Death and Death Rates Due to

Extreme Weather Events

Global and U.S. Trends, 1900–2006

Indur M. Goklany
Death and Death Rates Due to About the Civil Society Coalition on
Extreme Weather Events Climate Change
By Indur M. Goklany The Civil Society Coalition on Climate Change seeks to
educate the public about the science and economics of
November 2007
climate change in an impartial manner. It was
International Policy Network established as a response to the many biased and
Third Floor, Bedford Chambers alarmist claims about human-induced climate change,
The Piazza which are being used to justify calls for intervention and
London WC2E 8HA UK regulation.
t: +4420 7836 0750
The Coalition comprises over forty independent civil
f: +4420 7836 0756
society organisations who share a commitment to
e: [email protected]
improving public understanding about a range of public
w: www.policynetwork.net
policy issues. All are non-profit organizations that are
Designed and typeset in Latin 725 by MacGuru Ltd independent of political parties and government.
[email protected]

First published by International Policy Press About the author

a division of International Policy Network
Dr. Indur M. Goklany has worked on environmental
© Indur M. Goklany 2007 and energy policy issues for over three decades in federal
and state governments, and the private sector. He has
All rights reserved. Without limiting the rights under
written over one hundred monographs, book chapters
copyright reserved above, no part of this publication may
and papers on topics ranging from climate change,
be reproduced, stored or introduced into a retrieval
human well-being, economic development, technological
system, or transmitted, in any form or by any means
change, and biotechnology to sustainable development.
(electronic, mechanical, photocopying, recording or
otherwise) without the prior written permission of both He has worked for the U.S. Department of the Interior,
the copyright owner and the publisher. which manages 20 percent of the U.S. land area, and
associated mineral, energy and water resources, the U.S.
Environmental Protection Agency and the State of
International Policy Network Michigan. He is the winner of the 2007 Julian Simon
International Policy Network (IPN) is a charity based in Prize. He was a visiting fellow with the American
the UK, and a non-profit (501c3) organisation in the US. Enterprise Institute, and the Julian Simon Fellow at the
It is a non-governmental, educational and non-partisan Property and Environment Research Center in Bozeman,
organization which relies on charitable donations from Montana. He has represented the U.S. at the
individuals, foundations and businesses to carry out its Intergovernmental Panel on Climate Change and in the
work. It accepts no money from government. negotiations leading to the UN Framework Convention
on Climate Change. His degrees, all in electrical
IPN aims to empower individuals and promote respect
engineering, are from the Indian Institute of Technology,
for people and property in order to eliminate poverty,
Bombay, and Michigan State University..
improve human health and protect the environment.
IPN promotes public awareness of the importance of this He is the author of Clearing the Air: The Real Story of the
vision for all people. War on Air Pollution, and The Precautionary Principle, and
The Improving the State of the World, all published by the
Cato Institute. Opinions and views expressed by Dr.
Goklany are his alone, and not necessarily of any
institution with which he is associated.
Death and Death Rates Due to Extreme
Weather Events
Global and U.S. Trends, 1900–2006

an average year, and seem to be declining in general.

Executive summary Equally important, mortality due to extreme weather
Despite the recent spate of deadly extreme weather events has declined despite an increase in all-cause
events – such as the 2003 European heat wave and the mortality, suggesting that humanity is adapting better to
2004 and 2005 hurricane seasons in the USA – aggregate extreme events than to other causes of mortality. In
mortality and mortality rates due to extreme weather summary, there is no signal in the mortality data to
events are generally lower today than they used to be. indicate increases in the overall frequencies or severities
of extreme weather events, despite large increases in the
Globally, mortality and mortality rates have declined by
population at risk.
95 percent or more since the 1920s. The largest
improvements came from declines in mortality due to
droughts and floods, which apparently were responsible Introduction
for 93 percent of all deaths caused by extreme events
Even prior to Hurricanes Katrina and Rita striking the
during the 20th Century. For windstorms, which, at 6
U.S. coast, there was speculation in medical and
percent, contributed most of the remaining fatalities,
scientific journals, the popular press and elsewhere that
mortality rates are also lower today but there are no
climate change would exacerbate extreme weather
clear trends for mortality. Cumulatively, the declines
events through, among other things, intensification of
more than compensated for increases due to the 2003
the hydrological cycle, hurricanes and other storms, and,
heat wave.
thereby, raise deaths and death rates globally as well as
With regard to the U.S., current mortality and mortality in the United States (see, e.g., IPCC 2001: Table SPM-1;
rates due to extreme temperatures, tornados, lightning, Patz 2004; MacMichael and Woodruff 2004; Schiermeier
floods and hurricanes are also below their peak levels of 2003, 2005; Trenberth and Taylor, no date).1
a few decades ago. The declines in annual mortality for
This speculation has been fueled by recent, somewhat
the last four categories range from 62 to 81 percent,
controversial, studies suggesting that hurricane
while mortality rates declined 75 to 95 percent.
intensities might be stronger because of climate change
If extreme weather has indeed become more extreme for (e.g., Webster et al. 2005; Emmanuel 2005a, 2005b;
whatever reason, global and U.S. declines in mortality Pielke, Jr., 2005; Landsea 2005). Moreover, the
and mortality rates are perhaps due to increases in occurrence of recent weather related disasters – ranging
societies’ collective adaptive capacities. This enhanced from the Central European floods of 2002, the 2003
adaptive capacity is associated with a variety of European heat wave, and the back-to-back disastrous
interrelated factors – greater wealth, increases in Atlantic hurricane seasons of 2004 and 2005 – has only
technological options, and greater access to and intensified suspicions. All of this was epitomized by a
availability of human and social capital – although luck recent Time magazine cover story warning all to worry
may have played a role. Because of these developments, more about global warming (Kluger 2006; see also, for
nowadays extreme weather events contribute less than example, Anderson and Bausch 2006).
0.06 percent to the global and U.S. mortality burdens in
Some have claimed that, all else being equal, climate

