Welcome to the Extreme Weather Sourcebook
Economic & Other Societal Impacts Related to Hurricanes, Floods, Tornadoes, Lightning, & Other Weather Phenomena

Data Sources and Methodology

Overview

Extreme weather accounts for many billions of dollars in damage every year in the United States. The size of the U.S., along with its particular geographical and societal conditions, make it susceptible to a wide range of weather phenomena. This website presents a summary of damage suffered from hurricanes, floods, and tornadoes in the United States and its territories.

The goal of the website is to educate and stimulate interest in the societal impacts of weather in the United States. To compare damage from different phenomena, recorded over different time intervals, is a challenging methodological task. The data presented in this website come from a variety of sources using a range of methodologies. The data have been adjusted in various ways (described below) to allow for an apples-to-apples comparison of losses between events and over time. Consequently, extreme caution should be exercised when using this data. It is strongly recommended that analyses requiring precise data -- such as trend analysis or investigation of particular events - use the Sourcebook as a starting point to further, more rigorous research. In addition, the information presented here should be understood as historical, not predictive.

Tornado Damage Figures

Data Source

The 1950 - 2006 tornado damage figures for the 50 states, District of Columbia, Puerto Rico, and the U.S. Virgin Islands were compiled from the Storm Predition Center's archived data. The Storm Predition Center's archived data is entirely derived from the National Weather Service Performance Branch's database, which compiles storm events data originally collected by each of the National Weather Service Weather Forecast Offices. Some loss values from the Storm Predition Center may exclude damages to crops and agriculture.

Adjustments

Three different adjustments have been made to the tornado damage figures.

First, all tornado damages recorded between 1950 and 1994 is reported by the Storm Predition Center's tornado dataset as one of nine classes (Class 1 = $50; Class 2 = $50 - 500; Class 3 = $500 - 5K; Class 4 = $5K - 50K; Class 5 = $50K - 500K; Class 6 = $500K - 5M; Class 7 = $5M - 50M; Class 8 = $50M - 500M; Class 9 = $500M - 5B). We have adjusted the 1950 - 1994 damage figures to the geometric mean of each damage class. We chose the geometric mean because it was the computation standard used in earlier editions of the Extreme Weather Sourcebook.

Second, all damages have been adjusted for inflation to 2006 dollars using Implicit Price Deflators for Gross Domestic Product as reported by the Bureau of Economic Analysis. This measure was used to maintain consistency with the inflation adjustments made on other statistics presented in the Extreme Weather Sourcebook.

Third, the "Total Wealth Adjusted Damages" column reports tornado damage totals which have been adjusted for changes in the nation's assets. We used Current-Cost Net Stock of Fixed Assets and Consumer Durable Goods as reported by the Bureau of Economic Analysis as a measure of wealth according to the methodology presented in Pielke et al. (2008)¹.

Footnotes

¹Compare Brooks, H.E. and C.A. Doswell III, 2000: Normalized damage from major tornadoes in the United States: 1890-1999. Weather and Forecasting (submitted September 2000). Available at http://www.nssl.noaa.gov/~brooks/damage/tdam1.html.

Flood Damage Figures

Data Source

For the years 1955-2003, flood loss data for all states was taken from Flood Damage in the United States: A Reanalysis of National Weather Service Estimate. This database was created by Mary Downton and Roger Pielke Jr. through reanalyzing the National Weather Service Hydrologic Information Center’s flood loss database.
For the years 2004-2006 flood loss data for all states was taken directly from the National Weather Service Hydrologic Information Center’s flood loss data set.

Adjustments

Two different adjustments have been made to the flood damage figures.

First, all damages have been adjusted for inflation to 2006 dollars using Implicit Price Deflators for Gross Domestic Product as reported by the Bureau of Economic Analysis. This measure was used to maintain consistency with the inflation adjustments made on other statistics presented in the Extreme Weather Sourcebook.

Second, the "Total Wealth Adjusted Damages" column reports tornado damage totals which have been adjusted for changes in the nation's assets. We used Current-Cost Net Stock of Fixed Assets and Consumer Durable Goods as reported by the Bureau of Economic Analysis as a measure of wealth according to the methodology presented in Pielke et al. (2008)¹.

