patrickbdevaney/mia9 · Datasets at Hugging Face

text stringlengths 0 6.44k
as well as for designing strategies and policies to mitigate those
impacts. For example, in many regions the National Weather
Service (NWS) uses a single quantitative threshold for heat advisories and warnings, which can in turn activate resources in cities
and counties to help mitigate risks to citizens. However, this approach can miss inequities in heat impacts that are due to differences in exposure and vulnerability to heat across urban regions
and across different groups (Turek-Hankins et al. 2020). Further,
the official thresholds NWS uses do not always align with local
heat-health impacts (Vaidyanathan et al. 2019; Hondula et al.
2022). In some cases, those thresholds are being reconsidered
based on past performance and future conditions under a warming climate (Metzger et al. 2010; Guirguis et al. 2018; Benmarhnia
et al. 2019; McElroy et al. 2020; Vanos et al. 2020). This is particularly relevant as new federal policies are developed to
protect workers and communities from extreme heat (White
House Briefing Room 2021), many of which may rely on exceedance of certain heat thresholds. Hyperlocal-scale data
more closely tied to the heat exposure of individuals and
groups is critical.
There are a number of existing approaches to characterizing extreme heat within and across cities. Some studies have
used a downscaling approach applied to global data to produce validated maximum temperature and heat index data at
approximately 5-km scale (Funk et al. 2019). However, models operating at urban scales are still in relatively early stages
of development (Talebpour et al. 2021; Zheng et al. 2021) and
will require high-resolution data for verification. Remotely
sensed land surface temperature from thermal infrared sensors on a number of platforms is available at scales of tens to
100 m, which can provide a more complete snapshot of an
urban area at a particular time of day (e.g., Hulley et al. 2019;
Moffett et al. 2019; Hoffman et al. 2020). Yet, there are still
Denotes content that is immediately available upon publication as open access.
Corresponding author: Amy Clement, [email protected]
DOI: 10.1175/JAMC-D-22-0165.1
Ó 2023 American Meteorological Society. This published article is licensed under the terms of the default AMS reuse license. For information regarding
reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).
JULY 2023 CLEMENT ET AL. 863
Unauthenticated \| Downloaded 02/29/24 07:14 AM UTC
questions about how land surface temperature relates to air temperature, and cloud masking can be a significant limitation. Networks of experimental weather stations have been deployed in
some cities (Santamouris et al. 2007; Yang et al. 2020; Foissard
et al. 2019; Richard et al. 2018). There is a growing effort around
mobile sensors, where transects across an urban area can capture more accurate observations of heat (Shandas et al. 2019;
Kousis et al. 2021). These observations are made on impermeable surfaces (roads) and may not be at locations where people
are spending extended periods of time outdoors. Further, these
studies provide temporal snapshots and do not contain information about how extreme heat can vary over the day, the season
or from year to year.
To help fill data gaps on exposure, there has also been a growing role for data collected by community members in cities.
These include stationary sensors (Shi et al. 2021), as well as
urban monitoring sites based at people’s homes or using smartphones (Potgieter et al. 2021; Venter et al. 2020). There has also
been some exploratory work using smart phones (Cabrera et al.
2021) and wearable sensors (Uejio et al. 2018; Hass and Ellis
2019). Further, some studies are also attempting to quantify
indoor heat with similar technologies, and results indicate that
conditions indoors can be very different from outdoor, and in
many cases are much hotter (Vant-Hull et al. 2018; Ayanlade
et al. 2019; Kapwata et al. 2018; Teare et al. 2020). The advantage of these approaches is that heat data can be more easily
tied to impacts since sensors are monitoring in places where people spend time and are closer to personal exposure (Hondula
et al. 2021). They also can provide detailed temporal information about high heat days and times of day. For example, Shi
et al. (2021) showed that parts of Baltimore, Maryland, could experience “hot days” (threshold of 358C in their study) even
when the forecast did not indicate it. This illustrates the significance of having accurate, quantitative data at appropriate scales,
particularly when thresholds are used to trigger various actions
for mitigation of impacts.
Across the state of Florida, the NWS issues heat advisories
when the forecasted heat index exceeds 1088F (42.28C) for
more than 2 h in the forecast period with an 80% chance or
greater of occurrence, and an excessive heat warning for a
heat index of 1138F (458C) for the same amount of time and
chance of occurrence. This threshold is the same for all NWS
offices across the state and is based on a study of emergency
room admissions for several counties (Florida Department of
Health 2011). Advisories are rarely triggered in Miami–Dade
County, Florida, based on the area forecast. A historical analysis of NWS stations in the county is consistent with this lack of
advisories since those stations rarely report values of heat index in excess of 1088F (Florida Department of Health 2011).
Taken at face value, this would seem to suggest that heat exposure tied with health impacts has not been an issue requiring
mitigation actions in Miami–Dade County. This view is supported by studies that have shown that mortality data has no
clear link between heat and excess deaths in Miami–Dade
(Kalkstein and Greene 1997; Medina-Ramon and Schwartz
2007; Anderson and Bell 2011). Indeed, in Miami–Dade
County’s 2020 Local Mitigation Strategy (LMS; Miami–Dade
County 2020), extreme heat was considered a low impact
hazard in its Threat and Hazard Identification and Risk Assessment (THIRA) and not further considered in its LMS owing to only one extreme heat event reported between 1950 and
2019 (2% chance of occurrence per year).
Here we challenge the notion that exposure to extreme
heat, as measured by daily maximum heat index, is rare in
Miami–Dade County. We utilize hyperlocal temperature data
throughout the county to show that maximum heat exposure is
significantly greater in localized conditions of Miami–Dade
County when compared with NWS observations. We argue that
these hyperlocal data, more closely aligned with where people
are living and working, are a critical complement to NWS
station-based observations of potential heat exposure. This
localized information helps to fill a gap in communicating about
the dangers of heat exposure and the attribution of impacts that
can be used to guide regional decision-making and planning.
2. Methods
From September 2018 to January 2021, six iButton sensor deployments took place during different times of the
year (summer, fall, and winter). Temperature sensor deployments were organized through the Shading Dade initiative, a heat monitoring project in Miami–Dade County
launched by Florida International University’s Sea Level
Solutions Center in the Institute of Environment, and its
Department of Journalism and Media, as well as the nonprofit Catalyst Miami, and the University of Miami.
In total, 130 sensors were deployed and recovered; of those,
65 were in sunny locations, 54 in shaded locations, and 11 were
reported as unknown if they were in a sunny or shaded location, as reported by participants deploying the sensors (Fig. 1).
The sensors recorded hourly ambient air temperature over a
period of approximately 3 months (86 days). Of the 130 sensors, 11 also were equipped to measure and record hourly humidity data. The iButtons were deployed by volunteers who
were instructed to place the sensor in a relatively hidden location away from exposure to direct sunlight, and at eye level. The
sensors were affixed to surfaces with 3M heavy-duty Dual Lock
recloseable fasteners providing insulation from the surface. During sensor deployments, volunteers placed these iButtons in

