ORIGINAL PAPER
A biometeorology study of climate and heat-related
morbidity in Phoenix from 2001 to 2006
Jay S. Golden & Donna Hartz & Anthony Brazel &
George Luber & Patrick Phelan
Received: 6 July 2007 /Revised: 16 December 2007 /Accepted: 16 December 2007 /Published online: 25 January 2008
#
ISB 2008
Abstract Heat waves kill more people in the United States
than hurricanes, tornadoes, earthquakes, and floods combined.
Recently, international attention focused on the linkages and
impacts of human health vulnerability to urban climate when
Western Europe experienced over 30,000 excess deaths
during the heat waves of the summer of 2003—surpassing
the 1995 heat wave in Chicago, Illinois, that killed 739. While
Europe dealt with heat waves, in the United States, Phoenix,
Arizona, established a new all-time high minimum tempera-
ture for the region on July 15, 2003. The low temperature of
35.5°C (96°F) was recorded, breaking the previous all-time
high minimum temperature record of 33.8°C (93°F). While an
extensive literature on heat-related mortality exists, greater
understanding of influences of heat-related morbidity is
required due to climate change and rapid urbanization
influences. We undertook an analysis of 6 years (2001–
2006) of heat-related dispatches through the Phoenix Fire
Department regional dispatch center to examine temporal,
climatic and other non-spatial influences contributing to high-
heat-related medical dispatch events. The findings identified
that there were no significant variations in day-of-week
dispatch events. The greatest incidence of heat-relatedmedical
dispatches occurred between the times of peak solar irradiance
and maximum diurnal temperature, and during times of
elevated human comfort indices (combined temperature and
relative humidity).
Keywords Health vulnerability
.
Heat Waves
.
Urban Climate
.
Morbidity
.
Emergency medical dispatch
Introduction
Concerns regarding system interactions and complexities
between global climate change and human health vulnera-
bility are increasing significantly. Recently, the summer of
2003 brought international focus on the linkages and impacts
of human health vulnerability to climate change not at the
global scale but rather at the urban scale. Western Europe
experienced over 30,000 excess deaths during the heat waves
of the summer of 2003 (Kosatsky 2005)—surpassing the
1995 heat wave in Chicago, Illinois, that killed 739 (Shrader-
Frechette 2002; Kalkstein and Greene 1997).
Concurrent with the European heat waves, in the United
States, Phoenix, Arizona, established a new all-time high
minimum temperature for the region on July 15, 2003. The
low temperature of 35.5°C (96°F) was recorded, breaking
the previous all-time high minimum temperature record of
33.8 (93°F) which was set on June 27 1990, July 20, 1989
and July 14, 2003. In the US, heat waves kill more people
than hurricanes, tornadoes, earthquakes, and floods com-
bined (Klineberg 1999; National Weather Service 2007a).
Int J Biometeorol (2008) 52:471–480
DOI 10.1007/s00484-007-0142-3
J. S. Golden (*)
National Center of Excellence, SMART Innovations for Urban
Climate & Energy, Arizona State University,
929 South Mill Avenue #151, P.O. Box 873211, Tempe, AZ
85287-3211, USA
e-mail: Jay.Golden@asu.edu
D. Hartz
:
A. Brazel
School of Geographical Sciences, Arizona State University,
Tempe, AZ, USA
G. Luber
Centers for Disease Control & Prevention,
National Center for Environmental Health,
Atlanta, GA, USA
P. Phelan
Mechanical & Aerospace Engineering, Arizona State University,
Tempe, AZ, USA

A strong volume of literature exists concerning the
system dynamics of heat waves and the urban heat islands
in regards to sustainable development (Golden 2004;
Golden et al. 2006) including heat-related mortality
(Centers for Disease Control 1995; Semenza et al. 1996;
Kalkstein et al. 1996). This paper is focused on heat-related
morbidity and is the result of a joint research effort by the
National Center of Excellence on SMART Innovations for
Urban Climate and Energy at Arizona State University
(NCE) and the National Center for Environmental Health at
the Centers for Disease Control and Prevention (CDC). The
NCE in partnership with the CDC is undertaking studies of
multiple urban regions to increase understanding of how
climate change, including heat waves and electricity black-
outs, influence human health vulnerability. This project
examines findings from 6 years of emergency response
dispatches for heat-related health incidents in the Phoenix
metropolitan region. These fire/EMS dispatches were
tracked and analyzed in comparison to meteorological
conditions including heat waves and National Weather
Service Heat Advisory Warnings.