Hurricane Damage Figures

Data Source

Hurricane loss data for the years 1900-2005 was taken from the data set for Roger Pielke Jr.’s paper Normalized Hurricane Damage in the United States: 1900–2005. Both the data and the paper can be found here. 

This data set only documents the years up to 2005, so for the year 2006, data was taken directly from the National Hurricane Center’s Tropical Cyclone Report, which is believed to be the original data source of Pielke's data set for the years 1998-2005.  The NHC Tropical Cyclone Reports can be found here.

Since the late 1980's the National Hurricane Center (NHC) bases their loss estimates off of an insurance claims database run by Property Claims Services (PCS). The NHC does not use the received insured loss estimate directly, but instead doubles the PCS figure in an attempt to consider uninsured losses, flood losses, and other losses not typically included in an insurance claim. Since PCS has a $25 million threshold to be considered and included in their database, hurricane/tropical storm events with small dollar damage amounts may not have damages mentioned in the NHC Tropical Cyclone Reports. Therefore, these events will also not have damages listed in the Extreme Weather Sourcebook totals.

Since Pielke’s data set does not provide individual state loss assignments for multi-state hurricanes, some decisions were necessary to divide hurricane losses to individual states. The decisions process used for individual state assignments is described below:

For multi-state hurricanes which occurred during the years 1900-1954, the Monthly Weather Review’s (MWR) Annual Hurricane Summaries, which are believed to be the original sources for Pielke's hurricane data set, were referenced to aid in assigning losses to individual states. If the loss total in Pielke’s data set matched the MWR total, or differed by only a slight amount, and the MWR suggested a state assignment, then this state assignment was used. Otherwise, the loss was equally divided and equal parts were all assigned to states.

For multi-state hurricanes which occurred during the years 1955-2006, the MWR was used to determine state damage assignments. If the loss total in Pielke’s data set matched the MWR total, or differed by only a slight amount, and the MWR suggested a state assignment, then this state assignment was used. In the case that the MWR does not provide individual state loss assignments, Property Claims Services' catastrophe claims database was used to determine a ratio of state damages to total damages. Then the hurricane loss total listed in Pielke's data set was distributed to affected states according to this ratio. For three events, Claudette (1979), David (1979), and Chris (1982), no consistent state assignments could be determined. Therefore, the loss total provided in Pielke's data set was evenly distributed among the states also listed there.

Only damage from storms classified as hurricanes or tropical storms by the National Weather Service-National Hurricane Center has been included. These figures focus on damage due to wind and/or storm surge and usually exclude damage from flooding.

Adjustments

The hurricane damage data have been adjusted for inflation and wealth using the same methodology as was used with flood damages.

Footnotes

³Storms reviewed: Hurricanes Bertha, Fran, Danny, Bonnie, Earl, Georges, Mitch, Brett, Dennis, Floyd, Irene, Jose, and Lenny; Tropical Storms Josephine, Charley, Frances, and Harvey.

⁴Herbert, P.J., J.D. Jarrell, and M. Mayfield, 1996: Report No. NOAA-TM-NWS-TPC-1, Table 14.

U.S. Composite Damage Figures

Data Source

Damage values used for the composite total were directly taken from the data used for their individual Extreme Weather Sourcebook sections. For the specific data sources used to determine these, please see the “Hurricane Damage Figures”, "Flood Damage Figures" and "Tornado Damage Figures" paragraphs in the "Data Sources and Methodology" section

It should also be noted that there is no flood loss data for the years 1980-1982. Reliable data for these years do not exist due to a lack of funding for collecting flood loss data during this time period.

Adjustments

The composite damage values presented in the "Total Damages" column were adjusted for inflation only. The "Total Wealth Adjusted Damages" column presents damage values adjusted for both inflation and changes in wealth. Both the inflation and wealth adjustments were computed using the same methodology as was used for the hurricane, flood, and tornado sections. Any additional adjustments that were made to the tornado damage data also apply here since the same data was used for this section.

Hail, Thunderstorms, Heavy Rainfall Events, Winter Storms, Lightning, Wind Storms, Federal Disaster Relief for Weather-Caused Events

Data Source1950 - 1997 national data provided courtesy of Stanley Changnon, Climatologist, 801 Buckthorn Circle, Mahomet, Illinois, 61853. Adjustments

Where applicable, the data have been adjusted for inflation to 1997 values or as described in the figure.

