“Wind Gust Climatology for Southern South Carolina and Coastal North Georgia”
Ryan Kramer
Penn State University
Meteorology
NOAA Goal: Weather-Ready Nation
NWS Charleston, SC
Frank Alsheimer
4/11/2013 1
About Me
• Ryan Kramer
• Penn State University, Meteorology
• Chose Charleston, SC NWS because:
– Experience operational meteorology
– Different style of research
4/11/2013 2
Outline
• Goals
• Background
• Methods
• Results
• Conclusions
• Next Steps
4/11/2013 3
Goals
• Develop guidelines an NWS Charleston, SC forecaster can use to forecast wind gusts.
– No clear blueprint currently in place
• Base these guidelines on trends of various time scales found in an analysis of surface observations
4/11/2013 4
Background
• Wind Gust: Rapid fluctuation in wind speed – 10 knot variation between wind peaks and lulls
Gusts are only reported if ≥ 14 knots
• Land Sites – Sustained Wind: 2 minute average – Gusts: 5 second average
• Marine Sites (buoys) – Sustained Wind: 8 minute average – Gusts: 5 second average
4/11/2013 5
Background
• Gust Factor (G)
G = Umax/Ū Gust over sustained wind speed
• Ex: “1.25 or 125%” – Should always be ≥ 1
• G multiplied by Sustained Wind = Gust 4/11/2013 6
Background
• NWS Charleston wind gust forecasting:
– Forecasters estimate G from experience
– Forecasting Smart Tool’s default value of 1.15 is applied
• Project purpose: Analyze climatological data to determine appropriate gust factor values for forecasting
4/11/2013 7
Sites
4/11/2013 8 google.maps.com
Methodology
• Hourly, or more, time-stamped surface observations at each location – Gust, sustained wind, wind direction, etc. – Data from Jan 1, 2007 to Dec 31, 2011
• Quality Control – Remove erroneous wind reports – Keep only data within 1 standard deviation of mean G
• Ensure dataset representative of “fair-weather” conditions
4/11/2013 9
Methodology
• Geographical Groupings – All-Land vs. Marine
– Land vs. Marine/Shoreline
• Data set with 3,000 observations per location – Noticed a bias towards locations with many points
4/11/2013 10
Wind Statistics
• Gusts found in 17% of all land observations and 50% of all marine observations
• Sustained winds:
4/11/2013 11
Sustained Wind % of Obs.
< 10 knots 82.5%
10 – 20 knots 16.3%
20 – 30 knots 1.2%
30 – 40 knots .02%
40+ knots .001%
Sustained Wind % of Obs.
< 10 knots 35.3%
10 – 20 knots 53%
20 – 30 knots 11.3%
30 – 40 knots .45%
40+ knots .02%
All Land Sites Marine Sites
Wind Statistics
4/11/2013 12
0
20
40
60
80
100
1 6 11 16 21 26 31 36 41 46 51 56% O
f O
bs.
wit
h a
Gu
st R
ep
ort
ed
Sustained Wind Speed (knots)
All Land Sites: Frequency of Gusts for a Given Sustained Wind Speed
All Data
Day
Night
•How often are gusts reported when the sustained wind is X knots?
Wind Statistics
0
10
20
30
40
50
60
70
80
90
100
1 6 11 16 21 26 31 36 41
% O
f O
bs.
wit
h a
Gu
st R
ep
ort
ed
Sustained Wind Speed (knots)
Marine Sites: Frequency of Gusts for a Given Sustained Wind Speed
All Hours
Day
Night
4/11/2013 13
Wind Statistics
4/11/2013 14
0
5
10
15
20
25
30
0 3 6 9 12 15 18 21
% O
f O
bs.
Wit
h a
Gu
st R
ep
ort
ed
Hour (z)
All Land Sites: Frequency of Gusts for a Given Hour
•How often do observations at hour X include a gust report?
