cdot thermal mapping report · thermal mapping thermal mapping is a process by which the spatial...
TRANSCRIPT
CDOT
Applied
T THER
R
d Research
RMAL
Co
Report NoJu
h and Inno
L MAP
lin Walsh
o. CDOT‐2ne 2014
vation Bra
PPING
2014‐5
anch
G REPOORT
The contents of this report reflect the views of the
author(s), who is(are) responsible for the facts and
accuracy of the data presented herein. The contents
do not necessarily reflect the official views of the
Colorado Department of Transportation or the
Federal Highway Administration. This report does
not constitute a standard, specification, or regulation.
Technical Report Documentation Page
1. Report No.
CDOT-2014-5 2. Government Accession No.
3. Recipient's Catalog No.
4. Title and Subtitle
CDOT THERMAL MAPPING REPORT 5. Report Date
June 2014
6. Performing Organization Code
7. Author(s)
Colin Walsh 8. Performing Organization Report No.
CDOT-2014-5
9. Performing Organization Name and Address
Vaisala Inc Boulder Operations 194 South Taylor Avenue Louisville, CO 80027
10. Work Unit No. (TRAIS) 11. Contract or Grant No.
40.04
12. Sponsoring Agency Name and Address
Colorado Department of Transportation Applied Research and Innovation Branch 4201 E. Arkansas Ave. Denver, CO 80222
13. Type of Report and Period Covered
14. Sponsoring Agency Code
15. Supplementary Notes
Prepared in cooperation with the US Department of Transportation, Federal Highway Administration
16. Abstract
Thermal Mapping surveys were carried out on approximately 1000 miles of the Colorado Department of Transportation’s (CDOT’s) roads. The purpose of these surveys was to identify road surface variations across the network to determine whether forecast Thermal Maps or the data from the surveys would be useful to decision-makers in the CDOT regions. The distribution of road surface temperatures across the network and identifying Climatic Domains for the regions enabled Vaisala to look at current weather station locations and whether there were any gaps in coverage. Implementation
The use of forecast Thermal Maps is one that will be determined by the Boulder maintenance area in winter 2014/15. The Thermal Maps will be tested using Vaisala RoadsDSS and the final decision on whether the Thermal Mapping could be valuable to CDOT operations will be made in conjunction with the decision-makers prior to any statewide implementation plan. Vaisala has advised on potential weather station locations based on the Thermal Map findings which would complement the current setup. 17. Keywords
Climatic Domains, weather station locations, road surface temperatures (RSTs)
18. Distribution Statement
This document is available on CDOT’s website http://www.coloradodot.info/programs/research/pdfs
19. Security Classif. (of this report)
Unclassified 20. Security Classif. (of this page)
Unclassified 21. No. of Pages
46 22. Price
Form DOT F 1700.7 (8-72) Reproduction of completed page authorized
CDO
Co
OT TH
olorad
HERM
do De
MAL
epartm
©
ii
L MA
ment
June 2014© Vaisala 201
APPI
t of Tr
14
NG R
ransp
REPO
ortat
ORT
ion
iii
Published by:
Vaisala Ltd Phone (int.): +44 (0)121 683 1200
349 Bristol Road Fax: +44 (0)121 683 1299
Birmingham B5 7SW
United Kingdom
Visit our Internet pages at www.vaisala.com
© Vaisala 2014
This material is subject to copyright protection, with all copyrights retained by Vaisala and its individual partners. All rights reserved. Any logos and/or product names are trademarks of Vaisala or its individual partners. The reproduction, transfer, distribution or storage of information contained in this document in any form without the prior written consent of Vaisala is strictly prohibited. All specifications — technical included — are subject to change without notice.
iv
Table of Contents CHAPTER 1 .......................................................................................................................... 1
General information ................................................................................................. 1 About Vaisala ............................................................................................... 1
CHAPTER 2 .......................................................................................................................... 2
Introduction ............................................................................................................. 2 Thermal Mapping ......................................................................................... 2
Meteorological Conditions ........................................................................... 2
Thermal Mapping Results ............................................................................. 4
Thermal Mapping Applications .................................................................... 5
The Survey Network ...................................................................................... 5
Summary ....................................................................................................... 6
CHAPTER 3 .......................................................................................................................... 7
Terminology ............................................................................................................. 7
CHAPTER 4 .......................................................................................................................... 9
The Spatial Variations of Road Surface Temperatures ................................................. 9 Temporal Variations ..................................................................................... 9
Seasonal Variations .................................................................................... 10
Systematic Spatial Variations ..................................................................... 10
Summary ..................................................................................................... 15
CHAPTER 5 ........................................................................................................................ 16
Meteorological Conditions ...................................................................................... 16 Thermal Mapping Surveys .......................................................................... 16
Summary ..................................................................................................... 17
CHAPTER 6 ........................................................................................................................ 18
The Thermal Maps .................................................................................................. 18 Introduction ................................................................................................ 18
The Extreme Themal Map .......................................................................... 18
Extreme Map: Summary ............................................................................. 22
The Intermediate Thermal Map ................................................................. 23
Intermediate Thermal Map: Summary ....................................................... 24
The Damped Thermal Map ......................................................................... 24
Damped Thermal Map: Summary .............................................................. 25
The Thermal Maps: Summary .................................................................... 26
CHAPTER 7 ........................................................................................................................ 27
Climatic Domains .................................................................................................... 27
v
Introduction ................................................................................................ 27
Defining Climatic Domains ......................................................................... 27
Summary ..................................................................................................... 29
CHAPTER 8 ........................................................................................................................ 31
Weather Station Recommendations ........................................................................ 31 Locating Sensors ......................................................................................... 31
Sensor Designation ..................................................................................... 31
Ice Prediction / Ice Detection ...................................................................... 32
Weather Station Recommendations .......................................................... 34
Summary ..................................................................................................... 38
CHAPTER 9 ........................................................................................................................ 39
Next Steps .............................................................................................................. 39
C
G
A
V
m
in
A
T
c
m
T
d
T
d
u
T
p
m
y
a
T
a
CHAPT
Genera
About V
Vaisala is th
meteorologica
ndustry.
A truly intern
Traffic Safety
complete wea
many of the p
The thermal c
differ widely a
Thermal Map
distribution o
uniform weat
Thermal Map
providing the
make crucial
years through
accuracy and
To date Vaisa
across a world
TER 1
al Info
Vaisala
he world’s
al organizati
national comp
service incor
ather predicti
products with
characteristics
across a give
pping is the o
of warm and
her condition
pping highligh
winter main
decisions. Va
h continual in
repeatability,
ala has comp
dwide custom
ormatio
leading sup
ons, transpo
pany, Vaisala
rporates a ran
ion and mana
in this range.
s of any secti
n area. These
only techniqu
cold areas ac
ns.
hts potentiall
ntenance eng
aisala has rem
nvestment in
, and has see
pleted contra
mer network.
1
on
plier of me
ort authoritie
a serves custo
nge of produ
agement serv
on of road or
e differences
ue generally
cross a netw
y hazardous
gineer with a
mained at th
Research an
n subsequent
acts amountin
eteorological
es, research
omers in mo
cts from sim
vices. Therma
r runway acro
can be surve
accepted as
work is consist
sections acr
n enhanced
he forefront o
d Developme
t endorseme
ng to over 5
instrument
institutes, d
ore than 100
ple weather
al Mapping p
oss a network
eyed, quantif
being able
tent on each
ross a road o
information
of Thermal M
ent, regular s
nt in many Sc
500,000kms o
ts and serv
defence forc
countries. V
detecting dev
lays a centra
k are unique a
ied and categ
to confirm t
h winter nigh
or runway th
base from w
Mapping for o
statistical tes
cientific Journ
of Thermal M
ices to
ces and
Vaisala’s
vices to
l role in
and can
gorized.
hat the
t under
herefore
which to
over 25
ting for
nals.
Mapping
2
CHAPTER 2
Introduction
Thermal Mapping
Thermal Mapping is a process by which the spatial variation of minimum night time road surface
temperatures (RSTs) is measured using a high resolution infrared thermometer. The thermometer
is mounted in a specially equipped vehicle and connected to an automatic data logger. Readings
are taken along the road surface and recorded electronically at a rate of 10 readings per second.
Thermal Maps are produced using collected data to provide a representation of the relative
spatial variation of minimum RST under different weather conditions.
Meteorological Conditions
The prevailing weather conditions are the overriding large scale influence on RST. The scale of
variation in RST is dependent on the weather. Cloud cover reflects, absorbs and emits heat in the
form of radiation, and wind mixes air layers over the surface. Both these factors can effectively
reduce variations in RST. Within Thermal Mapping, weather conditions are categorized into three
scenarios: Extreme, Intermediate and Damped.