3
Death and Death Rates Due to Extreme Weather Events

change will increase the frequency or severity of the episodic nature of extreme events, such an
weather-related extreme events (see, e.g., IPCC 2001; examination should ideally be based on several decades’,
Patz 2004; MacMichael and Woodruff 2004). This study if not centuries’, worth of data. Any such examination
examines whether losses due to such events (as should, of course, be cognizant that adaptive capacity
measured by aggregate deaths and death rates2) have and exposure of human populations to risk also change
increased globally and for the United States in recent over time.
decades. It will also attempt to put these deaths and
In particular, one should examine mortality rates so as
death rates into perspective by comparing them with
to filter out the effect of population growth on the
the overall mortality burden, and briefly discuss what
population at risk. However, it may be argued that the
trends in these measures imply about human adaptive
use of mortality rates is inadequate to eliminate the
capacity.
effect of increases in populations at risk since, as the
Trends in deaths and death rates, while of intrinsic population becomes larger, people will migrate to riskier
interest for public policy purposes, may also have a and more vulnerable locations as the less vulnerable
bearing on trends in economic losses. Goklany (2000) locations are occupied. In addition, some state policies
speculates that while a wealthier society may invest may inappropriately create a “moral hazard” situation in
more in limiting loss of human lives, it may be less which individuals have incentives to bear less than their
concerned about property losses, even though a full burden of risk, effectively transferring portion of
wealthier society is also likely to have more property at their risk to other segments of society; this may place
risk. This suggestion finds some support in data which even wealthier populations at greater physical risk (in
show that the ratio of death-to-property-loss for addition to increasing financial risk; Goklany 2000).
tornadoes in the U.S. has declined in recent decades
Figure 1 displays data on aggregate global mortality and
(Doswell et al. 1999, Brooks and Doswell 2001).
mortality rates between 1900 and 2006 for the following
weather-related extreme events: droughts, extreme
Trends in mortality and mortality rates temperatures (both extreme heat and extreme cold),

Global trends, 1900–2006

In general, climate change could change the frequencies,

intensities and/or durations of extreme weather events Figure 1 Global death and death rates due to extreme
events, 1900–2006
such as floods, droughts, windstorms and extreme
temperatures – increasing them at some locations and 500 485 Deaths per year (in 1000s)
446 Death rates per year (per million)
for some periods, while decreasing them at other
locations and other periods. Some of the effects of these 400
370
changes will tend to offset each other and/or be
redistributed over space and time. 300
242
For instance, an increase in deaths due to heat waves at 209 211
one location might be compensated for by a decline in 200
168
157
deaths due to fewer or less intense cold waves at the 128

same or another location. Alternatively, climate change 100 79 71

66
might redistribute the temporal and spatial pattern of 48 54
33
25 14 14 14 22
6
rainfall, droughts and other such events. 0
3

1900– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000–
Accordingly, to estimate the net impact of climate 09 1919 1929 1939 1949 1959 1969 1979 1989 1999 2006

change on mortality (if any), it is probably best to Note that in figures 1 through 4, data for the last period are averaged over seven years worth
of data.
examine cumulative deaths at the global level Sources; EM-DAT (2007); McEvedy and Jones (1978); WRI (2005, 2007)

aggregated over all types of extreme events. Because of

4
 Death and Death Rates Due to Extreme Weather Events

Table 1 Global deaths and death rates for various types of events, 1900–1989 and 1990–2006

Deaths per year Death Rates per year

(per million people)

1900–1989 1990–2006 1900–1989 1990–2006

Droughts 130,042 185 57.99 0.03

Floods 75,212 7,637 31.95 1.29
Windstorms 10,856 13,650 3.96 2.45
Slides 469 868 0.16 0.15
Waves/Surges 128 207 0.06 0.03
Extreme Temperatures 110 5,671 0.03 0.91
Wild Fires 21 47 0.01 0.01
TOTAL 216,839 28,266 94.16 4.87

Sources: EM-DAT (2007); McEvedy and Jones (1978); WRI (2007)

floods, slides, waves and surges, wild fires and the 1900–1989 and 1990–2006 periods, while the
windstorms of different types (e.g., hurricanes, cyclones, annual mortality rate dropped by 95 percent.
tornados, typhoons, etc.).3,4 It indicates that both death
n Declines in mortality between the two periods were
and death rates have declined at least since the 1920s.
mainly due to declines in annual fatalities owing to
Specifically, comparing the 1920s to the 2000–2006
droughts and floods (see also Figures 2 and 3). The
period, the annual number of deaths declined from
remarkable 99.9 percent drop in annual drought
485,200 to 22,100 (a 95 percent decline), while the death
fatalities indicates that, for whatever reason,
rate per million dropped from 241.8 to 3.5 (a decline of
99 percent).

