 17 FEMA 366, p. 37. Congressional Research Service 13 Earthquakes: Risk, Detection, Warning, and Research February 27, 2010, Magnitude 8.8 Earthquake in Chile: Is There a Similar Risk to the United States? A massive magnitude 8.8 earthquake struck Chile on February 27, 2010, along a subduction zone plate boundary fault 70 miles north-northeast of the city of Concepcion and offshore of the Chilean coast. The earthquake occurred at a depth of approximately 22 miles below the seafloor, much deeper than the earthquake that struck Haiti near Port-auPrince on January 12. Several cities, including Concepcion, experienced intensity VIII shaking on the Modified Mercalli Intensity Index, corresponding to severe perceived shaking. The capital city of Santiago, located 200 miles northeast of the epicenter, experienced intensity VII shaking corresponding to very strong perceived shaking. The earthquake has caused widespread damage that may amount to $3-$8 billion in insured losses, and probably more in total economic losses. Over 700 deaths have been reported, many from the tsunami generated by the subsea earthquake, and possibly 2 million people have been affected. Far fewer fatalities are expected in Chile compared to the Haiti earthquake, even though the Chilean earthquake was much more powerful, in part because the epicenter was farther from major population centers and the earthquake occurred 14 miles deeper than the Haitian event. In addition, details are beginning to emerge about the effectiveness of earthquake-resistant design and building construction in Chile as compared to Haiti. Because the earthquake occurred offshore, it generated a tsunami, which struck parts of the Chilean coastline and offshore islands, causing damage and fatalities. Tsunami warnings were issued by the National Weather Service Pacific Tsunami Warning Center for Hawaii, Japan, and other regions bordering the Pacific Ocean that may have been vulnerable to a damaging tsunami wave, although most regions far from the epicenter did not experience any serious damage. A tsunami caused significant damage to the city of Hilo, Hawaii, following the May 1960 magnitude 9.5 earthquake that also occurred along the subduction zone fault about 143 miles south of the February 27, 2010, earthquake. Why the 1960 earthquake generated a tsunami that caused damage and fatalities in Hawaii, Japan, and the Philippines while the 2010 earthquake did not is not yet well understood and is being actively studied. The magnitude 8.8 earthquake occurred along the boundary between the Nazca tectonic plate and the South American tectonic plate, which converge at a rate of about 3 inches per year. The Nazca plate is subducting under the South American plate, which rides over the top of the Nazca plate. In geologic terms, this is known as a thrust fault or megathrust, in contrast to a strike-slip fault, where the rocks on either side of the fault slide past each other. The San Andreas fault and the Enriquillo fault that caused the January 2010 Haiti earthquake are strike-slip faults. The Sumatran-Andaman megathrust fault, which triggered the December 2004 Indonesian earthquake and tsunami, is a subduction zone fault or megathrust geologically similar to the Nazca-South American tectonic plate subduction zone. Subduction zone megathrust faults generate the largest earthquakes in the world. The Cascadia Subduction Zone megathrust that stretches from mid-Vancouver Island in southern British Columbia southward to Cape Mendocino in northern California has the potential to generate a very large earthquake, similar in magnitude to the February 2010 Chilean earthquake. The fault’s proximity to the northwestern U.S. coastline—approximately 50-100 miles offshore— also poses a significant tsunami hazard; destructive waves from a large earthquake along the fault could reach the coast of Oregon and Washington in less than an hour, possibly in tens of minutes. The Cascadia Subduction Zone fault forms the boundary between the subducting Juan de Fuca tectonic plate and the overriding North American plate, very similar to the relationship between the Nazca Plate and the South American Plate off the Chilean coast. If the Cascadia Subduction Zone megathrust were to “unzip” or rupture along a large section of its entire length, it would likely generate a megathrust earthquake near magnitude 9 or more, similar to the 1964 Alaskan earthquake, the 1960 and 2010 Chilean earthquakes, and the 2004 Indonesian earthquake. Scientists have documented that the last time this occurred along the Cascadia Subduction Zone fault was in 1700. The 1700 earthquake spawned a tsunami that traveled across the Pacific Ocean and struck Japan. Because of the tectonic similarities between the Cascadia Subduction Zone megathrust and the Nazca-South American plate megathrust, scientists hope to learn a great deal about the seismic hazard in the Pacific Northwest by studying the unique strong ground motion recordings from the 2010 Chilean magnitude 