Extreme This scenario produces the greatest degree of variation in RSTs. It occurs when skies are clear and
wind speeds are very low, or even calm. The absence of cloud and wind allow for maximum
radiative cooling and thus heat loss from road surfaces. In addition, the lack of any significant
mixing of air layers allows katabatic drainage (cold air pooling) to occur and the generation of
temperature inversions as cooler ground surfaces chill the air immediately above them so that
the usual decrease of temperature with height is reversed. This weather scenario is usually
associated with anticyclonic (high pressure) systems, or high pressure ridges between cyclonic
systems.
3
NIGHT: OUTGOINGLONG-WAVE RADIATION(Overall energy balance)
CONVERGINGAIR
DIVERGING AIR
N. HemisphereClockwise & Outward
ANTICYCLONIC CONDITIONS
NIGHT: OUTGOINGLONG-WAVE RADIATION(Overall energy balance)
NIGHT: OUTGOINGLONG-WAVE RADIATION(Overall energy balance)
CONVERGINGAIR
DIVERGING AIR
N. HemisphereClockwise & Outward
ANTICYCLONIC CONDITIONS
CONVERGINGAIR
DIVERGING AIR
N. HemisphereClockwise & Outward
ANTICYCLONIC CONDITIONS
Damped This scenario represents the opposite end of the weather spectrum, consisting of total low level
(below 13,000ft) cloud cover and moderate or stronger wind speeds. This produces a significantly
reduced (or damped) variation in RST.
Similar results can also be obtained with lower wind speeds if conditions have remained cloudy
throughout the preceding day. This kind of weather is commonly associated with cyclonic (low
pressure) systems, which often bring associated bands of precipitation.
Intermediate Intermediate conditions are typically the result of either: a) medium to high level cloud cover
with the absence of significant wind, or b) clearer skies with moderate wind speeds. These two
situations produce similar results in that the degree of variation of RST is reduced from that found
under Extreme conditions but is greater than that found under Damped conditions. This scenario
can make up the bulk of winter weather and can be a result of various meteorological patterns,
such as transient ridges and frontal approaches.
WIND
DIVERGINGAIR
N. HemisphereCounter-clockwise & Inward
CONVERGING AIR
CYCLONIC CONDITIONS
WINDWIND
DIVERGINGAIR
N. HemisphereCounter-clockwise & Inward
CONVERGING AIR
CYCLONIC CONDITIONS
DIVERGINGAIR
N. HemisphereCounter-clockwise & Inward
CONVERGING AIR
CYCLONIC CONDITIONS
T
T
W
r
S
le
r
T
s
n
g
d
T
r
2
Thermal
The road/runw
Wherever po
epresent the
Sectors are fu
ength depend
educe the ch
The raw data
software. A m
noise and the
graphical rep
distribution of
Thermal Finge
elative distrib
2.0ºF bands u
l Mappi
way network
ossible Secto
network tha
urther split int
ding on the n
ance of chan
a collected d
mathematical
e data plotte
resentation
f RST over the
erprints are t
bution of RST
nder each of
ing Resu
is planned in
rs are plann
t can be surv
to a series of
network. The
ges in weathe
uring winter
Gaussian Filt
d as graphs
of the patte
e survey netw
then combin
T around the
the different
4
ults
nto Sectors, w
ned to oper
eyed by one v
f overlapping
e purpose of
er conditions
nights is do
ter is applied
known as Th
ern of RST a
work.
ed to produc
e average for
t survey cond
which are sub
ate within a
vehicle on an
Routes, each
splitting the
s affecting the
ownloaded an
d to the surve
hermal Finge
along the su
ce Thermal M
the road ne
itions.
bdivisions of t
areas of clim
ny one night.
h being appro
Sectors into
e data.
nd processed
ey data to eli
rprints. Thes
rveyed roads
Maps. These
twork survey
the overall n
matic similar
oximately 40
shorter rout
d through co
iminate any r
e Fingerprint
s and illustra
maps repres
yed in colour
etwork.
ity and
miles in
tes is to
omputer
random
ts are a
ate the
ent the
r coded,
T
T
s
s
t
n
F
p
O
f
d
T
T
s
H
in
c
a
D
D
Thermal
The Thermal
stations (if no
surface condit
he road wea
network can
Finally, where
practice, The
Optimization
or the road
different mete
The Surv
Thermal Map
survey netwo
However, due
naccessible fo
completed on
adjusting the
Drop‐Off). Da
Drop‐Off.
Figure
l Mappi
Maps can th
ot already in
tions and a w
ather outstati
also be prod
e anti‐icing
ermal Maps
of this nature
user by takin
eorological co
vey Net
pping was ca
ork was divid
e to the rec
or the Therm
n the data b
data for am
ta from exist
e 1 CDOT Ther
ing App
hen be used
place), whic
wide range of
ions and a m
duced for up
or deicing m
can be use
e aims to max
ng account o
onditions.
work
rried out alo
ded into five
ent floods to
mal Mapping s
by means of
mbient tempe
ting CDOT RW
rmal Mapping
5
plication
to locate th
ch use senso
meteorologic
meteorologica
to 24‐hours
materials (usu
ed to optimi
ximize the eff
of the Therma
ong approxim
e Sectors of
o the west o
surveys, leavi
a series of
rature chang
WIS stations
g network
ns
he optimum
or technology
cal paramete
al forecast, a
ahead using
ually salt) ar
ize the wint
ficiency of tre
al characteris
mately 1000
four Routes
of the surve
ing 4 Sectors
overlaps bet
ge during the
was also use
position for
y to measure
rs. Combining
prediction o
a computer
re used in w
ter maintena
eatment route
stics of the r
miles of road
s, making a
y area, US 3
. During the s
tween survey
e time of the
ed to cross‐re
r the siting o
e and monito
g the Therma
of RST over th
model and
winter maint
ance routes.
es to enhance
road network
ds in Colorad
total of 20
36 and US 3
surveys check
y Routes to
e surveys (kn
eference the
of RWIS
or RSTs,
al Maps,
he road
system.
tenance
Route
e safety
k under
do. The
Routes.
34 were
ks were
help in
nown as
rate of
6
The Surveys Thermal Mapping surveys were conducted across the network between October 1013 and
December 2013. The project specification was to collect data under Extreme conditions (calm and
clear) and Damped conditions (cloudy and windy), to illustrate the degree of variation present
across the region. As a result, Thermal Maps were produced displaying the variation of RST
around the average for each of the three weather scenarios.
Meteorological conditions often vary across an area on the same night with associated effects on
the distribution of RSTs. Therefore it is important to delineate relatively homogenous areas
where conditions could be expected to be similar. This leads to the identification of Climatic
Domains, which must also be considered when recommending or locating road weather stations.
Summary
Thermal Mapping is a technique that measures RST under key winter weather conditions, leading
to the production of isothermally colour coded maps depicting the distribution of RST over a
surveyed network.
The results provide invaluable information for highway engineers who deal with potentially
dangerous winter driving conditions. The value of Thermal Mapping is then greatly enhanced
when combined with road weather stations and meteorological forecasts under an ice prediction
or Decision Support system.
Thermal Mapping was completed in Colorado along I‐25, I‐70, I‐76, I‐225, US 36, US 287, US 85,
US 40 and US 6 between during the latter stages of 2013. This report discusses the results of
those surveys, including the production of the Thermal Maps.
7
CHAPTER 3
Terminology
Thermal Mapping The measurement of the spatial variation of RST along a road using an infrared radiation
pyrometer.
Thermal Fingerprints The graphical representation of RSTs plotted against distance, relative to the average RST for that
particular survey Route on that particular night.
Relative Average Not an actual temperature but the relative average for the road network covered by Thermal
Mapping surveys. All roads within the network are related to this average using different colour
bands.
Thermal Maps Maps of the distribution of RST under different meteorological conditions by isothermally colour
coded (2ºF) bands. These are related to the relative average RST for the region, which will change
from one night to the next according to prevailing air mass and weather system.
Climatic Domains Regions within a survey area that may experience differing meteorological conditions on any one
night.
Marginal Night Nights when the lowest RST for the region is forecast to be at or close to freezing.
Extreme Conditions Calm, clear nights typical of an anticyclone (high pressure system) producing the maximum RST
development profile along a survey route.
Intermediate Conditions Clear and windy nights, or calm nights with total medium level cloud present. Such conditions will
generally produce similar though less severe results than on an Extreme survey, with the
exception of frost hollows which may be greatly reduced or even totally absent.
8
Damped Conditions Windy nights with total, low level cloud cover and possibly precipitation typical of cyclonic (low
pressure) systems giving rise to a general flattening of the development of the RST profile.
Frost Hollow Under Extreme conditions cold air may descend under the effects of gravity (katabatic drainage)
and gather in small valleys or dips in the road. This pooling of air may produce the coldest RSTs in
the network.
Katabatic Drainage The “draining” of cold air down slopes under the force of gravity. Often causing Frost Hollows to
occur as cold air gathers in dips or valleys.