Table 1 provides a breakdown of the average annual

global deaths and death rates for the various categories Figure 2 Droughts: global deaths & death rates,
1900–2006
of extreme events for 1900–1989 and 1990–2006. The
500 Deaths per year (in 1000s)
columns are arranged in order of declining mortality 472
Death rates per year (per million)
ascribed to the various events (highest to lowest) for the
former period. This shows that: 400
345
n During much of the 20th Century, the deadliest
300
extreme events were droughts, followed by floods
235
and windstorms. Over the 107-year record, droughts
and floods were responsible for 59 and 35 percent of 200
147 151
all fatalities worldwide due to all extreme weather 126

events, while windstorms contributed an additional 100 78

55
6 percent. Thus, these three categories together 43

9 5 12 12
accounted for over 99 percent of the fatalities due to 0
0 0 0 0 3 0 0 0 0

extreme events. 1900– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000–
09 1919 1929 1939 1949 1959 1969 1979 1989 1999 2006
n Aggregate annual mortality for the seven categories
Sources: EM-DAT (2007); McEvedy and Jones (1978); WRI (2005, 2007)
of extreme events declined by 87 percent between

5
Death and Death Rates Due to Extreme Weather Events

Figure 3 Floods: global deaths and death rates, Figure 4 Windstorms: global deaths and death rates,
1900–2006 1900–2006
500 Deaths per year (in 1000s) 40 Deaths per year (in 1000s)
Deaths
Deaths per
per year
year (in
(in 1000s)
1000s)
36
436 Death rates per year (per million) Death rates per year (per million)

Death
Death rates
rates per
per year
year (per
(per million)
million)
400
30

300
21
20
204 206
200
13
12 12
10
10 9

100 6 6
70 5
4 4 4 4 4
3 3
2
10 6 10 4 1 1 1
0 0 0 0 3 1 5 1 5 1 9 2 5 1 0
0 0
1900– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000– 1900– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000–
09 1919 1929 1939 1949 1959 1969 1979 1989 1999 2006 09 1919 1929 1939 1949 1959 1969 1979 1989 1999 2006

Sources: EM-DAT (2007); McEvedy and Jones (1978); WRI (2005, 2007) Sources: EM-DAT (2007); McEvedy and Jones (1978); WRI (2005, 2007)

available food supplies per capita have increased in The spike in deaths and death rates owing to
marginal areas. This is possibly due to greater food extreme temperatures during the 1990–2006 period
production at the global level, and an enhanced occurred because of the 2003 European heat wave.
ability to move food from food surplus areas to However, these fatalities were more than
deficit areas through institutions such as compensated for by declines in flood and drought
international trade, and governmental and fatalities.5
nongovernmental aid agencies and philanthropies
(e.g., through the World Food Program or the
The global mortality burden from extreme events
International Red Cross). All of this is facilitated by
better transportation and communication networks, To place the current death toll due to all extreme
and irrigation facilities (Goklany 1998). weather events in a wider context, consider that the
The 89.8 percent decline in annual flood fatalities average annual death toll for 2000–2006 due to all
between the two periods possibly reflects better weather-related extreme events according to EM-DAT
control, prevention and management of floods (2007) was 19,900. By contrast, the World Health
through construction of dams and other Organization (2004) estimates that in 2002, a total of
infrastructure, supplemented by better emergency 57.0 million people died worldwide from all causes,
response measures facilitated by improvements in including 5.2 million from other kinds of accidents. Out
transportation systems, flood forecasting, and of these, road traffic was responsible for 1.2 million
management of water facilities, among other things. deaths, violence (other than war) for 0.6 million, and
war for 0.2 million (see Table 2). Thus, while extreme
n While average annual fatalities due to windstorms
weather-related events garnish plenty of attention
increased from around 11,000 to 14,000 per year
worldwide because of their episodic and telegenic
between the two periods, the annual mortality rates
nature, their contribution to the global mortality burden
declined by 38 percent (see also Figure 4).
is relatively minor: 0.03 percent of global deaths. Their
n Annual mortality rates dropped for virtually every contribution to the global burden of disease should be
category except extreme temperatures and slides. similarly small.

6
 Death and Death Rates Due to Extreme Weather Events

Table 2  Global deaths per year due to various causes, early 2000s. Note: All data are for 2002, except for deaths due to
extreme weather events, which are based on the annual average from 2000–2006

Cause of death No. of deaths Percent of total deaths

I. Communicable Diseases 18,324,000 32.13%

Tuberculosis 1,566,000 2.75%
HIV/AIDS 2,777,000 4.87%
Diarrhoeal diseases 1,798,000 3.15%
Malaria 1,272,000 2.23%
Other tropical diseases 129,000 0.23%
Other infectious & parasitic diseases 3,362000 5.90%
Subtotal – Infectious and parasitic diseases 10,904,000 19.12%
Respiratory infections 3,963,000 6.95%
Nutritional deficiencies 485,000 0.85%
Maternal and perinatal conditions 2,972,000 5.21%
II. Non-communicable Conditions 33,537,000 58.81%
Malignant neoplasms 7,121,000 12.49%
Cardiovascular diseases 16,733,000 29.34%
Respiratory diseases 3,702,000 6.49%
Other non-communicable conditions 5,981,000 10.49%
III. Injuries 5,168,000 9.06%
Road traffic accidents 1,192,000 2.09%
Violence 559,000 0.98%
War 172,000 0.30%
Extreme weather events 19,868 0.03%
All other injuries 3,225,600 5.66%

Sources: WHO (2004), EM-DAT (2007)