8.8 earthquake. Sources: USGS Earthquake Hazards Program, Magnitude 8.8–Offshore Maule, Chile, at earthquakes/eqinthenews/2010/us2010tfan/; Eqecat, “M8.8 Earthquake Offshore Maule, Chile,” CatWatch, February 27, 2010, at (Eqecat provides loss estimates for insurance and financial institutions); Business Week, “Chile Asks for International Aid as Looting Spreads,” March 2, 2010, at; Brian F. Atwater et al., The Orphan Tsunami of 1700—Japanese Clues to a Parent Earthquake in North America, USGS, Professional Paper 1707, 2005,; e-mail from Richard Aster, Professor of Geophysics, New Mexico Institute of Mining and Technology, March 2, 2010. Congressional Research Service 14 Earthquakes: Risk, Detection, Warning, and Research Annualized earthquake loss (AEL) addresses two components of seismic risk: the probability of ground motion and the consequences of ground motion. It enables comparison between different regions with different seismic hazards and different building construction types and quality. For example, earthquake hazard is higher in the Los Angeles area than in Memphis, but the general building stock in Los Angeles is more resistant to the effects of earthquakes. The AEL annualizes the expected losses by averaging them by year. A Decrease in Estimated Loss? In its most recent publication estimating earthquake losses, FEMA noted that the $5.3 billion in annualized earthquake loss nationwide was 21% higher than the $4.4 billion calculated in FEMA’s previous report, published in February 2001.18 However, the 2001 report calculated losses using 1994 dollars, and when adjusted to reflect 2005 dollars the earlier estimate increased to $5.6 billion, indicating a small decrease in nationwide annualized earthquake loss potential since the 2001 report was published. According to FEMA, this loss occurred even though the national building inventory increased by 50% over this same period. What factors led to a decreased estimate in potential loss despite growth in building inventory? According to FEMA, two primary factors were responsible: (1) a slight decrease in estimated earthquake hazard in the western United States (namely California) except for some parts of Washington and Utah, and (2) a change in the distribution of building inventory in California, with an increase in wood frame buildings of 17% and a reduction in the amount of masonry (-6%), steel (-5.8%), and concrete (-3%) buildings in the state.19 Wood frame buildings are less vulnerable to earthquake damage, generally, compared to other construction types. Because California accounts for 66% of the overall nationwide annualized earthquake loss, a 17% increase in wood frame buildings had a proportionally large effect. In fact, FEMA attributed 78% of the loss reduction between 2001 and 2008 to the change in building inventory distribution, and 22% to the decrease in earthquake hazard for California.20 A single large earthquake can cause far more damage than the average annual estimate. Annualized estimates, however, help provide comparisons of infrequent, high impact events like damaging earthquakes, with more frequently occurring hazards like floods, hurricanes, or other types of severe weather. The annualized earthquake loss values shown in Table 6 represent future estimates, and are calculated by multiplying losses from potential future ground motions by their respective frequencies of occurrence, and then summing these values. 21 Table 6 also shows that annualized earthquake losses in the New York-Northern New JerseyLong Island metropolitan area are $30 million (ranked 21 out of 43 metropolitan areas with losses greater than $1 million per year), even though no destructive earthquakes have struck that area for generations. 22 This area has a relatively low seismic hazard, but also has an extensive 18 Ibid., p. 32. FEMA 366, p. 32 and p. 36. 20 Ibid., p. 36. 21 Ibid., p. 10. 22 The largest earthquakes in New York, New Jersey, and Massachusetts were, respectively: 1944, Massena, NY, magnitude 5.8, felt from Canada south to Maryland; 1783, New Jersey, magnitude 5.3, felt from New Hampshire to Pennsylvania; and 1755, Cape Ann and Boston, MA, intensity of VIII on the Modified Mercalli Scale, felt from Nova Scotia to Chesapeake Bay (USGS Earthquake Hazards Program). 