Mapping Sector The highway network is divided into Sectors. These are smaller, sub‐networks that can be
surveyed by one vehicle within a typical operational window of between midnight and sunrise.
Mapping Route A Mapping Sector is divided into Mapping Routes that are approximately 20 miles in length
(depending on the nature of the survey network) in order that the effects of changing weather
conditions upon surveys are minimized.
Road Surface Temperature (RST) The actual temperature of the road surface of any given point of the network.
Sky View Factor “The ratio of the amount of the sky “seen” from a given point on a surface to that potentially
available (i.e., the proportion of the sky hemisphere subtended by a horizontal surface)” OKE,
Boundary Layer Climates (1987) Sky view thus describes and quantifies the amount of
unobstructed sky/horizon that is visible at the surface.
Thermal Memory This is the term used to describe the length of time, which a highway structure, or any structure,
retains the stored heat, which it gains due to day‐time solar radiation. The Thermal Memory of a
section of road depends on the depth of construction, the construction materials used and the
amount of incident solar radiation received.
C
T
T
R
M
c
T
T
o
n
I
p
It
t
b
s
CHAPT
The Sp
Tempe
RSTs vary in b
Mapping proj
carried out wh
Tempora
Thermal Map
on any given
normally occu
mmediately a
part of the nig
t is therefore
hose recorde
both time an
surveys.
TER 4
patial V
erature
both space an
ect, as well a
hen the avera
al Varia
ping must be
n night. The
urs in the ea
after sunset,
ght RSTs chan
Figure 2
e possible to
ed on anothe
d distance),
Variati
es
nd time and t
s when interp
age RST is wit
ations
e conducted a
diurnal rhyt
arly afternoo
RSTs fall very
nge very little
Diurnal varia
directly comp
er. This comp
which reduc
9
ions of
his has to be
preting the re
thin the range
at night in or
hm of RST i
on and the m
y rapidly but t
.
ation at a wea
pare tempera
parison is faci
ces the amou
f Road
e taken into a
esults. Therm
e ‐4ºF to +50
rder to captu
is one in wh
minimum tem
this decline le
ather station
atures measu
ilitated by th
unt of surfac
d Surfa
ccount when
mal Mapping s
0ºF.
re the minim
hich the max
mperature oc
evels off so t
ured on one s
he use of sho
ce temperatu
ace
n planning a T
surveys are n
mum RST distr
ximum temp
ccurs around
hat during th
section of ro
ort survey Ro
ure change b
Thermal
ormally
ribution
perature
d dawn.
he latter
ad with
utes (in
etween
10
Solar Radiation Incoming solar radiation varies throughout the winter in proportion with the total amount of
daylight and the height of the sun in the sky. Minimum solar input occurs on the shortest day
(December 21st) but the actual incident solar radiation at one place is also dependent on two
variables: a) cloud cover and b) sky view factor.
Cloud Cover Clouds reflect and absorb solar radiation, thereby reducing the amount of direct solar radiation
reaching the surface. They absorb heat not only from above but also from below due to re‐
radiation from the earth's surface. This absorbed heat is then re‐radiated and at night this can
significantly offset surface cooling.
Seasonal Variations
The variation of the height of the sun in the sky and the angle of incidence of incoming solar
radiation means that the influence of sky view factor will change with the seasons. The sky view
factor will also be influenced by the amount of foliage left on the trees in winter. The shedding of
leaves by deciduous trees will allow more long wave radiation to escape at night, increasing the
rate of cooling. For these reasons, the distribution of RST can show slight differences at the
winter margins (e.g., April compared to December or January), when solar input is that much
greater due to increasing hours of sunlight and the increased angle of the sun in the sky. This can
cause some open road stretches with high sky view factors to absorb enough solar radiation
during daylight hours to offset their normally rapid cooling regime after sunset.
Latitude Latitude affects the length of the days and the winter, as well as the height of sun in the sky.
These are all important influences on the amount of solar radiation reaching the surface during
daylight hours. For example, a stretch of road that has a high sky view factor will receive greater
solar input. In higher latitudes, if all other factors are equal, such a stretch of road could be
expected to display relatively low RSTs due to unrestricted cooling after dark. However, in lower
latitudes where the angle of incidence of solar input is higher and days are longer, there may be
sufficient heat absorbed during the day to offset the cooling after dark. This must be taken into
account when examining the pattern and profile of RST distribution.
Systematic Spatial Variations
Although there are seasonal and diurnal variations to be considered, during the majority of the
winter months, the combined effect of buildings, topography and tree cover is generally
systematic at a single point. Each of these factors has a bearing on the distribution of RST but
they rarely act in isolation and the RST at any given point is a complex interaction of a number of
factors.
S
a
r
t
s
a
G
b
r
in
d
c
N
r
a
t
1
d
h
r
n
t
a
h
lo
MnMn
Sky VSky view facto
and is used to
adiation that
hat of the ac
sky is visible
are no obstr
Generally, the
building cove
adiation to t
nput thus no
depending on
cover, topogra
AltitNormally, the
esult of the
atmosphere. T
emperature w
10ºF per 100
decline with
hollows can
ecorded in v
nights due to
he pooling of
air, will tend
hollow will oc
ower RSTs. Th
Maximum sky view othing restricting
Maximum sky view othing restricting
View Factoor relates to
o determine t
t could theor
tual. It range
(e.g., inside a
ructions visib
e sky view fa
er, which r
the surface
ot only varie
n weather con
aphy and bui
Fi
ude higher the a
e decrease i
The environm
with height a
00m in altitu
altitude at
cause the
valley bottom
o either the
f cold air. Cold
to sink unde
ccur where a
he size of the
factor (1.0):the surfacefactor (1.0):the surface
or the amount
the maximum
retically occu
s from 0.0 wh
a tunnel) to
ble (e.g., an
ctor depends
educe the
via shading.
es seasonally
nditions, the
ldings.
igure 3 Effect
ltitude, the lo
n air tempe
mental lapse
above sea lev
ude. RST cou
a similar rat
lowest tem
ms, especially
formation of
d air, being d
r the influenc
slope is suff
e cold air poo
11
of “visible s
m incoming so
ur, compared
hen none of t
1.0 when the
open hillto
s upon tree a
incoming so
Solar radiat
but also da
amount of t
ts of sky view
ower the road
erature with
rate (the fall
vel) is usually
ld be expect
te. However,
mperatures t
on clear and
f inversions a
enser than w
ce of gravity
iciently steep
l is related to
Minimum sky vietotally restrict
Minimum sky vietotally restrict
ky”
olar
to
the
ere
op).
and
olar
ion
aily,
ree
factor
d minimum t
height that
l of air
about
ted to
, frost
to be
d calm
and/or
warmer
and this is c
p for such dra
o the length a
ew factor (0.0):ted surfaceew factor (0.0):ted surface
temperature
t occurs in
called kataba
ainage to tak
and steepness
expected. Th
a normal, u
tic drainage.
ke place, resu
s of the slope
is is the
unstable
A frost
ulting in
e. A cold
a
a
H
d
c
g
a
b
t
W
g
h
p
d
n
H
w
A
c
t
lo
t
t
s
t
t
It
t
t
c
s
s
s
s
f
o
air pool at the
a short, steep
However, the
difference in
circumstances
give cold tem
at lower altitu
between the
thermal belt.
WateWater reacts
generally take
heat for longe
proximity to s
due to the am
noticeable wh
However, wat
warming effec
TopoAt night a ro
cooling from t
he amount o
oss to the en
raffic and cut
he road back
surface and m
rees and/or b
unnels will te
t must also b
han more ex
he road surfa
conditions lea
surrounding f
sublimated b
shaded sectio
surface, RSTs
rost into ice.
of time than e
e bottom of a
slope as a re
e air at the b
height and th
s where ther
peratures at
udes. In these
cold hilltops
er Bodies more slowly
e longer to w
er than road
significant wa
meliorating e
hen wind spe
ter bodies m
ct.
ography oad surface c
the surface b
of long wave
nvironment is
ttings. Such f
k to the surfa
maintaining t
buildings, roa
end to stay wa
be remembe
xposed roads,
ace. This can
ad to the fo
fields. After s
y the incide
ons where sol
may remain
Hence, areas
exposed secti
a long, shallow
esult of the gr
bottom of th
herefore tem
e is sufficien
higher altitud
e situations th
and the col
y than land
warm, their re
surfaces. As
ater bodies (e
ffect of warm
eeds are gre
may freeze if
cools by radi
by influencing
radiation tha
s further redu
features refle
ce, thereby o
emperatures
ads running th
armer than m
red that she
, since early
be an impo
ormation of
sunrise, thes
nt solar rad
lar radiation
below freez
s with a low s
ons if the RST
12
w slope will b
reater volume
he short, ste
mperature bet
t relative rel
de and cold a
he warmest te
d valley bott
surfaces to
elatively poor
a consequen
e.g., coastal r
m air circulat
ater than 4m
ambient tem
ation. Topog
g the sky view
at can escape
uced by build
ct, absorb an
offsetting rad
. As a result,
hrough cuttin
more exposed
ltered roads
morning sola
rtant factor i
hoar frost o
se hoar frost
iation on ex
is unable to
ing and early
sky view facto
T falls below 3
be greater in e
e of cold air o
ep slope ma
tween top an
ief, the norm
air drainage w
emperatures
toms; this ph
changes in s
r thermal con
nce, during t
roads) can ex
ting off the w
miles/h as air
mperatures ar
graphy restric
w factor and t
e. The effect
dings, trees, c
nd re‐emit rad
iation loss fro
, at night roa
ngs, under br
d sections.
