Synoptic and Mesoscale Aspects of Ice Storms in the Northeastern United States
A43C-0145
American Geophysical Union
2012 Fall Meeting
San Francisco, CA
3–7 December, 2012
Research supported by NOAA/CSTAR Grant: NA01NWS4680002
Christopher M. Castellano1*, Lance F. Bosart1, Daniel Keyser1, John Quinlan2, and Kevin Lipton2
1Department of Atmospheric and Environmental Sciences, University at Albany, State University of New York, Albany, NY
2NOAA/NWS/WFO Albany, NY
*Corresponding author email: ccastellano@albany.edu
Ice Storm Climatology
Motivation
Summary: Climatology
N = 137
• Ice storms endanger human life and safety,
undermine public infrastructure, and disrupt local
and regional economies
NTOT = 137
NSYN = 25
• 81.8% of ice storms were local, regional, or
subsynoptic, while only 18.2% were synoptic
• Ice storms are historically prevalent and destructive
in the northeastern U.S.
• Ice storms present a major operational forecasting
challenge
Figure 1. Map indicating the 14 county warning
areas (CWAs) included in the climatology.
Figure 2. Archetypical surface synoptic weather patterns associated with freezing precipitation east of the
Rocky Mountains. Caption and figure reproduced from Fig. 2 in Rauber et al. (2001).
Objectives
Figure 3. Number of ice storms impacting each
county during the 1993–2010 period.
Figure 4. Frequency distribution of ice storms by
spatial coverage.
• Ice storms were primarily associated with Type G
(N=65), Type BC (N=30), or Type EF (N=25)
synoptic weather patterns
Figure 5. Distribution of total ice storms (blue) and
synoptic ice storms (red) by event type.
Summary: Composite Analysis
Composite Analysis: Type G Events (N=24)
Similarities
• Establish a cool-season climatology (Oct 1993–Apr
2010) of ice storms in the northeastern U.S.
A
• Ice storms are coincident with an upstream trough,
an amplifying downstream ridge, and confluent
upper-level flow over eastern North America
• Identify environments conducive to ice storms and
dynamical mechanisms responsible for freezing rain
A′
• Provide forecasters with greater situational
awareness of the synoptic and mesoscale
processes that govern the evolution of ice storms
10 m s−1
• Ice storms are accompanied by low- to midlevel
moisture transport and warm advection
10 x 10−11 K m s−1
A
Data and Methodology
Figure 6. Composite 500-hPa geopotential height (black, every
6 dam) and anomalies (shaded, every 30 m).
Ice Storm Climatology
Figure 7. Composite 850–700-hPa layer-averaged wind
(arrows, m s−1) and 0°C isotherm (dashed red), precipitable
water (green, every 4 mm), and standardized precipitable
water anomalies (shaded, every 0.5 σ).
Figure 8. Composite 300-hPa wind speed (shaded, every 5 m
s−1), 1000–500-hPa thickness (dashed, every 6 dam), and
mean sea level pressure (black, every 4 hPa).
Figure 9. Composite 700-hPa temperature (dashed, every 3
K), Q-vectors (arrows, 10−11 K m−1 s−1), and RHS Q-vector form
of QG omega equation (shaded, every 2.5 x 10−16 K m−2 s−1).
A′
5 m s−1
5 cm s−1
Figure 10. Frontogenesis [shaded, every 1 K (100 km)−1 (3
h)−1], potential temperature (black, every 2 K), wind speed
(green, every 5 m s−1), vertical velocity (dashed red, every 5
μb s−1), and ageostrophic circulation (arrows).
1. Any event listed as an “Ice Storm”
2. Any event with significant ice accretion (≥ 0.25 in)
3. Any event with damage attributed to ice accretion
Differences
A
• Type G events are preceded by a deep cold-air
intrusion, whereas Type BC events are preceded by
persistent warm air aloft
• Classified ice storms by spatial coverage:
Counties Affected
Local
≤3
AND
≤3
Regional
4–12
AND
≤6
A′
CWAs Affected
Subsynoptic
13–48
AND
≤6
Synoptic
> 48
OR
>6
• Partitioned ice storms by synoptic weather patterns
commonly associated with freezing rain (Fig. 2)
Composite Analysis
• Identified 35 ice storms impacting the ALY CWA
10 m s−1
10 x 10−11 K m s−1
A
Figure 11. As in Fig. 6, except for Type BC events.