Region of study
Phoenix, Arizona, was selected as the region of study. A
Centers for Disease Control (Centers for Disease Control
and Prevention 2005) study revealed that, from 1979 to
2002, a total of 4,780 heat-related deaths in the US resulted
from weather conditions and that, from 1993 to 2002, the
total incidence of such deaths was three to seven times
greater in Arizona than in the US overall. Additionally, over
the twentieth century, average annual temperatures in the
arid subtropical Phoenix region increased 1.7°C (3.1°F)
(Brazel et al. 2000). However, the urban portions of the
region have realized mean annual temperature increases of
4.2°C (7.6°F), a rate of three times the total regional mean
increase representing the pronounced influence of the built
environment (Fig. 1).
The setting, Phoenix, (elevation 345.9 m, 33°25′40″N,
112°0′14″W) was incorporated in 1881 and is one of the
nation’s fastest growing cities and fifth largest in population
(1,475,834 as of September 1, 2005 (U.S. Census 2007).
Geographically the city is over 1,295 km
2
(500 square
miles) and larger than the City of Los Angeles. The
Phoenix Fire Department 911 call center dispatches for
the majority of the regional fire departments. Maricopa
County, the regional jurisdiction containing the City of
Phoenix, has a population of approximately 3.6 million (U.
S. Census 2007) and has a land area of 23,836 km
2
(9,203
square miles). It is the fourth most populous county in the
nation, and is home to more people than 21 states and the
District of Columbia. The jurisdictions located within
Maricopa County and dispatched by Phoenix Fire include
Tempe, Chandler, Scottsdale (added in 2005), Glendale,
Surprise, Buckeye, Tolleson, Peoria, Paradise Valley,
Guadalupe and Goodyear. Phoenix Fire does not dispatch
for the City of Mesa, Arizona, which is located only 9.5 km
(6 miles) from the Phoenix border. Mesa with a population
of 452,000 is larger than (or similar in population to)
Miami, FL; Atlanta, Cincinnati, Pittsburgh, PA; St. Louis,
MO; Atlanta, GA; and Minneapolis, MN. Therefore, heat-
related dispatches by the Mesa Fire Department are not
included within this research, but should be considered in
regards to the totality of heat-related impacts for the region.
Materials and methods
Climate data used in this study are from the National
Weather Service Automated Surface Observing System
located at Sky Harbor International Airport in Phoenix. Daily
normal temperature data were obtained from the National
Weather Service. Daily maximum and minimum temperature
data were acquired from the National Climate Data Center.
Hourly temperature, dew point and relative humidity, and
cloud cover data were provided by the Arizona Office of
Climatology. Solar radiation data were acquired from the
Maricopa County Flood Control District’s Durango weather
station (central Phoenix). To correspond to the daily normal
temperature data, the 30-year average daily dew point
“normal” was calculated using the hourly dew point temper-
ature data from January 1, 1971 through December 31, 2000
(which had <0.02% missing data points).
We examined the Heat Index (HI) (National Weather
Service 2007b) as a possible explanation for annual
variances. The heat index, also known as the “apparent
temperature,” is an index commonly used by the US
National Weather Service and incorporates temperature with
relative humidity to estimate the “feels like” temperature.
Daily HI was calculated using hourly temperature and
Fig. 1 The role of urbanization and engineered materials to urban
climate. A 50-year analysis of annual minimum temperatures for a
urban vs rural site. Source: Golden (2004)
472 Int J Biometeorol (2008) 52:471–480