may warm m
ar radiation c
if, for examp
on the road
deposits are
xposed road
penetrate dir
y morning tra
or can be mo
32ºF.
extent than o
on the longer
ay be colder
nd bottom of
mal environm
will also lead t
are obtained
henomenon i
solar input. W
nductivity me
he winter mo
xperience rela
water. This ef
r layers are m
re sufficiently
cts radiative
thus limiting
of radiation
cloud cover,
diation from
om the road
ads lined by
ridges and in
more slowly
annot reach
ple overnight
surface and
e melted or
sections. In
rectly to the
affic can the
ore hazardous
one at the bo
slope.
due to the
f the slope. I
ental lapse r
to low tempe
d at middle al
s referred to
While water
eans that they
onths roads
atively warm
ffect is usual
more readily
y low to neg
n compact th
s for a longer
ottom of
greater
n some
rate will
eratures
titudes,
o as the
bodies
y retain
in close
mer RSTs
ly most
mixed.
ate any
he hoar
r period
R
f
e
T
t
o
F
S
t
R
e
p
v
RoadRoad construc
rom different
example, conc
The depth of
he warmer t
other roads b
Finally, the se
Spring when f
he daytime in
Road constru
effects of min
pooling by ra
variations and
BridgWhere a roa
result it will
time that a r
solar radiatio
The thermal
construction
and the am
received. Ce
over water,
radiation to
relatively wa
elevated via
to the effect
despite their
Steel bridges
and poor h
maintenance
d Construcction is anoth
t construction
crete roads a
construction
he road. As a
ut this can als
easonal variat
frost formatio
nput to offset
ction, particu
nor topograph
ising the roa
d also restrict
ge Decks ad crosses a b
possess a sm
road structur
on.
l memory de
n, the const
mount of in
rtain bridge d
may appear w
the undersid
armer water
duct sections
t of the urban
r limited dept
s normally co
eat retention
e as they can
tion her importan
n materials a
re generally w
n is also impo
a result, main
so be additio
tion in incide
on is still a ha
t the nighttim
ularly that of
hical features
ad above the
the sky view
bridge it is lik
maller therma
e, or any stru
epends on th
ruction mate
ncident sola
decks, particu
warmer as a r
de of the bri
surface. In u
s can remain
n heat island
th of construc
ool much qui
n. This low
produce sho
13
t influence o
t different ra
warmer than
ortant and ty
n highways a
nally influenc
nt solar radia
azard, sufficie
me cooling.
f highways a
. For example
e base of hol
factor of the
ely to be coo
al memory. Th
ucture, retain
he depth of
erials used
r radiation
ularly those
result of re‐
dge from a
urban areas
n warm due
and traffic,
ction.
cker than the
thermal mem
rt icy sections
n RST pattern
ates according
blacktop roa
ypically the g
and arterial ro
ced by traffic
ation must be
ent radiation
and other ma
e, embankme
llows and va
e road.
oler due to its
his is the ter
ns the stored
e road due to
mory can cr
s on an other
ns. Energy is
g to their the
ds.
reater the de
outes are no
flow volumes
e considered.
n may be stor
ajor roads, t
ents reduce t
lleys. Cutting
s shallower co
m used to de
heat that it g
o their high t
reate clear p
rwise safe roa
absorbed / r
ermal propert
epth of const
ormally warm
s.
. In late Autu
red in the roa
tends to redu
he effects of
gs reduce alt
onstruction a
escribe the le
gains due to d
thermal cond
problems for
ad.
radiated
ties. For
truction
mer than
mn and
ad from
uce the
cold air
titudinal
and as a
ength of
daytime
ductivity
winter
T
w
T
s
T
u
in
w
f
e
r
r
o
t
v
T
m
r
W
in
b
d
UrbaThe urban he
whereby the
The actual m
season and la
The temperat
urban and ru
ndustrial and
within the to
act that the
environment,
etain heat to
ural area. Th
of the constru
he urban arc
view factor.
The urban he
most noticeab
ural air with
Within the cit
nfluential on
budget is stro
diagram below
Urban He
entire D
an Heat Islaat island effe
built‐up area
agnitude of
nd use at tha
ture differenc
ral areas is t
d domestic h
wn or city, a
e fabric of
including
o a greater de
his is the resu
uction materi
chitecture an
eat island effe
ble when win
the warmer,
y, the heat is
RST than on
ongly recognis
w.
eat Island inte
omain at z =
and ect is the nam
can be seve
the heat isla
t location.
ce between
he result of
eat sources
allied to the
the urban
roads, will
egree than a
ult not only
ials but also
d lower sky
ect is stronge
nd speeds are
, urban air, t
sland effect m
n other non‐u
sable betwee
ensity over th
10m at 04:00
14
me given to th
ral ºF warme
and at any si
est in winter
e low. Highe
hereby reduc
means that to
urban roads.
en the hours
he
0
he phenomen
er than the su
ingle location
r, at its weak
r wind speed
cing the mag
opography, we
The urban m
of 18:00 and
Urban Heat
entire Dom
non observed
uburbs or sur
n in the city
kest in summ
ds promote t
gnitude of the
eather and tr
modification
06:00, which
Island intens
main at z = 10m
d in towns an
rrounding rur
will depend
mer and the e
he mixing of
e urban heat
raffic are usua
of the net ra
h is illustrated
ity over the
m at 04:00
nd cities
ral area.
on the
effect is
colder,
t island.
ally less
adiation
d in the
T
r
c
g
h
M
la
V
w
h
S
R
s
w
m
T
h
m
TraffTraffic tends
adiation. Tra
colder layers
generated by
helps to keep
Minimum RST
anes of up t
Vehicles tend
warmer than
high traffic vo
Summar
RST is a funct
systematic sp
will vary acco
maximum va
Thermal Map
highway engi
maintenance
fic to keep road
ffic also stirs
nearby and
vehicle tyres
roads with h
Ts can also va
o 4ºF can oc
to concentra
the outside la
olumes at nigh
ry
tion of a com
patial variatio
ording to the
riation and
ping measure
neers to use
decision proc
ds warm at n
s the air abov
again can lim
s and the gain
igher traffic f
ary across th
ccur due to t
ate in the ne
anes and slip
ht.
mplex relation
ns (e.g., sky
e prevailing
Damped con
es the spatial
as a tool to
cesses.
15
night, acting
ve the surfac
mit cooling o
n of radiant e
flows warmer
e lanes on In
the differenc
arside lane a
roads. This p
nship betwee
view factor a
weather con
nditions lead
l variation of
o supplement
like a shadow
ce, promoting
on calmer ni
energy by the
r than less he
nterstates an
ces in the vo
and as such a
phenomenon
en temporal v
and altitude)
nditions, with
ding to minim
f RST and disp
t local knowl
w, offsetting
g the mixing
ghts. In addi
e road from e
eavily trafficke
nd highway. D
olume of traf
at night these
is most signi
variations (e.
). The amplitu
h Extreme co
mum variatio
plays the resu
ledge and ex
the loss of h
of warmer a
ition, friction
ngines and e
ed roads.