Figure 12. As in Fig. 7, except for Type BC events.
Figure 13. As in Fig. 8, except for Type BC events.
Figure 14. As in Fig. 9, except for Type BC events.
A′
5 m s−1
5 cm s−1
Case Study: 11–12 Dec 2008 Ice Storm
Impacts
Synoptic Evolution
Mesoscale Dynamics
• 74 counties and 8 CWAs affected
• Numerous reports of ice accretion exceeding
0.50 in (1.27 cm)
• More than 1 million customers lost power
1200 UTC 11 Dec 2008
1800 UTC 11 Dec 2008
081212/0600 GYX (Gray, ME)
Case Study
• QG forcing for ascent, frontogenesis, and
ageostrophic transverse circulation are notably
stronger during Type BC events
Summary: Case Study
• Significant icing occurred on the cold side of a warm
front associated with a northeastward moving
surface cyclone
(a)
Figure 16. Map highlighting the 74 counties
impacted by the 11–12 Dec 2008 ice storm.
• During Type G events, the Gulf of Mexico is the
primary moisture source, but during Type BC
events, the Atlantic Ocean also serves a crucial role
Figure 15. As in Fig. 10, except for Type BC events.
• Generated synoptic composite maps (2.5° NCEP–
NCAR reanalysis data) and composite cross
sections (0.5° Climate Forecast System Reanalysis
[CFSR] data) centered on Albany, NY (KALB), for
the two most common synoptic patterns
• Performed analyses at the synoptic time nearest the
midpoint of each event
• Ice storms occur in association with quasigeostrophic (QG) ascent and frontogenetical forcing
beneath an equatorward jet entrance region
• Ice storms occur within a region of ageostrophic
northerlies on the cold side of a surface warm font
Composite Analysis: Type BC Events (N=7)
• Catalogued ice storms impacting 14 NWS CWAs
(Fig. 1) using NCDC Storm Data:
Size
• Ice storm occurrence is modulated by complex
topography and proximity to large bodies of water
(b)
(a)
(b)
Figure 18. (a) 300-hPa wind speed (shaded, every 5 m s−1), 1000–500-hPa thickness (dashed, every 6
dam), and mean sea level pressure (black, every 4 hPa). (b) Precipitable water (shaded, every 4 mm),
850–700-hPa layer-averaged θe (white, every 10 K), and 850–700-hPa layer-averaged moisture transport
(arrows, expressed as the product of wind velocity and mixing ratio, multiplied by 100).
Figure 20. (a) Mean sea level pressure (black, every 2 hPa) and 10-m streamlines (blue). (b) 2-m
temperature (red, every 2°C; dashed contour marks the 0°C isotherm). Surface elevation is shaded every
200 m.
0000 UTC 12 Dec 2008
0600 UTC 12 Dec 2008
Figure 22. Skew-T with temperature (red, °C),
dewpoint (blue, °C), and wind (barbs, kts). The 0°C
isotherm is highlighted in purple.
• Southerly low- to midlevel winds transported high θe
air poleward, providing ample moisture for heavy
and prolonged freezing rain
• Complex terrain enhanced the northerly surface
flow poleward of the warm front, which thereby
reinforced subfreezing air and strengthened the
confluence and associated frontogenesis
• Used 0.5° CFSR data to create maps illustrating the
synoptic evolution and dynamical forcing during the
11–12 Dec 2008 ice storm
• Sounding profiles were characterized by surfacebased subfreezing layers, elevated melting layers,
saturated conditions, and veering low-level winds
• Utilized radiosonde data from the University of
Wyoming and calculated backward trajectories from
the NOAA HYSPLIT model to examine the local
thermodynamic environment in the ice storm region
• Air parcels in the subfreezing layer arrived from
interior Canada (cP air mass), while air parcels
above the inversion had extensive contact with the
western Atlantic Ocean (mT air mass)
Figure 17. Quantitative precipitation estimates for
the 24-h period ending at 1200 UTC 12 Dec 2008.
From the National Mosaic & Multi-sensor QPE.
(a)
Figure 19. As in Fig. 18, except at 0000 UTC 12 Dec 2008.
(b)
(a)
Figure 21. As in Fig. 20, except at 0600 UTC 12 Dec 2008.
(b)
Figure 23. 120-h backward trajectories ending at
Gray, ME (GYX).