AmericanLifelinesAlliance

A public-private partnership to reduce risk to utility and transportation systems from natural hazards and
manmade threats

Extreme Ice Thicknesses from

Freezing Rain

September 2004
Extreme ice thicknesses from freezing rain Final Report

AmericanLifelinesAlliance
A public-private partnership to reduce risk to utility and transportation systems from natural hazards and
manmade threats

Extreme Ice Thicknesses from

Freezing Rain

September 2004

www.americanlifelinesalliance.org

This report was written under contract to the American Lifelines

Alliance, a public-private partnership between the Federal Emergency
Management Agency (FEMA) and the National Institute of Building
Sciences (NIBS). This report was reviewed by a team representing
practicing engineers, academics, and state climatologists.

September 2004 Page ii

Extreme ice thicknesses from freezing rain Final Report

Acknowledgements

The following people (with their affiliations) contributed to this report.

Project Team:
Kathleen F. Jones, Project Manager CRREL, Hanover, NH
Neal Lott National Climatic Data Center, Asheville, NC
Ronald Thorkildson Bonneville Power Administration, Vancouver, WA

Advisory Committee:
A draft report was provided for review to the State Climatologists of the contiguous 48 states and
Alaska, with the exception of states currently without climatologists (Montana, Rhode Island,
and West Virginia) and State Climate Offices (SCOs) without current email addresses (Arkansas,
Massachusetts, Nevada, and Virginia). Personnel at 27 of the remaining 42 SCOs provided
reviews. The thoughtful and informative comments and suggestions from personnel associated
with the SCOs from:

Alabama Iowa New Hampshire Oklahoma

Colorado Kansas New Mexico Oregon
Delaware Michigan New York Pennsylvania
Florida Minnesota North Carolina
Idaho Mississippi North Dakota
Illinois Nebraska Ohio

were used in preparing this final report.

ALA Oversight Committee:

Douglas G. Honegger D.G. Honegger Consulting, Arroyo Grande
Joseph Steller American Lifelines Alliance, Washington, D.C.

September 2004 Page i

Extreme ice thicknesses from freezing rain Final Report

Table of Contents

1.0 Introduction......................................................................................................... 1
1.1 Project Objective .............................................................................. 1
1.2 Project Scope ................................................................................... 1
1.3 Notation ............................................................................................ 2
2.0 Determining extreme ice thicknesses from freezing rain ............................... 3
2.1 Ice accretion models......................................................................... 3
2.2 Equivalent radial ice thickness ......................................................... 3
2.3 Application to weather data .............................................................. 4
2.4 Extreme value analysis..................................................................... 4
References................................................................................................................... 11
Acronym List ............................................................................................................... 13
Appendix. Maps for mean recurrence intervals of 50, 100, 200, and 400 years..... 15
Appendix A English units…………………………………………………….A-1 of 21
Appendix B Metric units……………………………………………………...B-1 of 21

September 2004 Page ii

Extreme ice thicknesses from freezing rain Final Report

List of Tables

Table 2-1. Factors on 50-yr thicknesses ......................................................................... 5

September 2004 Page iii

Extreme ice thicknesses from freezing rain Final Report

List of Figures

Figure 2.2-1 Some shapes of ice accreted from freezing rain on cylinders .................... 7
Figure 2.2-2 Equivalent radial ice thickness .................................................................... 8
Figure 2.3-1 Weather stations in contiguous 48 states.................................................... 9
Figure 2.3-2 Footprints of damaging ice storms 1948-2002 .......................................... 10

September 2004 Page iv

Extreme ice thicknesses from freezing rain Final Report

1.0 Introduction
The Federal Emergency Management Agency (FEMA) formed in 1998 the American Lifelines
Alliance (ALA) as a public-private partnership. In 2002, FEMA contracted with NIBS through
its Multihazard Mitigation Council (MMC) to, among other things, assist FEMA in continuing
ALA earlier guideline development efforts. In 2003, ALA requested Kathleen F. Jones of the
U.S. Army Corps of Engineers (USACE) Engineer Research and Development Center (ERDC)
Cold Regions Research and Engineering Laboratory (CRREL) Snow and Ice Branch to provide
maps of extreme equivalent radial ice thicknesses from freezing rain for mean recurrence
intervals of 50, 100, 200 and 400 years for both English and metric units. This effort
complements the map of 50-yr ice thicknesses in English units that is published in the ASCE 7
Standard Minimum Design Loads for Buildings and Other Structures (ASCE 2003). The latest
revision is ASCE 7-2005 (ASCE in draft), which is expected to be published early in 2005. The
50-yr maps are also in the current draft for ASCE Manual 74 Guidelines for Electrical
Transmission Line Structural Loading (the current revision is ASCE 1991) and are proposed to
be included in the 2007 revision of the National Electrical Safety Code (the current revision is
NESC 1997).