Differences b
ffic using eac
e lanes are ge
ificant on roa
g., time of d
ude of the va
onditions lea
on. The pro
ults in map fo
xperience for
heat by
air with
nal heat
xhausts
etween
ch lane.
enerally
ads with
ay) and
ariation
ding to
cess of
orm for
r winter
C
M
W
k
w
T
T
w
c
c
T
e
n
D
D
A
t
CHAPT
Meteo
Weather is a
knowledge of
when interpre
Thermal
The size and l
wide range o
coupled with
conditions acr
This range of
even on the
necessary to
Domains. As
Damped weat
A summary of
he following
TER 5
orologi
a critical fact
f the actual c
eting the resu
l Mappi
location of th
of topograph
the lower p
ross the regio
local geograp
same night.
divide the re
mentioned
ther condition
f the readings
table.
ical Co
tor in decidi
conditions th
ults.
ing Surv
he Colorado n
hy with diver
plain regions
on.
phy gives the
In order to
egion into a
earlier in th
ns.
s taken for ea
16
onditio
ing when to
at prevailed
veys
network mea
rse character
s to the eas
e potential fo
ensure that
number of c
e report, su
ach night of t
ons
carry out a
during the s
ns that the s
ristics. The h
t frequently
or varying wea
the Therma
limatically sim
rveys were
the Thermal M
a Thermal M
survey period
surveys were
high altitude
lead to var
ather conditi
l Maps are t
milar regions
carried out
Mapping surv
Mapping surv
d are also ne
carried out a
areas to th
rying meteor
ons across th
thus applicab
s, known as C
under Extrem
veys can be f
vey and
ecessary
across a
he west
ological
he area,
ble it is
Climatic
me and
found in
17
TABLE 1: Survey Meteorological Conditions
Date Sector Weather Scenario
10‐15‐2013 1 Extreme
10‐16‐2013 1,2 and 4 Damped
10‐17‐2013 3 Damped
10‐18‐2013 3 Extreme
10‐21‐2013 4 Extreme
10‐22‐2013 2,3 Damped
10‐23‐2013 2 Extreme
Summary
Weather conditions are critical when interpreting the results of Thermal Mapping data. Surveys
were conducted under Extreme and Damped conditions to establish the different variations in
RST for each meteorological scenario.
C
T
In
T
h
c
Th
T
n
s
d
U
H
t
“
s
c
CHAPT
The Th
ntroduc
Thermal Maps
highway netw
colours for ea
he Extre
The Extreme
network. This
system, allow
development
TemUnder Extrem
However, this
he road and
“spikes”' the t
subsequent E
conditions.
TER 6
hermal
ction
s are a repres
work. Variatio
ch 6.0°F varia
eme The
Thermal Ma
s will occur u
wing maximu
of a tempera
Figure
perature Rme conditions
s includes the
d RST cooling
temperature
Extreme nigh
l Maps
sentation of t
ons in RST fr
ation:
ermal M
ap exhibits t
under calm, c
um radiative
ature inversio
e 4 Thermal M
Range s, the survey
e particularly
g is offset by
range is appr
ts, providing
18
s
the relative sp
rom the over
Map
the maximum
clear conditio
cooling of
on.
Mapping Key
yed roads ex
warm “spike
y a low sky v
roximately 25
the whole a
patial variatio
rall survey av
m nighttime
ons, typical o
the road, co
xperience an
es” of RST fo
view factor.
5ºF. This is th
area surveye
on of minimu
verage are d
RST variatio
of an anticycl
old air drain
approximate
und where a
Without the
he range that
ed experience
m RST of a su
denoted by d
on across th
onic (high pr
nage and th
e variation o
bridge cross
e influence o
can be expe
es uniform E
urveyed
differing
he road
ressure)
e likely
of 40ºF.
ses over
of these
cted on
Extreme
T
v
t
w
c
c
t
s
A
m
t
c
T
a
L
w
GeneThe wide ran
variations of R
o certain infl
west of the re
cut as may be
centers. This c
he highest on
surrounding s
Although the
many of the m
rue on I‐70
changes in roa
InfluThe normal tr
appear true i
Loveland Pass
whole region.
eral Patternge in altitud
RSTs apparen
uencing facto
egion. The dis
e expected b
can be seen o
n the networ
treet architec
vast majorit
more rural hig
in the south
ad constructio
ence of Alrend for RSTs
n the southw
s where temp
Figure 5 CD
rns of RST De and topog
nt in the Ther
ors. The colde
stinction ofte
but may be d
on I‐25 as it p
rk and can be
cture.
y of the war
ghways also
h‐east of the
on cause RST
titude & T is that the h
west of the R
peratures alo
19
DOT Extreme
Developmgraphy across
mal Maps. It
est RSTs gene
n associated
due to the na
passes throug
e directly attr
rmer RSTs are
display sectio
region, whe
Ts to remain w
opographyhigher the alti
Region, in par
ng both thes
Thermal Map
ent s the survey
is possible to
erally relate to
between rur
ature of high
gh Denver. T
ibuted to vol
e found with
ons of above
ere localized
warmer than
y itude, the co
rticular along
se parts of th
p
region is ref
o relate certa
o the higher
ral and urban
ways, which
The RSTs in th
lume of traffi
hin the dense
average RST
changes in s
average.
lder the road
g US 40 Berth
e network ar
flected on th
in RSTs distri
altitude area
areas is not
often bypas
his area are a
ic and density
ely populated
s. This is part
sky view fact
d surface. This
houd Pass an
re the lowest
he large
butions
s to the
as clear
s urban
amongst
y of the
d areas,
ticularly
tor and
s would
nd US 6
for the
W
c
is
Oattt
E
c
r
a
w
S
w
a
T
e
s
s
E
f
r
Within this o
complicated b
s a wide rang
On the aforemalways found he lowest RSopography whe coldest se
Even away fro
cold air draina
oad which a
adjacent secti
where small p
Sky VSky view facto
weather cond
amount of sky
The closer to
emitted back
stands of tre
surrounding a
Examples can
actor along t
roads in the
verall trend
by other facto
e of RSTs eve
Figure
mentioned roat the highesSTs of the netwithin these aections of RST
om the highe
age in the for
llow cold air
ions. There a
pockets of col
View Factoor is a crucia
ditions. Roads
y visible at an
the roadsid
to the road
ees along th
area.
be observed
this stretch o
vicinity.
the distribut
ors such as sl
en at these hig
e 6 CDOT Extre
oads the coldst altitudes. Ttwork due toltitude areas Ts on the netw
r altitudes th
rm of frost h
to pool, pro
re various ex
d air drainage
or l determinan
side features
ny one point.
e and the ta
surface, thus
he roadside,
US287 S Col
of the highw
20
tion of RSTs
ope aspect a
gher altitude
eme Thermal
er sections, mThe US6 as it o cold air draalso allows f
work.
e topography
ollows. These
oducing sectio
xamples of th
e lead to fluct
nt in the deve
s, such as tre
aller the feat
s offsetting co
RSTs are us
lege Avenue
ay leads to R
at these hig
nd cold air dr
s.
l Map US 6
more than 15passes the Loinage from afor cold air dr
y of the lands
e are dips an
ons where RS
his on I‐70 be
tuations in RS
elopment of
ees and build
ture, the mo
ooling after d
sually relativ
to the south
RSTs being a
gher altitudes
rainage. As a
5.0ºF below toveland Ski Aadjacent highrainage this c
scape can cre
d depression
STs appear o
etween Denve
STs.
RSTs, particu
dings can sign
re radiation
dark. Where
vely warmer
of Fort Collin
lmost 6.0ºF
s can becom
consequence
the average, rea displays sh ground. Whcan result in s
eate further a
s on the path
over 6ºF cold
er and Idaho
ularly under E
nificantly red
is reflected
there are sig
than those
ns. The high s
colder than
e more
e, there
are not some of here the some of
areas of
h of the
er than
Springs
Extreme
uce the
and re‐
gnificant
in the
sky view
n other
B
f
e
H
t
R
lo
T
a
r
T
s
in
W
A
d
Figure
Buildings act
actor, they ca
effect are mo
However, alth
he sun rises
RSTs over the
onger, presen
InfluThe main urb
average, parti
oad will offse
The general c
sources, and
ncreases the
Within Denve
Away from th
densities and
7 CDOT Extre
in a similar b
an also act as
re common t
hough trees a
these same
e entire netw
nting potentia
ence of Urban areas of
icularly in the
et heat loss fr
oncentration
higher traffi
shadowing ef
er in particula
he central ar
traffic volum
eme Thermal
but more exa
s heat source
hroughout th
and buildings
areas may be
work fall belo
al problems f
rbanizationDenver and
e centers. Wi
rom the road
of buildings
ic volumes c
ffect to furth
ar away from
eas the fall i
es.
21
Map US 287
aggerated wa
s, promoting
he more urba
reduce sky v
e the last to
ow freezing,
or morning c
n Fort Collins
thin the urba
surface.
will also emi
cause friction
er offset radi
m I‐25 RSTs a
in RSTs gene
ay to trees. A
g RSTs in the i
nized parts o
view factor an
receive sola
these areas
ommuter tra
both display
an centers, b
it radiation fr
nal heating f
ative losses.
re closer to t
erally corresp
As well as re
immediate ar
f the survey n
nd therefore
r input. In a
may remain
ffic.
y a concentra
uildings in clo
rom industria
from engines
the average,
ponds with d
ducing the s
rea. Examples
network.
offset cooling
situation wh
n below freez
ation of RSTs
ose proximity
al and domes
s and exhau
often falling
ecreases in b
ky view
s of this
g, when
here the
zing for
s above
y to the
tic heat
sts and
g below.
building
C
R
d
p
M
a
a
r
C
s
c
A
E
E
O
f
t
c
in
t
a
RoadChanges in ro
RSTs. Differen
differing relat
priority meet.