19 Congressional Research Service 15 Earthquakes: Risk, Detection, Warning, and Research infrastructure and is densely populated. That combination of seismic risk, extensive infrastructure, and dense population produces a significant risk to people and structures, according to some estimates.23 In the absence of any significant or damaging earthquakes for that region in recent memory, however, the actual risk may be difficult to grasp intuitively. The New Madrid Seismic Zone The New Madrid Seismic Zone in the central United States is vulnerable to large but infrequent earthquakes. A series of large (magnitude greater than 7.0) earthquakes struck the Mississippi Valley over the winter of 1811-1812, centered close to the town of New Madrid, MO. Some of the tremors were felt as far away as Charleston, SC, and Washington, DC. The mechanism for the earthquakes in the New Madrid zone is poorly understood,24 and no earthquakes of comparable magnitude have occurred in the area since these events. Estimating earthquake damage is not an exact science and depends on many factors. As described above, these are primarily the probability of ground motion occurring in a particular area (see Figure 2), and the consequences of that ground motion, which are largely a function of building construction type and quality, and of the level of ground motion and shaking during the actual event. Such factors contribute to the difficulty of making a reasonable damage estimate for a lowfrequency, high-impact event in the New Madrid region based on the probability of an earthquake of similar magnitude occurring. This uncertainty has implications for policy decisions to ameliorate risk, such as setting building codes, and for designing and building structures to withstand a level of shaking commensurate with the risk. Developers of building codes tend to err on the side of caution; presumably the same seismic hazard should lead to similar building codes in urban areas (e.g., compare the seismic hazard for the New Madrid area with parts of California shown in Figure 2). Some researchers have questioned whether erring on the side of caution in the New Madrid Seismic Zone is justified. 25 These researchers question whether the benefits of building structures to conform with the earthquake probability estimates merit the costs, in light of the uncertainty in making those probability estimates. 26 These analyses may call into question whether the probability of ground motion estimates for the New Madrid Seismic Zone (the bulls-eye-shaped area shown in Figure 2 that includes parts of Arkansas, Illinois, Tennessee, and Missouri), and other regions of the country that experience infrequent earthquakes, are too high.27 An uncertainty analysis of the seismic hazard in the New Madrid Seismic Zone is beyond the scope of this report. 23 USGS Circular 1188, Table 3. In contrast to California, where earthquakes occur on the active margin of the North American tectonic plate, the New Madrid seismic zone is not on a plate boundary but may be related to old faults in the interior of the plate, marking a zone of tectonic weakness. 25 Andrew Newman et al., “Slow Deformation and Lower Seismic Hazard in the New Madrid Seismic Zone,” Science, v. 284 (April 23, 1999), pp. 619-621. 26 Seth Stein, Joseph Tomasello, and Andrew Newman, “Should Memphis Build for California’s Earthquakes?” Eos, v. 84, no. 19, (May 13, 2003), pp. 177, 184-185. 27 Seth Stein, “Code Red: Earthquake Imminent?” Earth, vol. 54, no. 1 (January 2009), pp. 52-59. 24 Congressional Research Service 16 Earthquakes: Risk, Detection, Warning, and Research Monitoring Congress authorized the USGS to monitor seismic activity in the United States in the 1990 reauthorization of the National Earthquake Hazards Reduction Act (P.L. 101-614). The USGS operates a nationwide network of seismographic stations called the Advanced National Seismic System (ANSS), which includes the National Strong-Motion Project (NSMP). Globally, the USGS and the Incorporated Research Institutions for Seismology (IRIS) operate 140 seismic stations of the Global Seismic Network (GSN) in more than 80 countries. The GSN provides worldwide coverage of earthquakes, including reporting and research.28 Advanced National Seismic System (ANSS) According to the USGS, “the mission of ANSS is to provide accurate and timely data and information products for seismic events, including their effects on buildings and structures, employing modern monitoring methods and technologies.”29 If fully implemented, ANSS would encompass more than 7,000 earthquake sensor systems covering parts of the nation vulnerable to earthquake hazards. As envisioned, the system would consist of dense urban networks, regional networks, and backbone stations. ANSS Funding Congress first authorized the ANSS program in P.L. 106-503 at a level of $38 million for FY2002 and $44 million for FY2003. The 2004 reauthorization of NEHRP (P.L. 108-360) authorized $30 million for ANSS in FY2005 and then $36 million per year through FY2009. The USGS spent $1.6 million in FY2000 and $3.6 million in FY2001 on ANSS-directed funding, but expenditures have never reached authorized levels since Congress first authorized appropriations for ANSS. From FY2005 through FY2009, the USGS spent a total of approximately $42.5 million within its Earthquake Hazards Program on ANSS, or approximately 24% of the total authorized levels over the five-year period since NEHRP was last reauthorized.30 The FY2010 budget request stated that the USGS plans to install a cumulative total of 822 ANSS monitoring stations by the end of 2009. That would represent approximately 12% of the 7,000 seismic stations originally envisioned for the program. According to its budget justification, the USGS plans to devote its ANSS-directed resources to operating and maintaining the installed system. 