Notably, the contribution of extreme events to the and (c) 27,000 out of over 1 million deaths due to
mortality burden for accidental causes of death is also malaria (see WHO 2002). It also ascribed 2,000 deaths to
small (at 0.4 percent). Also, over the last fifty years at floods in 2000, based on the EM-DAT database.
least, the general decline in annual mortality due to
Although the review paper’s estimates for non-flood-
extreme weather events (see Figure 1) has occurred
related deaths are problematic,6 if one accepts them as
despite an increase in all-cause mortality (WRI 2005).
valid, that means that climate change currently accounts
A review paper in Nature (Patz et al. 2005) estimates that for less than 0.3 percent of all global deaths. Accordingly,
climate change may have been responsible for over based on current contributions to the global mortality
150,000 deaths worldwide in 2000. This estimate is burden, other public health issues outrank climate
largely based on an analysis put out under the auspices change.
of the World Health Organization (McMichael et al.
2004). The latter study arrived at its estimate by
ascribing to climate change (a) 77,000 out of about
250,000 deaths due to protein malnutrition, (b) 47,000
out of about 2 million deaths due to diarrhoeal disease,

7
Death and Death Rates Due to Extreme Weather Events

U.S. trends in mortality and mortality rates,

1900–2006 Figure 5 Cumulative U.S. deaths and death rates,
1979–2002, for hurricanes, floods, lightning,
Among the problems in developing a long time series for tornados, extreme heat and extreme cold
U.S. deaths due to all (or most) extreme weather
3,000 12
phenomena is that the length of the U.S. (data) record
11
varies according to the type of event.
2,500 10

The Annual Summaries (e.g., NCDC 2004, 2007) 9

Death rates (per million)

published by National Oceanic and Atmospheric 2,000 8
7
Administration’s National Climatic Data Center (NCDC)

Deaths
1,500 6
provides time series data on fatalities due to hurricanes,
5
floods, tornadoes and lightning, respectively, from 1900,
1,000 4
1903, 1916 and 1959 onward. Each year’s summary also
3
gives that year’s death toll due to a variety of other Deaths
500 Death rate (per million) 2
weather-related phenomena such as extreme cold, Death rate (per million), R2=0.36 1
Deaths, R2=0.18
extreme heat, drought, mudslides, winter storms, 0 0
avalanches, etc., but it does not provide any time series 1979 1982 1985 1988 1991 1994 1997 2000

data for these other categories of natural disasters.

Linear trend lines were generated using EXCEL.
Sources: for hurricanes, Blake et al. (2007); for lightning and tornados, NCDC (2004, 2007) and
Another problem is that the data for several phenomena Hinson (2005); for floods, HIC (2007); for extreme heat and cold, CDC (2005)

from these summaries are at variance with other data

sources. Specifically, there are discrepancies between
Center for Disease Control’s WONDER database was
mortality data from these Annual Summaries and (a) the
used for extreme heat and cold, because it is based on
Hydrologic Information Center’s (HIC 2007) estimates
actual death certificate records, which, in turn, are based
for floods, (b) the National Hurricane Center’s data for
on medical opinion as opposed to the National Weather
hurricanes (Blake et al. 2007), and (c) CDC’s WONDER
Service’s expert opinion.7
database for extreme cold and extreme heat.
Figure 5 shows the trend in cumulative deaths and
Based on previous conversations with personnel from
death rates for a subset of extreme weather events –
the various agencies (Goklany 2000), it was decided to
specifically, hurricanes, floods, tornados, lightning, and
use HIC data for floods (as adjusted per Goklany 2000),
extreme heat and cold – from 1979–2002.8 It shows that
and Blake et al. (2007) for hurricanes through 2006. The
despite any warming that may have occurred, both

Table 3 US deaths due to weather-related events, 1979–2002. Sources: for extreme events, see text; for total all-cause
mortality, USCB (2004).

Cumulative deaths Deaths per year Percent of annual all-cause deaths

Extreme cold (XC) 16,313 680 0.031%

Extreme heat (XH) 8,589 358 0.016%
Flood (F) 2,395 100 0.005%
Lightning (L) 1,512 63 0.003%
Tornado (T) 1,321 55 0.003%
Hurricane (Hu) 460 19 0.001%
Sum 30,590 1,275 0.058%
Total deaths, all causes, 1979–2002 average 2,189,000 100.000%

8
 Death and Death Rates Due to Extreme Weather Events

Figure 6 U.S. Deaths due to hurricanes (1900–2006), Figure 7 U.S. deaths rates due to hurricanes
floods (1903–2006), tornados (1916–2006), (1900–2006), floods (1903–2006), tornados
lightning, (1959–2006) and total (1959-2006) (1916–2006), lightning, (1959–2006) and total
10-year moving averages (1959–2006)
800 Lightning (L) 10-year moving averages
Tornados (T) 12 Lightning (L)
Floods (F) 11 Tornados (T)
600 Hurricanes (H) 10 Floods (F)
Total (=L+T+F+H) 9 Hurricanes (H)
Deaths (in 1000s)

Death rate (per million)

8 Total (=L+T+F+H)

400 7
6
5
200 4
3
2
0 1
1900 1915 1930 1945 1960 1975 1990 2005 0
1900 1915 1930 1945 1960 1975 1990 2005
Sources: for hurricanes, Blake et al. (2007); for lightning and tornados, NCDC (2007) and NWS
(2007); for floods, HIC (2007) and Goklany (2000) Sources: for hurricanes, Blake et al. (2007), for lightning and tornados, NCDC (2007) and NWS
(2007); for floods, HIC (2007) and Goklany (2000); and for population, USCB (2007)

deaths and death rates have not increased over this

period. If anything, they might have declined over this Hurricane Katrina – 80 percent for hurricanes. Such
period, during which all-cause mortality increased by 28 declines are consistent with results of earlier analyses
percent (USCB 2004). (Goklany 2000).