1.1 Project Objective

The purpose of this guide is to
• explain the concept of equivalent radial ice thicknesses
• provide references to background information
• describe how ice thicknesses are estimated for long mean recurrence intervals
• provide maps in English units for mean recurrence intervals of 100, 200 and 400 years to
extend the 50-yr map in ASCE Standard 7
• provide maps in metric units for mean recurrence intervals from 50 to 400 years, for use
in hard metric standards, codes, and guidelines

1.2 Project Scope

This guide addresses extremes of equivalent radial ice thicknesses from freezing rain with
concurrent gust speeds.

This guide does not address accreted ice loads from in cloud icing or from sticky snow (see the
various Proceedings of the International Workshop on Atmospheric Icing of Structures). These
winter weather phenomena occur with different spatial distributions from freezing rain. It also
does not address dynamic loads associated with icing, including galloping of conductors (e.g.
Rawlins 1979), ground wires, and tower guys, which occurs with relatively small ice thicknesses
in moderate to high winds and can cause significant damage, either in a single episode or over
many winters through accumulated fatigue damage. Spatial factors, which quantify the increase
in risk environmental loads to distributed structures, such as transmission lines and distribution
line systems, are not discussed. The spatial effect for a variety of environmental loads is

September 2004 Page 1

Extreme ice thicknesses from freezing rain Final Report

discussed in a number of papers, including Golikova (1983) on ice storms, Twisdale (1982) on
tornadoes, and Vickery and Twisdale (1995) on hurricanes.

1.3 Notation
d = wire, branch diameter
L = sample length
m = ice mass
P = depth of freezing rain
t = equivalent radial ice thickness
V = average wind speed
Vo = 10 knots or 5 m/s
π = 3.14
ρ = density of ice = 0.9 g/cm3

September 2004 Page 2

Extreme ice thicknesses from freezing rain Final Report

2.0 Determining extreme ice thicknesses from freezing rain

2.1 Ice accretion models

The CRREL model for the accretion of ice on cylinders (e.g. wires and branches) from freezing
rain is described in Jones (1996). This model determines the amount of ice that freezes both
directly to the wire and as icicles, as any initially unfrozen water starts to drip off the wire. The
Simple model is also introduced in the above report, but is described in more detail in Jones
(1998).

In both models the severity of icing is quantified in terms of the equivalent radial ice thickness t.
A back-of-the envelope formulation of the Simple model that can be used, for example, to
estimate the potential damage to trees and overhead wires in forecasted freezing-rain storms is:

2
⎛V ⎞
t = 0.35 P 1 + ⎜ ⎟ (2-1)
⎝ V0 ⎠

where P is the total depth of freezing rain expected and V is the average wind speed
accompanying the freezing rain. The equivalent radial ice thickness t is in the same units as P,
and V0 is 10 for V in mph (or knots) and 5 for V in m/s. Note that t is independent of the wire
diameter. Based on our measurements and observations in the field in freezing rain storms, we
expect significant tree damage when t ≥ 0.25 in.

The mass m of ice with density ρi on a wire with diameter d and length L is calculated from the
equivalent radial ice thickness t:

m = ρiπ L ( dt + t 2 ) (2-2)

The ice mass increases with wire diameter for a given t.

2.2 Equivalent radial ice thickness
The equivalent radial ice thickness that describes the mass or weight of ice on a wire (or other
cylinder) is different from the maximum dimension of the ice accretion provided in Bennett
(1959) and Changnon (2003) and used in various papers by Hay (e.g. Hay 1957). These authors
use the measurements made by the American Railroad Association (ARA) during a 9 year period
in the 1920s and 1930s. According to Stanley Changnon (personal communication, August 2004)
he and William W. Hay were contemporaries at the University of Illinois. Hay had gotten the
ARA to provide him with all the data from the 9 year study. At one point, Hay gave Changnon a
summary list he had compiled from the original data of all the measurements of ice that were
associated with damage. This list is the 1689 point measurements that Changnon refers to in his
2003 paper. The summary (which is not publicly available) lists the location, date, and total ice
thickness, but with no description of how the thickness was measured. Changnon believes the
measurements were made by someone at the depot by climbing up the signal mast ladder with a