More localize
and elevated
adjacent sect
adiation. This
Consequently
surrounding
conditions.
Across the net
Examples incl
Figure 8 CDO
Extreme
Overall it may
rom well bel
he majority o
construction d
n RST. While
he relationsh
away from the
d Construcoad construct
nt constructio
ive RSTs, as is
d influences o
sections of
tions and acc
s is suppleme
, bridge dec
roads and s
twork there a
ude I‐25 as it
OT Extreme Th
e Map: S
y be seen th
ow to well a
of the roads w
depth and tra
there are som
hip at any one
e average, res
tion tion provide
on material,
s often the ca
of constructio
road are ge
cordingly hav
ented by heat
cks and eleva
should be c
are many exa
traverses the
hermal Map I
Summa
at RSTs on t
bove the rel
within the ne
affic flow), m
me dominant
e point is very
sulting in stre
22
a significant
age and surf
ase where tw
on changes c
nerally of a
ve a reduced
t loss from bo
ated sections
considered a
mples of brid
e South Platte
I‐25
ry
he roads ma
ative survey
etwork posses
any localized
t influences o
y complex and
etches of well
contribution
face dressing
wo roads of di
an be observ
much shallo
d capacity to
oth above and
s usually exh
potential i
dge decks bei
e River at Mil
pped within
average und
ss a number
factors com
on the RST dis
d involved, al
l above and w
to the varyi
g type across
iffering const
ved on bridge
wer depth o
o retain as m
d below the r
hibit relative
ce hazard i
ing colder tha
e High.
the region e
er Extreme c
of similar cha
bine to result
stribution, m
llowing RSTs t
well below av
ng character
s lanes will r
ruction mate
e decks. Bridg
of constructio
much daytim
oad surface.
ely lower RST
n adverse w
an adjacent se
exhibit a wide
conditions. A
aracteristics (
t in wide diffe
ost notably a
to develop m
verage RST.
istics of
esult in
erial and
ge decks
on than
me solar
Ts than
weather
ections.
e range
lthough
(such as
erences
altitude,
markedly
T
T
(
T
a
r
T
T
“
c
A
M
d
U
r
The Inte
The Intermed
>12km/h) or
These result i
air drainage
adiative cool
Thermal Finge
TemThe Intermed
“spikes”’ rela
change in wea
PatteAlthough man
Map, there is
distance of ro
Under Interm
eduction in
rmedia
iate Thermal
calmer, overc
n a narrower
and restricti
ing. The over
erprints and t
perature Rdiate Map ha
ting to featu
ather conditio
Figure 9 CD
erns of RSTny of the gen
less overall
ad falls into t
mediate condi
the effects o
te Ther
Map depicts
cast condition
r temperature
ing frost hol
rall result of e
therefore the
Range as an overal
ures such as
ons.
DOT Intermed
T Developmneral trends
development
the Light Blue
tions there i
of cold air d
23
rmal Ma
s the relation
ns where clou
e range due t
llow formatio
either of thes
temperature
l temperatur
bridges are
diate Thermal
ment of the Extre
t around the
e (below aver
s a greater m
rainage. This
ap
nship of RSTs
ud is medium
to increased
on, or cloud
se scenarios i
e range of the
re range of a
discounted. T
Map
me Map are
average RST
age) and Red
mixing of air
s is evident a
that develop
m to high leve
mixing of air
d cover redu
is to reduce t
e road surface
approximate
This reductio
e borne out o
T. As a conseq
d (average) ca
layers and a
along I‐70 be
p under clear
l.
layers reduci
ucing the eff
the amplitude
e.
ly 15.0ºF, on
on is a result
on the Interm
quence a far
ategories.
s a result th
etween Denv
r, windy
ing cold
fects of
e of the
nce any
t of the
mediate
greater
ere is a
ver and
24
Silverthorne. Under Extreme conditions these valley roads displayed RSTs well below the average.
However, with the reduction in cold air drainage RSTs on these roads appear up to 6ºF closer to
the average than they appear on the Extreme.
Other, smaller frost hollows often become less apparent or even non‐existent under Intermediate
conditions, depending on their particular location and exposure to the mixing of air layers.
The importance of some features show under both Extreme and Intermediate conditions. Bridge
decks for example tend still to display lower RSTs than surrounding roads as can again be seen by
on I‐25 bridges particularly to the south of Denver.
Areas of concentrated low sky view factor also tend to retain their temperature profile and often
remain above average and this is once again demonstrated in the central areas of Denver and
Fort Collins.
Altitude also remains crucial and the coldest stretches of road on the survey network remain the
higher relief areas. For example, Berthoud Pass and Loveland pass still display RSTs of over 9.0ºF
below the average, reaching into 12.0ºF on some sections.
Intermediate Thermal Map: Summary
The distribution of colour bands on the Intermediate Thermal Map is generally consistent with
those on the Extreme Map. However, the influence on RST development of many localized factors
are not as marked as under Extreme conditions, with an apparent shift toward the relative
average as a result. The prime factor in the relative reduction of RST development is the influence
of wind, which mixes the air layers above the road surface and dampens out extremes of
temperature.
The Damped Thermal Map
The Damped Thermal Map demonstrates the pattern of RSTs that develop with increased cloud
cover (especially under 2000m) that offsets radiative cooling of the road surface, and higher wind
speeds that mix the air layers and reduce temperature inversions.
Consequently the importance of localized features is reduced while larger effects such as
proximity to the sea and changes in altitude are dominant. As a result, the Damped Thermal Map
displays a greatly reduced variation in RST from the overall average. The majority of the road
network falls within 4.0ºF either side of the network average.
Temperature Range Under Damped Mapping conditions very few spikes occur in the data where bridges pass over the
road, or at any similar locations. Overall, the variation in RST is approximately 15ºF.
T
f
a
t
A
a
w
c
D
U
a
c
h
f
s
PatteThe overall tre
ew instances
average RST.
emperature w
As a result th
again illustrat
which continu
continue to di
Damped
Under Dampe
average) rang
consequence
higher wind s
actors reduce
such as chang
Figure 1
erns of RSTend under Da
s factors aside
The most
with height.
he colder are
ted on the w
ue to display
isplay the ma
d Therm
ed conditions
ge, with the m
of cloud cov
peeds mixing
e the importa
ges in altitude
10 CDOT Dam
T Developmamped condit
e from the w
dominant of
eas of the reg
western sectio
the coldest R
jor concentra
mal Map
s RSTs fall in
majority of th
ver affecting
g the air layer
ance of locali
e.
25
mped Therma
ment tions is for RS
weather, will i
f these is al
gion are once
ons of the ne
RSTs under Da
ations of abov
p: Summ
nto either th
he road surve
the radiative
rs to inhibit t
zed features
l Map
ST to fall into
nfluence the
ltitude, as th
e again foun
etwork, in pa
amped condi
ve average RS
mary
e Green (ave
eyed falls int
e cooling of t
temperature
and promote
o the Red (ave
developmen
here continu
d at the high
articular on t
itions. The m
STs.
erage) or Lig
o the averag
the road surf
inversions. T
e the significa
erage) catego
nt of above o
ues to be a
her altitudes.
the US 40 an
ain urban are
ght Blue (just
ge category. T
face, combine
hese meteor
ance of other
ory. In a
r below
fall of
. This is
nd US 6
eas also
t below
This is a
ed with
ological
r affects
26
The Thermal Maps: Summary
There is an extremely complex relationship between many factors that affect the distribution of
RSTs. This relationship is at its most complex when there is no cloud to offset radiative cooling
and no wind to mix air layers (i.e., under Extreme meteorological conditions).
However, the temperature relationship between sections of road under different constant
weather scenarios is generally consistent, although the magnitude of variation in RST differs
depending on the prevailing meteorological conditions.
Thermal Maps are representative of the relationship between RSTs, assuming constant
meteorological conditions across the whole of the area surveyed. However, this is not always the
case in practice so it is important to identify any required Climatic Domains prior to using the
Thermal Maps operationally.
27
CHAPTER 7
Climatic Domains
Introduction
The Thermal Maps constructed from the data collected during the winter of 2013 relate to the
various weather scenarios previously outlined. This assumes that the meteorological conditions
are constant across the region. This is likely to be true of a number of winter nights, particularly
when anticyclonic conditions prevail, but there are many nights on which the weather will vary
from one part of the region to another, allowing some roads to cool more rapidly than others
where skies are clear, and offset cooling where cloud persists.
It is therefore possible that while one part of the region may be experiencing Extreme conditions,
another may be experiencing Intermediate, or even Damped conditions, due to a changing
meteorological situation. This variability in weather depends on a number of factors, not least the
size of the area in question and its position with regard to major upland areas and coastline.