Notably, the bulk of the weather-related deaths (53 Similarly, Figure 7 indicates that the corresponding
percent) during this period were caused by extreme cold. declines for death rates (comparing their peaks with the
In rank of importance, these were followed by extreme most recent 10-year period) were 80 percent for floods,
heat, floods, lightning, tornados, and hurricanes, which 75 percent for lightning, 95 percent for hurricanes and
contributed 28, 8, 5, 4 and 2 percent, respectively (see 92 percent for tornados.
Table 3.)
Both figures show:
Figures 6 and 7 show the 10-year moving averages for
n A large spike for hurricanes at the very beginning of
deaths (and death rates) due to hurricanes from
the record due to the Galveston Hurricane of 1900,
1900–2006, floods from 1903–2006, tornados from
and a smaller spike due to Hurricane Katrina at the
1916–2006, and lightning from 1959–2006, as well as for
very end. This is punctuated by a relatively high
cumulative deaths (and death rates) from these four
plateau in the late 1920s which extended into the
individual categories of events (from 1959 to 2006).
late 1930s, and troughs extending from the late
Death rates in Figure 7 are calculated using national
1940s to the early 1950s and, again, in the 1980s
population estimates from the US Census Bureau
and 1990s. 10 This is also illustrated in Figure 8.11
(USCB).9
n Mortality and mortality rates dropped more or less
Figure 6 shows that for the most recent 10-year period
steadily for lightning and tornados over the period
for which data are available, average annual deaths
for which records are available. These results are
declined below their earlier peaks in the 10-year moving
consistent with earlier analyses for both sets of
averages by 62 percent for floods, 63 percent for
lightning, 81 percent for tornados and – despite

9
Death and Death Rates Due to Extreme Weather Events

Figure 8 Annual U.S. deaths and death rates due to Figure 9 Annual U.S. deaths and death rates due to
hurricanes, 1900–2006 floods, 1903–2006
1,200 1,135 Deaths/year 200 Deaths/year
Death rate (per 100 million)/year Death rates/year/100 million people 181

1,000 R2=0.29 175 R2=0.05

873 R2=0.35 R2=0.27
150
800
128
125
112
600 110
101 103 99
97
100
85
400 293
244 76 79
75
238
75
62 62
200 59 48 48
104 102 115 24 11 50 44
89 88 81 38
52 13 5 36 39
22 16 30 25 9
29
0 25 21

–200 0
1900– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000– 1903– 1910– 1920– 1930– 1940– 1950– 1960– 1970– 1980– 1990– 2000–
09 19 29 39 49 59 69 79 89 99 06 09 19 29 39 49 59 69 79 89 99 06

Linear trend lines were generated using EXCEL. Linear trend lines were generated using EXCEL. The data for 1900–1909 are based on the seven-
Sources: Blake et al. (2007); USCB (2007) year average from 1903 through 1909 because the HIC’s series starts in 1903.
Sources: HIC (2007); Goklany (2000); USCB (2007)

events by Goklany (2000), and Doswell et al. (1999) have declined somewhat (see also Figure 9). Once
for tornados. final figures are compiled, mortality due to floods in
2005 may be higher than in other recent years
n Mortality from floods exhibits no particular trend
because of floods related to hurricanes that year.12
from 1903 to 2006, although mortality rates might
n Cumulative annual mortality and mortality rates for
the above four categories of extreme weather events
also indicate no specific trend from 1959–64 to
Figure 10 Cumulative annual U.S. deaths and death
2001–06, although death rates may have declined
rates from hurricanes, floods, lightning, and
tornados, 1959–2006 (Figure 10; see also Figures 6 and 7).
500 Total deaths
445 Total death rate (per 100 million) The previous figures (6 through 10) show that current
434
405 R2=0.01 deaths and death rates are, in general, lower than in the
400 R2=0.31
past. Predictably, death rates have declined more rapidly
than deaths per year, confirming results from previous
300
260
270 266 studies (Goklany 2000).
242
222
204 210
200
141 147 Discussion and conclusions
120
112
100 87 88 The information presented above indicates that globally
as well as for the United States, the aggregate
0 contribution of extreme weather events to the mortality
1959– 1965– 1971– 1977– 1983– 1989– 1995– 2001– burden is relatively minor at present, ranging from 0.03
64 70 76 82 88 94 2000 06
to 0.06 percent.
Linear trend lines were generated using EXCEL.
Sources: For hurricanes, Blake et al. (2007); for lightning and tornados, NCDC (2004, 2007) and
NWS (2007); for floods, HIC (2007); for population USCB (2007) Moreover, if the frequency, intensity and duration of
extreme weather events has increased in recent decades