31 Of the approximately $8.8 million for ANSS-directed funding in FY2009, about $1.5 million was devoted to the development, modernization, and expansion of the system; the remainder of FY2009 funding was used to operate the existing system. 32 However, the American Recovery and Reinvestment Act (ARRA, P.L. 111-5) provided approximately $19 million in ANSS-directed funding to be spent over FY2009-FY2011.33 All of the ARRA funding would be 28 The GSN also monitors nuclear explosions. USGS Earthquake Hazards Program, at 30 USGS FY2010 Budget Justification, at FY2010_USGS_Greenbook.pdf, p. I-10. 31 Ibid., p. I-11. 32 Email from William Leith, Advanced National Seismic System Coordinator, USGS, December 22, 2009. 33 USGS FY2010 Budget Justification, pp. T-32 and T-33. 29 Congressional Research Service 17 Earthquakes: Risk, Detection, Warning, and Research provided for modernization and expansion of the current system, and when supplemented by base program funds over the three-year period, would double the number of ANSS stations.34 A doubling of the number of current stations would total approximately 23% of the 7,000 stations originally planned for ANSS. Dense Urban Networks In the original conception for ANSS, approximately 6,000 of the planned stations would be installed in 26 high-risk urban areas to monitor strong ground shaking and how buildings and other structures respond. Currently, five high-risk urban areas have instruments deployed in sufficient density to generate the data to produce near real-time maps, called ShakeMaps, which can be used in emergency response during and after an earthquake.35 (See “ShakeMap,” below.) Regional Networks Approximately 1,000 new instruments would replace aging and obsolete stations in the networks that now monitor the nation’s most seismically active regions. The current regional networks contain a mix of modern, digital, broadband, and high-resolution instruments that can provide real-time data; they are supplemented by older instruments that may require manual downloading of data. Universities in the region typically operate the regional networks and will likely continue to do so as ANSS is implemented. Backbone Stations Approximately 100 instruments comprise the existing “backbone” of ANSS, with a roughly uniform distribution across the United States, including Alaska and Hawaii. These instruments provide a broad and uniform minimum threshold of coverage across the country. The backbone network consists of USGS-deployed instruments and other instruments that serve both ANSS and the EarthScope project (described below, under “National Science Foundation”). National Strong-Motion Project (NSMP) Under ANSS, the USGS operates the NSMP to record seismic data from damaging earthquakes in the United States on the ground and in buildings and other structures in densely urbanized areas. The program currently has 900 strong-motion36 instruments in 701 permanent stations across the United States and in the Caribbean. The NSMP has three components: data acquisition, data management, and research. The near real-time measurements collected by the NSMP are used by other government agencies for emergency response and real-time warnings. If fully implemented, the ANSS program would deploy about 3,000 strong-motion instruments. Many of the current NSMP instruments are older designs and are being upgraded with modern seismometers. 34 USGS FY2010 Budget Justification, pp. T-32 and T-33. The number of stations necessary to generate a data-based ShakeMap depends on the urban area and geology, but roughly correspond to about half the number of planned stations per urban area, at a spacing of about 20 kilometers between stations. Personal communication, William Leith, Advanced National Seismic System Coordinator, USGS, January 11, 2010. 36 Strong motion seismometers, or accelerometers, are special sensors that measure the acceleration of the ground during large (>6.0 magnitude) earthquakes. 35 Congressional Research Service 18 Earthquakes: Risk, Detection, Warning, and Research Global Seismic Network (GSN) The GSN is a system of broadband digital seismographs arrayed around the globe and designed to collect high-quality data that are readily accessible to users worldwide, typically via computer. Currently, 140 stations have been installed in 80 countries and the system is nearly complete, although in some regions the spacing and location of stations has not fully met the original goal of uniform spacing of approximately 2,000 kilometers. The system is currently providing data to the United States and other countries and institutions for earthquake reporting and research, as well as for monitoring nuclear explosions to assess compliance with the Comprehensive Test Ban Treaty. Funding for the GSN totaled approximately $9 million in FY2009.37 The Incorporated Research Institutions for Seismology (IRIS)38 coordinates the GSN and manages and makes available the large amounts of data that are generated from the network. The actual network of seismographs is organized into two main components, each managed separately. The USGS operates two-thirds of the stations from its Albuquerque Seismological Laboratory, and the University of California-San Diego manages the other third via its Project IDA (International Deployment of Accelerometers). Other universities and affiliated agencies and institutions operate a small number of additional stations. IRIS, with funding from the NSF, supports all of the stations not funded through the USGS appropriations. Detection, Notification, and Warning Unlike other natural hazards, such as hurricanes, where predicting the location and timing of landfall is becoming increasingly accurate, the scientific understanding of earthquakes does not yet allow for precise earthquake prediction. Instead, notification and warning typically involves communicating the location and magnitude of an earthquake as soon as possible after the event to emergency response providers and others who need the information. Some probabilistic earthquake forecasts are now available that give, for example, a 24-hour probability of earthquake aftershocks for a particular region, such as California. These forecasts are not predictions, and are currently intended to increase public awareness of the seismic hazard, improve emergency response, and increase scientific understanding of the short-term hazard.39 In the California example, a time-dependent map is created and updated every hour by a system that considers all earthquakes, large and small, detected by the California Integrated Seismic Network, 40 and calculates a probability that each earthquake will be followed by an aftershock41 that can cause strong shaking. The probabilities are calculated from known behavior of aftershocks and the possible shaking pattern based on historical data. 37 $9 million reflects the combined appropriations for USGS and NSF for FY2009. ARRA (P.L. 111-5) enabled approximately $10 million to be made available via the USGS and NSF for the GSN through FY2010. Email from William Leith, Advanced National Seismic System Coordinator, USGS, Dec. 21, 2009. 38 IRIS is a university research consortium, primarily funded by NSF, that collects and distributes seismographic data. 39 USGS Open-File Report 2004-1390, and California 24-hour Aftershock Forecast Map, at 40 The California Integrated Seismic Network is the California region of ANSS; see 41 Earthquakes typically occur in clusters, in which the earthquake with the largest magnitude is called the main shock, events before the main shock are called foreshocks, and those after are called aftershocks. See also Congressional Research Service 19 Earthquakes: Risk, Detection, Warning, and Research When a destructive earthquake occurs in the United States or in other countries, the first reports of its location, or epicenter,42 and magnitude originate either from the National Earthquake Information Center (NEIC), or from one of the regional seismic networks that are part of ANSS. Other organizations, such as universities, consortia, and individual seismologists may also contribute information about the earthquake after the event. Products such as ShakeMap are assembled as rapidly as possible to assist in emergency response and damage estimation following a destructive earthquake. National Earthquake Information Center (NEIC) The NEIC, part of the USGS, is located in Golden, CO. Originally established as part of the National Ocean Survey (U.S. Department of Commerce) in 1966, the NEIC was made part of the USGS in 1973. With data gathered from the networks described above and from other sources, the NEIC determines the location and size of all destructive earthquakes that occur worldwide and disseminates the information to the appropriate national or international agencies, government public information channels, news media, scientists and scientific groups, and the general public. With the advent of the USGS Earthquake Notification Service (ENS), notifications of earthquakes detected by the ANSS/NEIC are provided free to interested parties. Users of the service can specify the regions of interest, establish notification thresholds of earthquake magnitude, designate whether they wish to receive notification of aftershocks, and even set different magnitude thresholds for daytime or nighttime to trigger a notification. The NEIC has long-standing agreements with key emergency response groups, federal, state, and local authorities, and other key organizations in earthquake-prone regions who receive automated alerts—typically location and magnitude of an earthquake—within a few minutes of an event in the United States. The NEIC sends these preliminary alerts by email and pager immediately after an earthquake’s magnitude and epicenter are automatically determined by computer.43 This initial determination is then checked by