September 2004 Page 3

Extreme ice thicknesses from freezing rain Final Report

ruler to measure the maximum dimension of the ice on the wires where they entered the depot
above the double window, 8 to 10 ft above ground.
The CRREL Ice Storm Team has deployed in a number of freezing rain storms beginning in the
mid 1990s to measure equivalent radial ice thicknesses and document the distribution and
severity of icing on trees and wires. The photographs in Figure 2.2-1 were taken in those storms
and are chosen to illustrate some of the variety of ice accretion shapes that occur. This figure
shows that because of the great variety of ice accretion shapes, the maximum dimension of the
accretion that is reported in papers that use ARA data is not a good measure of the load of ice on
the wire. The determination of the equivalent radial ice thickness t from field measurements of
ice samples in freezing rain storms is presented in Figure 2.2-2.

2.3 Application to weather data

The approach used to apply the CRREL and Simple ice accretion models to U.S. and Canadian
hourly weather data to determine equivalent radial ice thicknesses in past freezing rain storms is
described in Jones et al. (2002). A map of the weather stations used in this analysis in the lower
48 states and neighboring portions of Canada is shown in Figure 2.3-1. Stations in Alaska and
portions of northwest Canada are shown on the extreme ice maps for Alaska in the Appendix. As
is mentioned in the 2002 paper, we obtained damage information from Storm Data (NOAA
1959-present), Climatological Data: National Summary (NOAA 1950-1958), and newspaper
reports on freezing rain storms for which the modeled ice thicknesses were significant. From the
model results and this qualitative damage information we mapped the region of each storm in
which there was enough ice to damage trees, overhead wires (telegraph, phone, power, cable TV,
etc.), and communication towers. A map of the compiled damaging ice storm footprints is in
Figure 2.3-2.
The weather conditions during ice storms are not extraordinary. Winds are typically light to
moderate, precipitation rates are relatively low, and temperatures are at or below freezing but not
bitterly cold. Thus, extraordinary ice storms typically occur when conditions favorable for
freezing rain or drizzle persist for many days. This also means there is no natural maximum
equivalent radial ice thickness. Weather conditions that produce 1 in. of ice in two days will
produce 2 in. after four days, or 4 in. after eight days, with the same freezing rain rate and wind
speed.

2.4 Extreme value analysis

A discussion of extreme events and various approaches for the calculation of extreme values is in
Jones and White (2002). The maps of equivalent radial ice thickness (henceforth shortened to
“ice thickness”) from freezing rain with concurrent gust speeds are organized by mean
recurrence interval (50, 100, 200 and 400 years) in the Appendix. For each recurrence interval,
there is a map of the lower 48 states, a map of Alaska, and detail maps of Lake Superior, the
Columbia River Gorge, and the Fraser Valley. For 50 and 100-yr mean recurrence intervals, ice
thicknesses are mapped in 0.25 in. increments. Each ice thickness zone includes values from
70% below the nominal value to 30% above. For example, the 0-in. zone includes ice thicknesses
up to 0.075 in.; the 1-in. zone includes ice thicknesses from 0.825 in. to 1.075 in. On the 200 and
400-yr maps, ice thickness zones are 0 in., 0.25 in., 0.5 in., 1 in., and continue in 0.5-in., rather
than 0.25-in. increments, reflecting the greater uncertainty in the values for these long mean

September 2004 Page 4

Extreme ice thicknesses from freezing rain Final Report

recurrence intervals. Similarly, the corresponding metric maps use 5-mm increments for the 50
and 100-yr maps, and 10-mm increments for the 200 and 400-yr maps. Major terrain features
were used in defining some of the ice zones on the maps. For comparison, a color relief map of
the United States is available at http://fermi.jhuapl.edu/states/us/big_us_color.gif .
As is described in Jones et al. (2002), the 50-yr ice thicknesses are calculated by grouping the
~500 weather stations into superstations to generate longer periods of record for the extreme
value analysis. By doing this, the 50-yr values are within the sample of extremes rather than far
out in the tail of the distribution. However, the 100-yr to 400-yr extremes are farther out in the
tail and thus may still be significantly affected by the tail shape parameter, which in turn is
affected by the random occurrence, or not, of storms with long mean recurrence intervals (e.g.
the January 1998 storm in the northeast) in the period of record. To decrease this variation
between superstations, 100-, 200- and 400-yr values are calculated for each superstation using an
average multiplier on the 50-yr value for each mean recurrence interval. The same multipliers are
used for all superstations in zones with 50-yr ice thicknesses of 0.5 in. or more. Separate
multipliers are used for stations in the 0.25-in. and 0-in. zones. These factors are calculated as the
average of the ratios of ice thickness (e.g. the 100-yr value divided by the 50-yr value) weighted
by the number of years in the period of record for each superstation. The factors are shown in
Table 2-1.