10
 Death and Death Rates Due to Extreme Weather Events

– all empirical issues best left to climatologists – there is comparing the most recent 10-year period against
no signal of that in the data on either mortality rates or the peak periods, mortality declined by 60–80
(more importantly) mortality, despite an increase in percent for floods, lightning, tornados and
populations at risk. hurricanes, while mortality rates declined by 75–95
percent. However, there are no consistent trends for
The data at the global level, while incomplete, indicate:
mortality due to floods, and both mortality and
n Aggregate annual mortality and mortality rates mortality rates for hurricanes spiked in 2005.
owing to extreme weather events have declined Nevertheless, these spikes are lower than levels than
between 95 and 99 percent, respectively, since the were reached in previous periods.
1920s regardless of whether the frequencies,
intensities and/or durations of extreme weather Thus, it appears that mortality rates and – more
events have increased (or not) due to human- significantly – mortality for the most deadly and
induced or natural climate change. destructive forms of extreme weather events have
declined substantially over the past several decades. This
n Much of the above improvement is due to a
suggests two possibilities if indeed climate change has
substantial decline in mortality due to droughts and
exacerbated extreme events. One possibility is that
floods, which apparently caused 93 percent of the
humanity has been extremely lucky in terms of where
fatalities due to extreme events between 1900 and
and when these events have struck. The other possibility,
2006.
a more likely explanation, is that society’s ability to cope
n Death rates for the different categories of extreme with extreme events has not only improved, it has also
events were generally lower in the 1990s and early- put its increased adaptive capacity to good effect.
to mid-2000s than in previous decades – with the
Several interrelated factors have contributed to this
notable exception of death rates for extreme
increase in adaptive capacity. First, today’s societies have
temperatures, which were higher because of the
a wider range of technological options at their disposal,
2003 European heat wave.
enabling them to finesse the consequences of extreme
n Regarding windstorms, both mortality and mortality events before they strike and to cope with their
rates peaked in the 1970s. The average annual aftermath after they have struck. Such options range
mortality due to windstorms from 1990–2006 from early warning systems, building codes, and better
exceeded the average over 1900–1989 by 26 percent, meteorological forecasts, to better construction,
but the mortality rate was 38 percent lower. communications and transportation systems, all of
which have increased the ability to transport people and
With respect to the United States: materiel (including food, medical and other essential
supplies) in and out of disaster zones.
n In an average year, more lives are lost to extreme
temperatures – both extreme heat and extreme cold Second, many of these options were learnt through
– than to more heavily publicized events such as experience and were enabled through the ability of
tornados, hurricanes and floods. According to data wealthier societies to research and develop new
from the Centers for Disease Control, on average technologies and practices. Greater wealth also allowed
extreme cold claims more lives than tornados, them to obtain and implement more effective
floods, lightning, hurricanes and extreme heat, technologies. Once those technologies and practices
combined. have been developed in wealthier, more technologically
sophisticated nations, it is possible for poorer societies to
n In general, mortality and mortality rates from the
learn from and adapt them to their own circumstances.
various categories of extreme events examined here
(tornados, hurricanes, floods, lightning and extreme Third, societies have greater access to human and social
temperatures) are lower today than they have been capital to protect themselves from, and cope with,
in the past. Based on 10-year moving averages, adversity in general and extreme weather events in

11
Death and Death Rates Due to Extreme Weather Events

particular (IPCC 1991; Goklany 1995, 2000, 2006; impossible to say at this time whether the upward trend
Goklany and Straja 2000). for economic losses would hold were global economic
losses to be measured in terms of global wealth, because
However, as indicated by the European experience with
appropriate data are lacking.
the 2003 heat wave and the U.S. experience with
Hurricane Katrina, the importance of human and social
capital cannot be overemphasized. They are just as Notes
important as wealth and increased access to technology.
1 This paper uses “extreme weather events”
Moreover, greater adaptive capacity is necessary but not
synonymously with “extreme events.”
sufficient to effectively cope with extreme events. Such
capacity must be deployed more rapidly and used more 2 The risk of death due to any cause, e.g., floods, is
fully. estimated by the number of deaths caused by floods
divided by the magnitude of the population
Nevertheless, the decline in deaths and death rates from
potentially exposed to floods. Thus the death rate
extreme events indicates that if the frequencies,
for floods (measured as the number of flood deaths
intensities and durations of extreme events have
either per 1,000 people or per million people) is a
increased, then adaptive capacity has evolved even
measure of the risk due to floods. If the risk of
faster. Thus far wealth, technology, and human and
floods increases, then if all else stays the same, the
social capital have for the most part (despite notable
flood death rate should increase. But, in fact, all
exceptions) apparently trumped natural climatic
else rarely stays the same. Increases in the
variability and human-induced climatic change.
frequencies and magnitudes of floods may be offset
Finally, it has been suggested that declining trends for by better protection from floods or disaster
deaths (and death rates) due to extreme events might be preparedness, for instance. Thus trends in death
accompanied by – or might contribute to – increasing rates from floods tell us whether overall risk to
trends in economic losses. This idea is based on the populations from floods are increasing, decreasing
general notion that individuals might accept greater or staying constant.
financial risks provided personal safety is more assured
3 Figure 1 is constructed using data the following
(Goklany 2000). In addition, a wealthier society is likely
sources: (1) For deaths, EM-DAT (2007). EM-DAT
to have more property at risk.
is the International Disaster Database maintained
For the U.S., results of this paper, combined with by the Office of Foreign Disaster Aid and Center for
previous studies of economic losses due to various Research on the Epidemiology of Disasters at the
categories of extreme events, suggest (but do not prove) Université Catholique de Louvain, Brussels,
that declining mortality is indeed accompanied by Belgium. (2) For population from 1900–1925,
higher economic losses if losses are measured in terms McEvedy & Jones (1978). (3) For population from
of real (constant) dollars for floods, hurricanes, and 1950–2006, World Resources Institute (2007). (4)
tornados. However, this does not seem to be the case if For population from 1926–1949, estimates were
losses are measured in terms of a fraction of wealth based on interpolation for each year using the 1925
(measured as fixed tangible reproducible assets, which estimate from McEvedy and Jones and the 1950
excludes land values) (Pielke and Landsea 1998, Doswell estimate from WRI (2005), assuming exponential
et al. 1999, Goklany 2000, Pielke, Jr, and Downton 2000, population growth. For 2004, I excluded the deaths
Downton et al. 2005). due to the Boxing Day Tsunami disaster (which,
according to EM-DAT killed 226,435 people). Death
Globally, the analysis in this paper and the estimates of
estimates, in particular, are approximate and,
economic losses in constant dollars provided by the
possibly, more prone to error as we go further into
IPCC’s Third Assessment Report suggest that declining
the past. As is evident from the following footnote,
mortality is also accompanied by increasing property
EM-DAT is not quite complete. While events in the
losses (see also, Faust et al. 2006). However, it is