as well as for designing strategies and policies to mitigate those

impacts. For example, in many regions the National Weather

Service (NWS) uses a single quantitative threshold for heat advisories and warnings, which can in turn activate resources in cities

and counties to help mitigate risks to citizens. However, this approach can miss inequities in heat impacts that are due to differences in exposure and vulnerability to heat across urban regions

and across different groups (Turek-Hankins et al. 2020). Further,

the official thresholds NWS uses do not always align with local

heat-health impacts (Vaidyanathan et al. 2019; Hondula et al.

2022). In some cases, those thresholds are being reconsidered

based on past performance and future conditions under a warming climate (Metzger et al. 2010; Guirguis et al. 2018; Benmarhnia

et al. 2019; McElroy et al. 2020; Vanos et al. 2020). This is particularly relevant as new federal policies are developed to

protect workers and communities from extreme heat (White

House Briefing Room 2021), many of which may rely on exceedance of certain heat thresholds. Hyperlocal-scale data

more closely tied to the heat exposure of individuals and

groups is critical.

There are a number of existing approaches to characterizing extreme heat within and across cities. Some studies have

used a downscaling approach applied to global data to produce validated maximum temperature and heat index data at

approximately 5-km scale (Funk et al. 2019). However, models operating at urban scales are still in relatively early stages

of development (Talebpour et al. 2021; Zheng et al. 2021) and

will require high-resolution data for verification. Remotely

sensed land surface temperature from thermal infrared sensors on a number of platforms is available at scales of tens to

100 m, which can provide a more complete snapshot of an

urban area at a particular time of day (e.g., Hulley et al. 2019;

Moffett et al. 2019; Hoffman et al. 2020). Yet, there are still

Denotes content that is immediately available upon publication as open access.

Corresponding author: Amy Clement, [email protected]

DOI: 10.1175/JAMC-D-22-0165.1

Ó 2023 American Meteorological Society. This published article is licensed under the terms of the default AMS reuse license. For information regarding

reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

JULY 2023 CLEMENT ET AL. 863

Unauthenticated | Downloaded 02/29/24 07:14 AM UTC

questions about how land surface temperature relates to air temperature, and cloud masking can be a significant limitation. Networks of experimental weather stations have been deployed in

some cities (Santamouris et al. 2007; Yang et al. 2020; Foissard

et al. 2019; Richard et al. 2018). There is a growing effort around

mobile sensors, where transects across an urban area can capture more accurate observations of heat (Shandas et al. 2019;

Kousis et al. 2021). These observations are made on impermeable surfaces (roads) and may not be at locations where people

are spending extended periods of time outdoors. Further, these

studies provide temporal snapshots and do not contain information about how extreme heat can vary over the day, the season

or from year to year.