around-the-clock staff who confirm and update the magnitude and location data.44 After the confirmation, a second set of notifications and confirmations are triggered to key recipients by email, pager, fax, and telephone. For earthquakes outside the United States, the NEIC notifies the State Department Operations Center, and often sends alerts directly to staff at American embassies and consulates in the affected countries, to the International Red Cross, the U.N. Department of Humanitarian Affairs, and other recipients who have made arrangements to receive alerts. ShakeMap Traditionally, the information commonly available following a destructive earthquake has been epicenter and magnitude, as in the data provided by the NEIC described above. Those two 42 The epicenter of an earthquake is the point on the earth’s surface directly above the hypocenter. The hypocenter is the location beneath the earth’s surface where the fault rupture begins. 43 Stuart Simkin, NEIC, Golden, CO, telephone conversation, Nov. 4, 2006. 44 In early 2006, the NEIC implemented an around-the-clock operation center and seismic event processing center in response to the Indonesian earthquake and resulting tsunami of December 2004. Funding to implement 24/7 operations was provided by P.L. 109-13. Congressional Research Service 20 Earthquakes: Risk, Detection, Warning, and Research parameters by themselves, however, do not always indicate the intensity of shaking and extent of damage following a major earthquake. Recently, the USGS developed a product called ShakeMap that provides a nearly real-time map of ground motion and shaking intensity following an earthquake in areas of the United States where the ShakeMap system is in place. Figure 4 shows an example of a ShakeMap. Figure 4. Example of a ShakeMap Source: USGS, Note: Earthquake occurred 23.1 miles west-northwest of Ferndale, CA, at 4:27 p.m. on January 9, 2010, with a magnitude of 6.5. The star indicates the epicenter of the earthquake. Viewed on January 12, 2010. According to preliminary news reports, some damage estimates are nearly $30 million for the town of Eureka itself from the earthquake. (See Congressional Research Service 21 Earthquakes: Risk, Detection, Warning, and Research The maps produced portray the extent of damaging shaking and can be used by emergency response and for estimating loss following a major earthquake. Currently, ShakeMaps are available for northern California, southern California, the Pacific Northwest, Nevada, Utah, Hawaii, and Alaska.45 With improvements to the regional seismographic networks in the areas where ShakeMap is available, new real-time telemetry from the region, and advances in digital communication and computation, ShakeMaps are now triggered automatically and made available within minutes of the event via the Web. In addition, better maps are now available because of recent improvements in understanding the relationship between the ground motions recorded during the earthquake and the intensity of resulting damage. If databases containing inventories of buildings and lifelines are available, they can be combined with shaking intensity data to produce maps of estimated damage. The ShakeMaps have limitations, especially during the first few minutes following an earthquake before more data arrive from distributed sources. Because they are generated automatically, the initial maps are preliminary, and may not have been reviewed by experts when first made available. They are considered a work in progress, but are deemed to be very promising, especially as more modern seismic instruments are added to the regional networks under ANSS and computational and telecommunication abilities improve. Prompt Assessment of Global Earthquakes for Response (PAGER) Another USGS product that is designed to provide nearly real-time earthquake information to emergency responders, government agencies, and the media is the Prompt Assessment of Global Earthquakes for Response, or PAGER, system. 46 This automated system rapidly assesses the number of people, cities, and regions exposed to severe shaking by an earthquake, and generally makes results available within 30 minutes. Following the determination of earthquake location and magnitude, the PAGER system calculates the degree of ground shaking using the methodology developed for ShakeMap, estimates the number of people exposed to various levels of shaking, and produces a description of the vulnerability of the exposed population and infrastructure. The vulnerability includes potential for earthquake-triggered landslides, which could be devastating, as was the case for the huge May 12, 2008, earthquake in Sichuan, China. The automated and rapid reports produced by the PAGER system provide an advantage compared to the traditional accounts from eye-witnesses on the ground or media reports, because communications networks may have been disabled from the earthquake. Emergency responders, relief organizations, and government agencies could make plans based on PAGER system reports even before getting “ground-truth” information from eye-witnesses and the media.47 Figure 5 shows an example of PAGER output for the January 12, 2010, magnitude 7.0 earthquake in Haiti. 