Defining Climatic Domains
In order to represent this variation it is necessary to subdivide the road network into smaller,
discrete areas referred to as Climatic Domains. The boundary of a Climatic Domain and the
number of climatic areas covering a particular network is dependent upon the proximity of
geographical features likely to influence the weather over the network. These influences are
generally threefold:
1. Proximity to large water bodies such as oceans, large lakes and rivers, which provide a
ready source of moisture and therefore have a tendency to produce cloud in the air
mass passing over them.
2. Proximity of high ground, which forces air to rise over it making the air unstable and
often producing cloud in the process.
3. Prevailing and dominant wind directions during the winter months ‐ which dictate the
extent of the influence of water bodies and high ground on any particular night.
Example Climatic Domains Figure 11 illustrates a typical example of how changes in topography can lead to variations in
localized weather conditions. Air that is forced to rise can become unstable and lead to cloud
formation, particularly if it has passed over a water body and become moist.
D
b
b
D
m
e
ic
I
p
h
t
F
e
s
F
a
a
T
w
Domain 1 mig
breezes at nig
body to provid
Domains 1 an
moisture lade
experience pr
ce coverage.
n Domain 4,
precipitation
higher ground
o less restrict
Finally, in Dom
existent. This
susceptible to
From this, it c
a crucial dete
account.
CDOThe CDOT net
will have an im
DOMA
ght typically
ght. During th
de an amelio
nd 2 have th
en air is force
recipitation in
as the air d
less likely. In
d, leading to
ted radiative
main 5 the a
allows unre
o fog if there i
can be seen h
rminant of m
T Climatic twork incorp
mportant influ
AIN 1 DO
Figure 11 Ex
experience w
he Winter mo
rating effect t
he potential f
ed to rise ov
n the form of
descends fro
n addition, th
a rain shadow
cooling of the
ir has becom
estricted cool
is localized so
how weather
minimum RSTs
Domainsorates a wide
uence on the
OMAIN 2
28
Example Clima
warm, moist
onths, there
to offset nigh
for thick clou
ver the land m
f snow and in
om the high
he moisture
w effect. Ove
e road surfac
me considerab
ing of the ro
ources of moi
conditions ca
s, it is clearly
e range of alt
meteorologi
DOMAIN 3
atic Domains
onshore win
may be suffi
httime radiati
ud and the p
mass. Domai
n extreme sit
ground it be
in the air m
erall, this allo
ce.
bly more stab
oad surface
sture.
an vary withi
important fo
titude, topog
ical condition
DOMAIN
nds during th
cient stored
ve cooling on
possibility of
in 3, if at suf
tuations have
ecomes more
ay have bee
ows cloud cov
ble and cloud
but may also
n the same r
r these variab
graphy and la
ns.
N 4 D
e day with o
energy in the
n the land.
precipitation
fficient altitu
e perpetual s
e stable and
n exhausted
ver to break,
d cover may b
o leave this
region. As we
bles to be tak
and use, all o
DOMAIN 5
offshore
e water
n as the
de may
snow or
makes
on the
leading
be non‐
Domain
eather is
ken into
of which
29
The Domains have also been delineated with the current number and location of Forecast
Stations already installed in Colorado in mind. Historical weather information has also been
utilized to develop the Domain boundaries.
Climatic Domain I: Mountain
This Domain covers the division of the continental divide from the far west of the network to
the east of Idaho Springs. Due to the high altitudes in this Domain, significant snowfall can be
expected during the winter months.
Climatic Domain II: Foothills
This Domain stretches from the far south of the network to the state‐line. It incorporates all the
front‐range roads which can be susceptible to snowfall because of the higher altitudes.
Climatic Domain III: West and South Metro
This Domain again stretches all the way from the north to the south of the network it extends
eastwards from the Foothills to east of the I‐25 corridor. From historical data, this Domain
experiences snowfall and weather fronts which differs from the Foothills in intensity but is
significantly more than further east of the I‐25 corridor.
Climatic Domain IV: Eastern Plains
This is the largest Domain in the region and is defined by all roads to the east of the I‐25 corridor
and as far south as the higher altitudes of Palmer Ridge.
Climatic Domain V: Palmer Ridge
Located in the south of the region, the higher altitudes here warrant a separate Domain to both
the Eastern Plains and Metro Domains. Dependent on the prevailing weather front,
temperatures here could be much colder than surrounding Domains and it will not be affected
as much from the mountainous relief to the west of the region
Summary
The survey network may experience differing weather conditions in different areas on the same
night, but the Thermal Maps have been produced for constant meteorological conditions.
To account for the possibility of changing weather conditions, the current survey network has
been divided into five regions, or Climatic Domains. While these Domains have been identified
from experience during surveys and topographical maps, they may not encompass the full range
of climatic areas in the region. They are intended for use as a general guide and are open to
change with the benefit of the knowledge of local highway engineers or meteorologists.
30
It is possible that the boundaries of the Domains may migrate slightly from night to night,
depending on local conditions (such as wind speed and direction) but their overall position is
representative of the major climatological influences.
31
CHAPTER 8
Weather Station Recommendations
Sensors can provide information about RSTs at selected points within an area. Related to the
Thermal Maps, this allows the maintenance engineer to know precisely what is happening at any
location in the Thermally Mapped network.
Locating Sensors
To obtain maximum use by the maintenance engineer, sensors should be sited in order to
represent a cross‐section of temperature regimes and a good geographical spread across the
area, highlighting some of the following:
1. Climatic differences across the road network
2. The range of temperature encountered from the Thermal Mapping
3. The geographical extent of the road network
4. Large temperature anomalies (e.g., large bridge decks or overhanging vegetation)
5. Areas of porous road surface
Depending on the focus of winter maintenance operations it is generally not advisable to locate
all stations in the coldest spots since this can lead to over‐treatment of those areas that remain
above 0°C on marginal nights. Locating sensors in a variety of temperature regimes will provide a
representative picture of the total temperature range over the road network on a particular
night. The larger the geographical area covered by the road network, the greater the possibility
for weather change on the same night and the greater the need to divide the region into smaller
Climatic Domains.
Sensor Designation
If there is an ice prediction system planned or in use, there are two possible designations for
weather stations: Forecast Stations, and Non‐Forecast Stations. For an ice prediction system,
there should be one Forecast Station per Climatic Domain and ideally one Non‐Forecast Station to
act as a back‐up and to give supplementary information.
Forecast Stations are used to drive the forecast model with the ice prediction system. They
should be sited on stretches of road that are representative of their Climatic Domain and also in
areas that demonstrate stable RSTs across all the weather scenarios (Extreme, Intermediate &
Damped).
N
s
p
b
a
o
a
c
n
n
Ic
T
in
h
t
p
m
T
o
t
m
a
e
m
m
I
t
d
M
r
t
t
c
lo
S
s
a
Non‐Forecast
significant shi
problem area
be two Non‐F
all weather st
obstructions s
an accurate
complete roa
nights on whi
network to an
ce Predi
The advanta
nstalling an
highway engin
he highway
precautionary
materials.
The technique
of minimum
emperature f
meteorologica
apply to their
existing probl
may also be a
moisture sour
n addition, t
raffic flows. S
distinct from
Mapped high
apidly than o
hese section
iming of trea
capacity, Ther
ocation on th
Stations can b
stations but a
avoiding “spik
Stations can
fts in relative
s or accident
Forecast Stati
tations should
such as trees
forecast of
d network o
ch the weath
nother.
iction /
ge of comp
ice predictio
neer a fuller
y network
y measures
e of Thermal
RST distribut
for each sect
al inputs. If st
r location. Fo
em areas, wh
dvisable to lo
rces, to give w
here may be
Site specific
Forecast Th
way network
others (such
s in real tim
atment. Rega
rmal Mapping
he network.
be sited used
also ensure t
ky” or noisy F
be sited in a
e RST betwee
t blackspots.
ons per Clim
d be located
s or buildings
RSTs to be
on any partic
her will vary f
/ Ice Det
pleting Ther
n system is
picture of wh
and allow
such as
Mapping me
tion across t
tion of highw
tations are fo
or example,
here accident
ocate stations
warning of po
e more huma
Forecast Gra
hermal Maps
k. There will
as bridge de
e in order to
ardless of wh
g provides vit
d Thermal M
hat they are
ingerprint are
32
areas of pote
en Thermal M
Ideally, there
atic Domain.
away from ro
s. This will al
e produced f
ular night, in
from one par
tection
rmal Mappin
that it will g
hat is occurri
him/her to
applying an
eans that few
the network,
way using info
or detection/w
it may be th
ts due to ice
s where the l
otential hazar
an or subject
phs show the
s, which sho
be some se
ecks) and in
o contribute
hether Statio
tal informatio
apping in ord
positioned i
ea that may s
ential variabil
Map type (e.g
e should
Ideally,
oadside
llow for
for the
ncluding
rt of the
ng and
give the
ng over
o take
nti‐icing
wer road senso
, as the ice
ormation gath
warning only
hat weather s
formation ha
owest RSTs a
rds.
tive consider
e forecast pr
ow the minim
ections of hig
some instanc
to the infor
ons will be us
on on the spe
der to provid
n areas with
skew sensor i
lity, i.e., thos
., frost hollow
ors are requi
prediction m
hered from w
y, then differe
stations will
ave been a fe
are found and
rations relate
rogress of RST
mum RST ac
ghway surfac
ces it can be
rmation used
sed in a pred
ecific therma
de both the o
stable Finge
nterpretation
se areas whic
ws) and/or in
red to give a
model calcula
weather statio
ent criteria m
be better pl
eature in the
d/or where th
ed to issues
T at single po
cross the Th
ce which coo
e helpful to m
d in deciding
diction or de
l characterist
optimum num
erprint profile
n).
ch show
known
picture
ates the
ons and
may well
aced in
past. It
here are
such as
oints as
ermally
ol more
monitor
on the
etection
tics of a
mber of
es (e.g.,
33
Weather stations combined with Thermal Mapping then give an overview of the entire network
RSTs rather than just receiving site specific information from individual sensor locations.