Factor to multiply 50-yr value

100-yr 200-yr 400-yr
Superstations in 50-yr 0.5” zone and greater 1.25 1.55 1.91
Superstations in 50-yr 0.25” zone 1.34 1.76 2.31
Superstations in 50-yr 0” zone 1.62 2.42 3.49

Table 2-1. Factors on 50-yr thicknesses

The 50-yr map was developed over the past 8 years as CRREL, utility organizations (e.g. EPRI,
CEA), individual utilities (e.g. VELCO, NYPA, BPA), and FEMA funded the mapping of
various regions of the country. Partial maps were included in ASCE Standard 7 Minimum
Design Loads for Buildings and Other Structures, initially in the 1995 revision, and then in the
1998 and 2002 revisions as additional regions were analyzed. The map in this document is in the
draft for ASCE 7-2005. There are discrepancies between this 50-yr map and the maps for the
longer recurrence intervals. On the 50-yr map, zones in southern Minnesota-northern Iowa and
eastern Pennsylvania-northern New Jersey are shown with ice thicknesses greater than the
surrounding area, but these zones are not delineated on the 100-, 200-, and 400-yr maps. Those
are only two of the many relatively small areas across the country with significant local
variations in the severity of icing associated with variations in terrain and thus, should not be
delineated. A list of examples for ASCE Standard 7 of areas where extreme ice thicknesses are
expected to be greater than in the surrounding terrain include: Signal and Lookout Mountains in
Tennessee; Ponatock Ridge and the edge of Yazoo Basin in Mississippi; Shenandoah Valley and
Poor Mountain in Virginia; and Mt. Washington in New Hampshire. As a result of the State
Climatologist review, Buffalo Ridge in southwest Minnesota is being added to this list.

September 2004 Page 5

Extreme ice thicknesses from freezing rain Final Report

The maps of extreme ice thicknesses should not be interpreted as predictions of the future. They
are simply an evaluation of the risk based on current information. Rare ice storms can and do
occur now. Because structures are designed for specified risk levels that correspond to relatively
long return periods, we need to evaluate the severity of ice storms with large equivalent radial ice
thicknesses based on the information now available. For example, the 1998 ice storm in the
northeast appears to be consistent with a mean recurrence interval of about 250 years in the
Montreal area. As time goes by, and no other extraordinary ice storms occur, then the estimate of
that mean recurrence interval will increase. Or, on the other hand, if another storm as severe as
the 1998 storm occurs 20 years from now, a reanalysis of the data will indicate a shorter mean
recurrence interval. Any nonstationarity in the climate, whether natural or anthropogenic, can
certainly further affect the estimates of extremes as well.

September 2004 Page 6

Extreme ice thicknesses from freezing rain Final Report

Figure 2.2-1 Some shapes of ice accreted from freezing rain on cylinders

September 2004 Page 7

Extreme ice thicknesses from freezing rain Final Report

Definition:
Ice

Wire, Equivalent radial

branch ice thickness t
diameter
d

Determining t from measurements in the field:

M

Measure:
sample ice mass m
branch or wire diameter d
sample length L

Calculate equivalent radial ice thickness t:

d d2 m
t =− + +
2 4 πρ i L
where
π = 3.14
ρi = density of ice = 0.9 g/cm3

Figure 2.2-2 Equivalent radial ice thickness

September 2004 Page 8

Extreme ice thicknesses from freezing rain Final Report

Figure 2.3-1 Weather stations in contiguous 48 states

September 2004 Page 9

Extreme ice thicknesses from freezing rain Final Report

Figure 2.3-2 Footprints of damaging ice storms 1948-2002

September 2004 Page 10

Extreme ice thicknesses from freezing rain Final Report

References
ASCE, in draft, Minimum Design Loads for Buildings and Other Structures, ASCE Standard 7-
05, Reston, Virginia.
ASCE, 2003, Minimum Design Loads for Buildings and Other Structures, ASCE Standard 7-02,
Reston, Virginia.
ASCE, 1991, Guidelines for Electrical Transmission Line Structural Loading, ASCE Manual 74,
American Society of Civil Engineers, New York.
Bennett, I., 1959, Glaze: its meteorology and climatology, geographical distribution, and
economic effects, Quartermaster Research and Engineering Center, Technical Report EP-105,
217 pages.
Changnon, S., 2003, Characteristics of ice storms in the United States, J. of Applied
Meteorology, 42, 630-639.
Golikova, T.N., B.F. Golikov and D.S. Savvaitov, 1983, Methods of calculating ice loads on
overhead lines as spatial constructions, Proceedings of the First International Workshop on
Atmospheric Icing of Structures, CRREL Special Report 83-17, pp 341-346.
Hay, W.W., 1957, Effects of ice storms on railroad transportation, The effect of weather on
railroad operation, maintenance, and construction, Geography Department, U. of Illinois at
Urbana Champaign, pp 88-117.
Jones, K.F., 1996, Ice accretion in freezing rain, CRREL Report 96-2,
http://www.crrel.usace.army.mil/techpub/CRREL_Reports/reports/CR96_02.pdf
Jones, K.F., 1998, A simple model for freezing rain ice loads, Atmospheric Research, pp 87-97.
ftp://ftp.crrel.usace.army.mil/pub/outgoing/kjones/SimpleModel.pdf
Jones, K.F., R. Thorkildson and J.N. Lott, 2002, The development of the map of extreme ice
loads for ASCE Manual 74, Electrical Transmission in a New Age, Omaha, ASCE, Reston
Virginia, pp 9-31. Published on the web as The development of a U.S. climatology of extreme
ice loads at ftp://ftp.ncdc.noaa.gov/pub/data/techrpts/tr200201/tr2002-01.pdf
Jones, K.F. and H.B. White, 2002, The estimation and application of extremes, Electrical
Transmission in a New Age, Omaha, ASCE, Reston Virginia, pp 32-47.
ftp://ftp.crrel.usace.army.mil/pub/outgoing/kjones/ETNAjonesandwhite.pdf
NESC, 1997, National Electrical Safety Code, National Bureau of Standards, Washington, D.C.
NOAA, 1950-1958, Climatological Data, National Summary, National Climate Data Center,
Asheville, North Carolina
NOAA, 1959-present, Storm Data, National Climate Data Center, Asheville, North Carolina.
Rawlins, 1979, “Galloping of conductors”, chapter 4 in Transmission Line Reference Book,
Electrical Power Research Institute.
Twisdale, L.A., 1982, Wind-loading underestimate in transmission line design, Transmission and
Distribution, December 1982, pp 40-45.

September 2004 Page 11

Extreme ice thicknesses from freezing rain Final Report

Vickery, P.J. and L.A. Twisdale, 1995, Prediction of hurricane wind speeds in the United States,
ASCE Journal of Structural Engineering, 121, pp 1691-1699.

September 2004 Page 12

Extreme ice thicknesses from freezing rain Final Report

Acronym List

ALA American Lifelines Alliance

ARA American Railroad Association
ASCE American Society of Civil Engineers
BPA Bonneville Power Administration
CEA Canadian Electricity Association
CRREL Cold Regions Research and Engineering Laboratory
EPRI Electrical Power Research Institute
ERDC Engineer Research and Development Center
MMC Multihazard Mitigation Council
NIBS National Institute of Building Science
NOAA National Oceanic and Atmospheric Administration
NYPA New York Power Authority
VELCO Vermont Electric Company

September 2004 Page 13

Extreme ice thicknesses from freezing rain Final Report

Terms and Definitions

equivalent radial ice thickness: The uniform radial thickness of ice on a cylinder; describes the
ice load.
galloping. Large amplitude, low frequency vibration, typically of overhead wires and guys for
tall towers.
mean recurrence interval. The inverse of the annual exceedance probability. For example, the
ice thickness with an annual exceedance probability of 0.02 has a 50-yr mean recurrence interval.
superstation. A grouping of weather stations with similar climatology for the environmental
load under consideration for the purpose of generating a longer period of record for an extreme
value analysis.

September 2004 Page 14

Extreme ice thicknesses from freezing rain Final Report

Appendix. Maps for mean recurrence intervals of 50, 100, 200, and 400
years

Maps with English units (ice thicknesses in inches, wind speeds in miles per hour) are in
Appendix A. Maps with metric units (ice thicknesses in millimeters, wind speeds in meters per
second) are in Appendix B. For each mean recurrence interval, the map of the 48 contiguous
states is followed by the Alaska map and then the three detail maps for Lake Superior, the
Columbia River Gorge, and the Fraser Valley. The maps are arranged in order of increasing
mean recurrence interval.

September 2004 Page 15