45 ShakeMaps for some areas outside the United States are also available. See shakemap/. 46 See the USGS Earthquakes Hazards Program for more information, at 47 See also USGS Fact Sheet 2007-3101 at Congressional Research Service 22 Earthquakes: Risk, Detection, Warning, and Research Figure 5. Example of PAGER Output for the January 12, 2010, Magnitude 7.0 Haiti Earthquake Source: USGS, Note: This is version 7 of the PAGER output, accessed on January 14, 2010. Congressional Research Service 23 Earthquakes: Risk, Detection, Warning, and Research Pre-disaster Planning: HAZUS-MH FEMA developed a methodology and software program called the Hazards U.S. Multi-Hazard (HAZUS-MH). 48 The program allows a user to estimate losses from damaging earthquakes, hurricane winds, and floods before a disaster occurs. The pre-disaster estimates could provide a basis for developing mitigation plans and policies, preparing for emergencies, and planning response and recovery. HAZUS-MH combines existing scientific knowledge about earthquakes (for example, ShakeMaps, described above), engineering information that includes data on how structures respond to shaking, and geographic information system (GIS) software to produce maps and display hazards data including economic loss estimates. The loss estimates produced by HAZUS-MH include • • • physical damage to residential and commercial buildings, schools, critical facilities, and infrastructure; economic loss, including lost jobs, business interruptions, repair and reconstruction costs; and social impacts, including estimates of shelter requirements, displaced households, and number of people exposed to the disaster. In addition to furnishing information as part of earthquake mitigation efforts, HAZUS-MH can also be used to support real-time emergency response activities by state and federal agencies after a disaster. Twenty-seven HAZUS-MH user groups—cooperative ventures among private, public, and academic organizations that use the HAZUS-MH software—have formed across the United States to help foster better-informed risk management for earthquakes and other natural hazards.49 HAZUS-MH software was first released to the public in 1997 and the first user group, the Bay Area HAZUS-MH User Group, was formed the same year. Research—Understanding Earthquakes U.S. Geological Survey Under NEHRP, the USGS has responsibility for conducting targeted research into improving the basic scientific understanding of earthquake processes. The current earthquake research program at the USGS covers six broad categories:50 • Borehole geophysics and rock mechanics: studies to understand heat flow, stress, fluid pressure, and the mechanical behavior of fault-zone materials at seismogenic51 depths to yield improved models of the earthquake cycle; • Crustal deformation: studies of the distortion or deformation of the earth’s surface near active faults as a result of the motion of tectonic plates; 48 See See 50 See 51 Seismogenic means capable of generating earthquakes. 49 Congressional Research Service 24 Earthquakes: Risk, Detection, Warning, and Research • Earthquake geology and paleoseismology: studies of the history, effects, and mechanics of earthquakes; • Earthquake hazards: studies of where, why, when, and how earthquakes occur; • Regional and whole-earth structure: studies using seismic waves from earthquakes and man-made sources to determine the structure of the planet ranging from the local scale, to the whole crust, mantle, and even the earth’s core; and • Strong-motion seismology, site response, and ground motion: studies of largeamplitude ground motions and the response of engineered structures to those motions using accelerometers. National Science Foundation NSF supports fundamental research into understanding the earth’s dynamic crust. Through its Earth Sciences Division (part of the Geosciences Directorate), NSF distributes research grants and coordinates programs investigating the crustal processes that lead to earthquakes around the globe.52 In 2003, NSF initiated a Major Research Equipment and Facilities Construction (MREFC) project called EarthScope that deploys instruments across the United States to study the structure and evolution of the North American Continent, and to investigate the physical processes that cause earthquakes and volcanic eruptions.53 EarthScope is a multi-year project begun in 2003 that is funded by NSF and conducted in partnership with the USGS and NASA. EarthScope instruments are intended to form a framework for broad, integrated studies of the four-dimensional (three spatial dimensions, plus time) structure of North America. The project is divided into three main programs: • The San Andreas Fault Observatory at Depth (SAFOD), a deep borehole observatory drilled through the