TABLE 2: Existing CDOT Weather Stations on the Thermal Mapping Network
SITE NAME HWY DIRECTION MP Latitude Longitude Altitude (m)025N190 SURREY RIDGE 025 N 189.45 39.486560 -104.871990 1946
025S185 FOUNDERS POINT 025 S 185.1 39.424830 -104.876720 1895
025S191 S. of Ridgegate 025 S 191 39.510482 -104.871270 1861
070E216.6 Loveland Pass Exit 070 E 216.7 39.687858 -105.882933 3238
070W205 Silverthorne 070 W 205.7 39.630650 -106.062820 2706
070W210 Lower Runaway Truck Ramp 070 W 209.79 39.656640 -105.996013 3038
070W213 EJT West Portal 070 W 213.3 39.676389 -105.943231 3380
070W218 Hermans Gulch Exit 070 W 218.1 39.701779 -105.861435 3157
070W221 Bakerville Exit 070 W 221.1 39.692206 -105.808015 2993
070W246 Floyd Hill 070 W 246.5 39.720928 -105.410115 2394
070W253 Genesse 070 W 253.60 39.710530 -105.294340 2370
070W304 BENNETT 070 W 304 39.738670 -104.433270 1689
070W348 CEDAR POINT 070 W 348.3 39.356990 -103.870130 1768
025N242 St Vrain 025 N 241.2 40.175571 -104.979362 1475
025N259 Crossroads Blvd 025 N 259.3 40.433370 -104.991810 1542
025N276 Wellington 025 N 276.4 40.682936 -105.000127 1573
025S251 Berthoud 025 S 251.1 40.321120 -104.980640 1514
025S298 Natural Fort 025 S 298.1 40.987810 -104.911760 1837
034W070 Cedar Mont 034 W 69.7 40.392191 -105.427581 2171
036W038 Baseline Road 036 W 38 40.001375 -105.259267 1637
076E037 Keenesburg 076 E 36.5 40.110230 -104.565763 1525
076W067 Wiggins/Bijou Creek 076 W 67.1 40.248500 -104.037560 1381
287 @ SE 19th St 287 N 331 40.369243 -105.072462 1543
025N194 C-470 EAST 025 N 194 39.556465 -104.870033 1798
025S194 C-470 WEST 025 S 194 39.556206 -104.871830 1810
025N213 I-70 EB/WB 025 N 213 39.779698 -104.988916 1560
025S213 I-70 West 025 S 213 39.779868 -104.990864 1580
025S229 SH-7 025 S 229 40.000169 -104.981524 1581
070W260 Rooney Rd 070 W 260 39.712470 -105.194760 1885
070W276 Colorado Blvd 070 W 276 39.780235 -104.943331 1603
070W283 Chambers 070 W 283 39.771518 -104.810140 1644
225N004 Parker Road 225 N 4 39.658749 -104.842258 1717
225N007 Mississippi Ave 225 N 7 39.697122 -104.828713 1702
W
F
C
g
T
T
in
f
s
c
Weather
Figure 12 Curr
CDOT has com
gaps in that c
Thermal Map
Thermal Map
nstalled. With
rom Palmer
surveys shou
communicatio
r Statio
rent CDOT sta
mprehensive
coverage. Tab
ping network
pping carried
h regard to Fo
Ridge Doma
ld be suppl
ons and powe
n Recom
ations (station
weather stat
ble 2 above l
k. Due to the
out across t
orecast statio
ain. Recomm
emented by
er.
34
mmend
ns actually on
tion coverage
lists the exist
size of the T
he region th
ons there is en
mendations fo
y closer site
dations
n the Therma
e and this rep
ting weather
Thermal Map
here may be
nough potent
or station lo
inspections
al Mapping ne
port will inves
stations tha
ped region a
scope for fu
tial coverage
ocations from
to validate
etwork in red)
stigate any p
t are specific
nd the exten
rther station
in all Domain
m Thermal M
the availab
)
otential
c to the
t of the
ns to be
ns apart
Mapping
bility of
C
L
n
L
c
Cl
T
o
c
Climatic Do
Figure 13 R
Located on U
network and w
Figure 14
Located on U
coldest of the
imatic Dom
The only secti
of road is ade
completed the
omain I: M
Recommended
S 40 in the a
would benefit
Recommende
S 6 to the ea
whole netwo
main II: Fo
ion currently
equately cove
e data collect
ountain
d Non‐Foreca
area of milep
t from a mon
ed Non‐Forec
ast of the Lov
ork
oothills
on the Therm
ered by statio
tion on US 36
35
ast Station 1
post 236, this
nitoring statio
cast Station 2
veland Ski Ar
mal Map for t
ons at the m
6 and US 34
s section is o
on.
2
rea entrance,
this Domain i
oment but w
one of the co
, this section
is the section
we may revisi
oldest on the
n is again one
of I‐70. This
it this once w
e entire
e of the
section
we have
Cl
L
t
le
C
L
5
imatic Dom
Figure
Located on I‐2
han other se
evel of traffic
Climatic Do
Figure 1
Located on US
5°F colder tha
main III: W
15 Recomme
25 at Mile Hig
ections in the
c in this area.
omain IV: E
6 Recommen
S 85 north of
an surroundin
West and S
ended Non‐Fo
gh as it trave
e vicinity and
Eastern Pla
nded Non‐Fore
La Salle as it
ng sections.
36
outh Metr
orecast Statio
rses the Sout
would benef
ains
ecast Station
crosses the S
ro
on 1
th Platte Rive
fit from a mo
1
South Platte R
er. This sectio
onitoring stat
River. This sec
on is relatively
tion due to t
ction of road
y colder
he high
is up to
L
o
C
T
H
is
L
d
r
Figure
Located on US
of road is up t
Climatic Do
The size of th
However, it m
ssues with th
F
Located on t
displays stabl
epresentative
17 Recomme
S 85 to the ea
to 5°F colder t
omain V: P
he current ne
may be bene
e current stat
Figure 18 Reco
he eastboun
e RSTs about
e of the Dom
nded Non‐Fo
ast of Greeley
than surroun
Palmer Rid
etwork that i
ficial to have
tion at Cedar
ommended F
d highway a
t 300 yards e
ain.
37
recast Station
y as it crosses
ding sections
dge
is part of thi
e another for
Point.
Forecast Statio
pprox. 2.3 m
either side o
n 2
s the Cache L
s.
s Domain do
recast station
on 1
miles east th
of the exact l
La Poudre rive
oes not warra
n in place in
e Limon off
location. The
er. Again this
ant another
case there a
ramp, this l
e RSTs here a
section
station.
are any
ocation
are also
38
Summary
The Thermally Mapped network has been split into five Climatic Domains to take account of any
variation in weather conditions on winter nights that could affect the temperature profiles and
any forecasts within an ice prediction system. Six weather stations have subsequently been
recommended (one Forecast sites within each Climatic Domain, along with a back up Non‐Forecast
weather station) to provide a comprehensive coverage and full picture of RSTs across the
Thermally Mapped network during the winter months.
39
CHAPTER 9
Next Steps
Obtain forecasts for each CDOT Domain
Identify maintenance area to trial Forecast Thermal maps and report back ‐ this will involve
support and training from Vaisala
Identifying potential improvements to aid with winter treatment decisions
Following trial period, assessing the benefits of Thermal Mapping and how it can be implemented
statewide
Utilizing the Thermal map data in the forthcoming route optimization research project.
Utilizing the Thermal Map data in the upcoming performance measurement research project.
One of the major benefits of Thermal Mapping is that it can be used in a Road Forecasting system
to extrapolate site specific information to the entire network. For that reason, CDOT will have
access to the Thermal Maps in Vaisala’s Navigator system to assess the benefits for winter
treatment operations.