12
 Death and Death Rates Due to Extreme Weather Events

earlier years might have been missed, EM-DAT on modeling studies, with quantification based on
should have captured the major natural disasters, anecdotal information. The temperature-disease
particularly in recent years. This suggests that relationship used to develop the estimate for
mortality and mortality rates might have been diarrhea, for example, was based on 6 years worth
higher in the early decades of the 20th century than of data from Lima, Peru, and 20 years of data from
is indicated in Figure 1, and subsequent figures and Fiji. In addition, the amount of climate change
tables. estimated for 2000 was based on the results of a
general circulation model at resolution of 3.750
4 EM-DAT contains data on the occurrence and
longitude and 2.50 latitude. The results of such
effects of over 7,700 hydro-meteorological disasters
models, which are inexact at best at the global level,
in the world from 1900 to present. Specifically,
tend to greater uncertainty as the resolution gets
these disasters are due to drought, extreme
finer.
temperatures, floods, slides, waves and surges,
wildfires, and wind storms. The data are compiled 7 To further complicate matters, the NWS website
from various sources, including UN agencies, non- provides a “67-Year List of Severe Weather
governmental organizations, insurance companies, Fatalities” (from 1940–2006. Unfortunately, the
research institutes and press agencies. For a disaster data in this list for lightning is inconsistent with
to be entered into the database one or more of the the data from its Annual Summaries. Enquiries to
following criteria must be met: (a) at least 10 NOAA, so far, have not resolved these discrepancies
people must have been reported killed, (b) at least satisfactorily.
100 people must have been reported as affected, (c)
8 In the WONDER database, mortality data for
a state of emergency must have been declared, or
1979–1998 are coded using the International
(d) there should have been a call for international
Classification of Disease, version 9 (i.e., ICD-9) for
assistance.
1979–1998, and ICD-10 for 1999 onward. To
5 EM-DAT (2005) ascribed over 45,700 deaths to the identify deaths due to extreme heat, I used codes
2003 European heat wave, while EM-DAT (2007) E900.0 and E900.9 for ICD-9 (per Goklany and
assigns over 72,200 deaths to that event. Straja 2000), and X30 for ICD-10. The
corresponding codes used for extreme cold were
6 The authors themselves note that among the
E901.0 and E901.9, and X31, respectively.
challenges in developing estimates of the health
impacts of climate change is that “climate change 9 For hurricanes it might have been more appropriate
occurs against a background of substantial natural to use annual estimates of the coastal population to
climate variability, and its health effects are estimate death rates, but that would have
confounded by simultaneous changes in many complicated calculations of cumulative death rates.
other influences on population health….Empirical
10 These figures are based on an estimate of 1,525
observation of the health consequences of long-
deaths in 2005 per Blake et al. (2007). The figures
term climate change, followed by formulation,
also assume a death toll of 8,000 for the 1900
testing and then modification of hypotheses would
Galveston Hurricane (Blake et al. 2007).
therefore require long time series (probably several
decades) of careful monitoring. While this process 11 Note that in Figure 8 the data for the last period is
may accord with the canons of empirical science, it based on a seven-year average, while that for other
would not provide the timely information needed to periods are based on ten-year averages.
inform current policy decisions on GHG emission
12 There is a probably an unavoidable degree of
abatement, so as to offset possible health
subjectivity involved in assigning deaths from
consequences in the future. Nor would it allow
hurricane-initiated events to either the hurricane or
early implementation of policies for adaptation to
flood categories.
climate changes.” Hence the estimates were based

13
Death and Death Rates Due to Extreme Weather Events

References Goklany, I.M. 1995. “Strategies to enhance adaptability:

technological change, economic growth and free trade.”
Anderson, J.; Bausch, C. 2006. Climate change and
Climatic Change, 30, 427–449.
natural disasters: Scientific evidence of a possible
relation between recent natural disasters and climate Goklany, I.M. 1998. “Saving habitat and conserving
change. Policy Brief for the EP Environment Committee, biodiversity on a crowded planet.” BioScience, 48,
IP/A/ENVI/FWC/2005–35. Available at <http://www. 941–953.
europarl.eu.int/comparl/envi/pdf/externalexpertise/
Goklany, I.M. 2000. “Potential consequences of
ieep_61eg/naturaldisasters.pdf>. Accessed on 27 April
increasing atmospheric CO2 concentration compared to
2006.
other environmental problems.” Technology, 7S, 189–213.
Blake, E.S.; Rappaport, E.N.; Landsea, C.W.; and Miami
Goklany, I.M. 2006. “Integrated strategies to reduce
NHC 2007. The Deadliest, Costliest, and Most Intense United
vulnerability and advance adaptation, mitigation, and
States Hurricanes from 1851 to 2006 (and other Frequently
sustainable development.” Forthcoming in Mitigation and
Requested Hurricane Facts).
Adaptation Strategies for Global Change .
Brooks, H.E.; Doswell III. C.A. 2001: Normalized
Goklany, I.M.; Straja, S.R. 2000. U.S. death rates due to
Damage from Major Tornadoes in the United States:
extreme heat and cold ascribed to weather, 1979–1997.
1890–1999. Weather and Forecasting, 16, 168–176.
Technology, 7S, 165–173.
Doswell III, C.A.; Moller, A.R.; Brooks, H.E. 1999. Storm
Hinson, S. 2005. National Climatic Data Center. Personal
Spotting and Public Awareness since the First Tornado
communication (August 22).
Forecasts of 1948. Weather and Forecasting, 14, 544–557.
Hydrologic Information Center. 2005. Flood Fatalities.
Downton , M.W.; Miller, J.Z.B.; Pielke Jr., R.A. 2005.
Available at http://www.nws.noaa.gov/oh/hic/flood_stats/
Reanalysis of U.S. National Weather Service Flood Loss
recent_individual_deaths.shtml. Accessed 22 August
Database. Natural Hazards Review, February 2005: 13–22.
2005.
Emanuel, K.. 2005a. Increasing destructiveness of
Intergovernmental Panel on Climate Change. 1991.
tropical cyclones over the past 30 years. Nature, 436,
Climate Change: The IPCC Response Strategies. Washington,
686–688, posted on-line August 4, 2005.
DC, Island Press.
Emanuel, K. 2005b. Meteorology: Emanuel replies.
Intergovernmental Panel on Climate Change. 2001.
Nature, 438, E13.
Climate Change 2001: Impacts, Adaptation, and Vulnerability.
EM-DAT: The OFDA/CRED International Disaster Cambridge, Cambridge University Press, p. 7.
Database. 2005. Available at www.em-dat.net, Université
Kluger, J. 2005. Climate change: The culprit? Time,
Catholique de Louvain, Brussels, Belgium. Data
October 3, 2005.
downloaded on 6 September 2005.
Kluger, J. 2006. Earth is at the tipping point. Time,
EM-DAT: The OFDA/CRED International Disaster
March 27, 2006.
Database. 2007. Available at www.em-dat.net, Université
Catholique de Louvain, Brussels, Belgium. Data Landsea, C.W. 2005. Hurricanes and global warming.
downloaded on 20 September 2007. Nature, 438, E12-E13.

Faust, E.; Höppe, P.; Wirtz, A.: Schmidt, S. 2006. Trends MacMichael, A, Woodruff, R. 2004. Climate change and
in natural catastrophes – potential role of climate risk to health BMJ , 329, 1416–1417 (18 December).
change, Climate Change and Disaster Losses Workshop:
McEvedy, C.; Jones, R. 1978. Atlas of World Population
Understanding and Attributing Trends and Projections,
History, New York, Penguin.
Hohenkammer, Bavaria, Germany, May 25–26, 2006.

14
 Death and Death Rates Due to Extreme Weather Events

McMichael, A., et al. 2004: Global climate change. In: U.S. Census Bureau. 2007. Statistical Abstract of the United
Comparative Quantification of Health Risks: Global and States, 2007. Available at http://www.census.gov/
Regional Burden of Disease due to Selected Major Risk Factors. compendia/statab/.
World Health Organization, Geneva, pp. 1543–1649.
Webster P., et al. 2005. Changes in tropical cyclone
National Climatic Data Center. 2005. 2004 Annual number, duration, and intensity in a warming
Summaries. Available at http://www.nws.noaa.gov/om/ environment. Science, 309, 1844–1846.
severe_weather/sum04.pdf.
World Health Organization. 2002. World Health Report
National Climatic Data Center. 2007. 2005 Annual 2002 – Reducing Risks, Promoting Healthy Life. WHO,
Summaries. Available at http://www.nws.noaa.gov/om/ Geneva, Annex Table 2, p. 186.
severe_weather/sum05.pdf.
World Health Organization. 2004. World Health Report
National Weather Service. 2007. Summary of U.S. Natural 2004 – Changing History. WHO, Geneva, Annex Table 2, p.
Hazard Statistics. Available at http://www.nws.noaa.gov/ 120.
om/severe_weather/sum04.pdf.
World Resources Institute. 2005. Earth Trends. Available
Patz J. 2004. Climate change: Health impacts may be at http://www.wri.org.
abrupt as well as long-term. BMJ , 328, 1269–70.
World Resources Institute. 2007. Earth Trends. Available
Pielke, Jr., R.A. 2005. Are there trends in hurricane at http://www.wri.org.
destruction? Nature, 438, E11.

Pielke, Jr., R.A.; Downton, M.W. 2000. Precipitation and

Damaging Floods: Trends in the United States, 1932–97.
Journal of Climate, 13, 325–337.

Pielke, Jr., R.A.; Landsea, C.W. 1998. Normalized

Hurricane Damages in the United States: 1925–95.
Weather and Forecasting,13, 621–631.

Schiermeier, Q. 2003. Climate change: The long road

from Kyoto. Nature, 426, 756.

Schiermeier, Q. 2005. Trouble brews over contested trend

in hurricanes. Nature, 435, 1008–1009.

Trenberth, K.; Taylor, G. Undated. “Does Climate Change

Cause More Extreme Weather? YES. Economic Effects of
Weather Disasters Warrant Action on Warming; NO.
Blaming Disasters on Global Warming Doesn’t Help.”
Washington Post Issues Forum: Global Climate Change,.
Available at www.washingtonpost.com/wp-adv/
specialsales/nei/global/article9.htm. Accessed 4 February
2003.

U.S. Census Bureau. 2004. Statistical Abstract of the United

States, 2004–2005. Available at http://www.census.gov/
compendia/statab/ .

15