San Andreas fault zone close to the hypocenter of the 1966 Parkfield, CA, magnitude 6 earthquake; • The Plate Boundary Observatory (PBO), a system of GPS arrays and strainmeters54 that measure the active boundary zone between the Pacific and North American tectonic plates in the western United States; and • USArray, 400 transportable seismometers that will be deployed systematically across the United States on a uniform grid to provide a complete image of North America from continuous seismic measurements. Through its Engineering Directorate, NSF funds the George E. Brown Jr. Network for Earthquake Engineering Simulation (NEES), a project intended to operate until 2014, aimed at understanding the effects of earthquakes on structures and materials.55 To achieve the program’s goal, the NEES 52 See See 54 A strainmeter is a tool used by seismologists to measure the motion of one point relative to another. 53 55 A non-profit NEES consortium (NEESinc.) has operated the facilities for the 10-year operating lifespan at the following institutions: Cornell University; Lehigh University; Oregon State University; Rensselaer Polytechnical Institute; University of Buffalo-State University of New York; University of California-Berkeley; University of California-Davis; University of California-Los Angeles; University of California-San Diego; University of California(continued...) Congressional Research Service 25 Earthquakes: Risk, Detection, Warning, and Research facilities conduct experiments and computer simulations of how buildings, bridges, utilities, coastal regions, and materials behave during an earthquake. Conclusion At present earthquakes can be neither accurately predicted nor prevented, and in its 1990 reauthorization NEHRP shifted its program emphasis from prediction to hazard reduction. The program’s focus has been on understanding the earthquake hazard and its risk to populations and infrastructure in the United States, developing effective measures to reduce earthquake hazards, and promoting the adoption of earthquake hazards reduction measures in vulnerable areas. In the 111th Congress, legislation introduced to reauthorize NEHRP, H.R. 3820, reemphasizes that approach but casts it in terms of hazard mitigation by stating that a major goal for the program should be “to reduce the loss of life and damage to communities and infrastructure through increasing the adoption of hazard mitigation measures.” The bill further emphasizes the social aspects of mitigating earthquake hazards, calling for research to better understand institutional, social, behavioral, and economic factors that influence how risk mitigation is implemented, in addition to the traditional research into understanding how, why, and where earthquakes occur. The emphasis on mitigation presented by H.R. 3820 reflects at least two fundamental challenges to increasing the nation’s resiliency to earthquakes, and to most other major natural hazards such as hurricanes and major floods. The first is to assess whether social, behavioral, and economic factors can be understood in sufficient degree to devise strategies that influence behavior to mitigate risk posed by the hazard. Put simply, what motivates people and communities to adopt risk mitigation measures that address the potential hazard? A second challenge, which is more squarely an issue for Congress, is how to measure the effectiveness of NEHRP more quantitatively. It is inherently difficult to capture precisely, in terms of dollars saved or fatalities prevented, the effectiveness of mitigation measures taken before an earthquake occurs. A major earthquake in a populated urban area within the United States would cause damage, and a question becomes how much damage would be prevented by mitigation strategies underpinned by the NEHRP program. A precise relationship between earthquake mitigation measures, NEHRP and other federal earthquake-related activities, and reduced losses from an actual earthquake may never be possible. However, as more accurate seismic hazard maps evolve, as understanding of the relationship between ground motion and building safety improves, and as new tools for issuing warnings and alerts such as ShakeMap and PAGER are devised, trends denoting the effectiveness of mitigation strategies and NEHRP activities may emerge more clearly. Without an ability to precisely predict earthquakes, Congress is likely to face an ongoing challenge in determining the most effective federal approach to increasing the nation’s resilience to low-probability but highimpact natural hazards, such as major earthquakes. (...continued) Santa Barbara; University of Colorado-Boulder; University of Illinois at Urbana-Champaign; University of Minnesota; University of Nevada-Reno; and University of Texas at Austin. See Congressional Research Service 26 Earthquakes: Risk, Detection, Warning, and Research Author Contact Information Peter Folger Specialist in Energy and Natural Resources Policy, 7-1517 Congressional Research Service 27