To help fill data gaps on exposure, there has also been a growing role for data collected by community members in cities.

These include stationary sensors (Shi et al. 2021), as well as

urban monitoring sites based at people’s homes or using smartphones (Potgieter et al. 2021; Venter et al. 2020). There has also

been some exploratory work using smart phones (Cabrera et al.

2021) and wearable sensors (Uejio et al. 2018; Hass and Ellis

2019). Further, some studies are also attempting to quantify

indoor heat with similar technologies, and results indicate that

conditions indoors can be very different from outdoor, and in

many cases are much hotter (Vant-Hull et al. 2018; Ayanlade

et al. 2019; Kapwata et al. 2018; Teare et al. 2020). The advantage of these approaches is that heat data can be more easily

tied to impacts since sensors are monitoring in places where people spend time and are closer to personal exposure (Hondula

et al. 2021). They also can provide detailed temporal information about high heat days and times of day. For example, Shi

et al. (2021) showed that parts of Baltimore, Maryland, could experience “hot days” (threshold of 358C in their study) even

when the forecast did not indicate it. This illustrates the significance of having accurate, quantitative data at appropriate scales,

particularly when thresholds are used to trigger various actions

for mitigation of impacts.

Across the state of Florida, the NWS issues heat advisories

when the forecasted heat index exceeds 1088F (42.28C) for

more than 2 h in the forecast period with an 80% chance or

greater of occurrence, and an excessive heat warning for a

heat index of 1138F (458C) for the same amount of time and

chance of occurrence. This threshold is the same for all NWS

offices across the state and is based on a study of emergency

room admissions for several counties (Florida Department of

Health 2011). Advisories are rarely triggered in Miami–Dade

County, Florida, based on the area forecast. A historical analysis of NWS stations in the county is consistent with this lack of

advisories since those stations rarely report values of heat index in excess of 1088F (Florida Department of Health 2011).

Taken at face value, this would seem to suggest that heat exposure tied with health impacts has not been an issue requiring

mitigation actions in Miami–Dade County. This view is supported by studies that have shown that mortality data has no

clear link between heat and excess deaths in Miami–Dade

(Kalkstein and Greene 1997; Medina-Ramon and Schwartz

2007; Anderson and Bell 2011). Indeed, in Miami–Dade

County’s 2020 Local Mitigation Strategy (LMS; Miami–Dade

County 2020), extreme heat was considered a low impact

hazard in its Threat and Hazard Identification and Risk Assessment (THIRA) and not further considered in its LMS owing to only one extreme heat event reported between 1950 and

2019 (2% chance of occurrence per year).

Here we challenge the notion that exposure to extreme

heat, as measured by daily maximum heat index, is rare in

Miami–Dade County. We utilize hyperlocal temperature data

throughout the county to show that maximum heat exposure is

significantly greater in localized conditions of Miami–Dade

County when compared with NWS observations. We argue that

these hyperlocal data, more closely aligned with where people

are living and working, are a critical complement to NWS

station-based observations of potential heat exposure. This

localized information helps to fill a gap in communicating about

the dangers of heat exposure and the attribution of impacts that

can be used to guide regional decision-making and planning.

2. Methods

From September 2018 to January 2021, six iButton sensor deployments took place during different times of the

year (summer, fall, and winter). Temperature sensor deployments were organized through the Shading Dade initiative, a heat monitoring project in Miami–Dade County

launched by Florida International University’s Sea Level

Solutions Center in the Institute of Environment, and its

Department of Journalism and Media, as well as the nonprofit Catalyst Miami, and the University of Miami.

In total, 130 sensors were deployed and recovered; of those,

65 were in sunny locations, 54 in shaded locations, and 11 were

reported as unknown if they were in a sunny or shaded location, as reported by participants deploying the sensors (Fig. 1).

The sensors recorded hourly ambient air temperature over a

period of approximately 3 months (86 days). Of the 130 sensors, 11 also were equipped to measure and record hourly humidity data. The iButtons were deployed by volunteers who

were instructed to place the sensor in a relatively hidden location away from exposure to direct sunlight, and at eye level. The

sensors were affixed to surfaces with 3M heavy-duty Dual Lock

recloseable fasteners providing insulation from the surface. During sensor deployments, volunteers placed these iButtons in