Lightning Fatalities, Injuries, Casualties, and Damage Reports by State 1959 - 1994

Data SourceCurran, E.B., R.L. Holle, and R.E. Lopez, 2000: Lightning casualties and damages in the United States from 1959 to 1994. J. of Climate 13, 3448 - 3464, Tables 3 and 4."Damage reports" refers to every instance in which damage from lightning was reported, regardless of the amount of damage.Adjustments

None

For Further Reading

Barnes, J., 1998: Florida's Hurricane History. University of North Carolina Press, Chapel Hill and London. 330 pp.

Barnes, J., 1995: North Carolina's Hurricane History. University of North Carolina Press, Chapel Hill and London. 206 pp.

Brooks, H.E., and C.A. Doswell III, 2000: Normalized damage from major tornadoes in the United States: 1890-1999. Weather and Forecasting (submitted September 2000).

Changnon, D., and S.A. Changnon, 1998: Evaluation of catastrophe data for use in climate change investigations. Climatic Change, 38, 453-445.

Changnon, S.A., 1982: Users beware: The upward trends in tornado frequencies. Weatherwise, 35, 64-69.

E.B. Curran, R.L. Holle, and R.E. Lopez, 2000: Lightning casualties and damages in the United States from 1959 to 1994. J. of Climate, 13, 3448 - 3464.

Grazulis, T.P., 1993: Significant Tornadoes 1680 - 1991: A chronology and analysis of events. The Tornado Project of Environmental Films, St. Johnsbury, Vermont, 1326 pp.

Grazulis, T.P., 1997: Significant Tornadoes Update 1992 - 1995. The Tornado Project of Environmental Films, St. Johnsbury, Vermont, 1444 pp. (continued from original version).

The H. John Heinz III Center for Science, Economics and the Environment, 1999: The Hidden Costs of Coastal Hazards: Implications for Risk Assessment and Mitigation. Island Press, Washington, D.C. 220 pp.

Kithil, R.: Results of investigations into annual U.S.A. lightning costs and losses. National Lightning Safety Institute.

Kunkel, K., R. A. Pielke Jr., S. A. Changnon, 1999: Temporal fluctuations in weather and climate extremes that cause economic and human health impacts: A review. Bull. Amer. Meteor. Soc., 80(6), 1077-1098.

Longshore, D., 2000: Encyclopedia of Hurricanes, Typhoons, and Cyclones. Checkmark Books, N.Y., 372 pp.

Lopez, R.E., R.L. Holle, and T.A. Heitkamp, 1995: Lightning casualties and property damage in Colorado from 1950 to 1991 based on Storm Data. Weather and Forecasting, 10, 114-126.

National Research Council, Committee on Assessing the Costs of Natural Disasters, 1999: The Impacts of Natural Disasters: A Framework for Loss Estimation. National Academy Press, Washington, D.C. 80 pp.

Pielke, R. A., Jr., and M. Downton, 1999: U.S. Trends in streamflow and precipitation: Using societal impact data to address an apparent paradox. Bull. Amer. Meteor. Soc., 80, 1435-1436. Available for download at Pielke, R.A., Jr., and M. Downton, 2000: Precipitation and damaging floods: Trends in the United States, 1932-1997. J. of Climate, 13(20), 3625-3637.

Pielke, R. A., Jr., and C. W. Landsea, 1998: Normalized hurricane damages in the United States: 1925-1995. Weather and Forecasting, 13, 621-631.

Pielke, R.A., Jr., and C.W. Landsea, 1999: La Niña, El Niño, and Atlantic hurricane damages in the United States. Bull. Amer. Meteor. Soc., 80(10), 2027-2033.

Relevant Links

Extreme weather and climate events (NCDC)

Billion dollar US weather disasters (NCDC)

El Nino/La Nina (NCDC)

Global climate change (NCDC)

Hurricanes (includes maps and special reports on numerous hurricanes) (NCDC)

Heavy precipitation (includes maps--such as 24-hour maximums by state, special reports, etc) (NCDC)

Societal Aspects of Weather (ESIG)

Temperature extremes (includes maps--such as extreme maximum by state, special reports, etc) (NCDC)

Tornadoes (includes maps--such as long-term averages by state and normalized per 10,000 sq miles, special reports, etc) (NCDC)

Weather events (numerous reports on recent and historical events)(NCDC)