Wind Summary
4/11/2013 15
35
39
43
47
51
55
1 4 7 10 13 16 19 22
% O
f O
bs.
wit
h a
Gu
st R
ep
ort
ed
Hour (z)
Marine Sites: Frequency of Gusts for a Given Hour
Seasonal Averages
4/11/2013 16
0
5
10
15
20
25
30
35
0 3 6 9 12 15 18 21% O
f O
bs.
Wit
h a
Gu
st R
ep
ort
ed
Hour (z)
All Land Sites: Frequency of Gusts for a Given Hour
Spring
Summer
Fall
Winter
•Changes in gust frequency magnitude, and daily “peak gust time” shifts with season
Gusts vs. Sustained Wind Speeds
• Gust factors and sustained wind speed trends are inversely related over land
– New grouping: Land vs. Marine/Shoreline
4/11/2013 17
Sustained G
<10 knots 1.65
10-15 knots 1.22
15-20 knots 1.21
20-25 knots 1.22
25+ knots 1.23*
Sustained G
<10 knots 1.94
10-15 knots 1.57
15-20 knots 1.43
20-25 knots 1.38
25+ knots 1.37*
Marine/Shoreline Land
* Statistically insignificant number of points
Hourly Averages
• Slight diurnal variation
4/11/2013 18
1.56
1.58
1.6
1.62
1.64
1.66
1.68
0 3 6 9 12 15 18 21Ave
rage
Gu
st F
acto
r
Hour (z)
Land: Average Gust Factor vs. Hour
1.19
1.2
1.21
1.22
1.23
1.24
0 3 6 9 12 15 18 21A
vera
ge G
ust
Fac
tor
Hour (z)
Marine/Shoreline: Average Gust Forecaster vs. Hour
Gust Factor Average
• Forecasters may want a general gust factor value they can confidently input into long-term forecasts
• Consider that the majority of observed gusts over land occur:
– with sustained winds of 11 to 16 knots
– from 15z to 22z
4/11/2013 19
Marine/Shoreline ≈ 1.21 Land ≈ 1.51
Wind Direction
4/11/2013 20
N
S
E W E W
N
S
Land Marine/Shoreline
1.63
1.64 1.63
1.59 1.21
1.22 1.22
1.25
Wind Direction
Direction G Sustained Wind
0-90° 1.25 15.2
90-180° 1.47 11.4
180-270° 1.45 12.1
270-360° 1.24 15.7
4/11/2013 21
N
google.maps.com
Ft. Pulaski
•More local variation •Gust factor variation explained by sustained wind speed variation
Seasonal Averages
• Slight gust factor variation monthly and seasonally
– Land and Marine/Shoreline appear inversed
4/11/2013 22
1.59
1.6
1.61
1.62
1.63
1.64
1.65
1.66
1 2 3 4 5 6 7 8 9 10 11 12
Ave
rage
Gu
st F
acto
r
Month
Land Sites: Average Gust Factor vs Month
1.17
1.18
1.19
1.2
1.21
1.22
1.23
1.24
1.25
1 2 3 4 5 6 7 8 9 10 11 12
Ave
rage
Gu
st F
acto
r
Month
Marine/Shoreline: Average Gust Factor vs. Month
Gusts vs. Sustained Wind Speeds
4/11/2013 23
0
10
20
30
40
50
60
7 12 17 22 27 32
Ave
rage
Gu
st (
kno
ts)
Sustained Wind Speed (knots)
Land Sites: Average Gust vs. Sustained Wind Speed
Correlation: 0.993
R2 = 0.9882
0
10
20
30
40
50
8 13 18 23 28 33
Ave
rage
Gu
st (
kno
ts)
Sustained Wind Speed (knots)
Marine/Shoreline: Average Gust vs. Sustained
Wind Speed
R2 = 0.9869
Correlation: 0.993
Conclusions • Relationship between surface sustained wind and gusts
is very strong – Accurate sustained wind forecast can lead to a strong gust
forecast
• Over land: Gust factor decreases with increasing sustained wind speed – Multiple factors can cause variability in gust factor
• Most gust trends, and land vs. marine differences,
explained by: – Surface Roughness, atmospheric turbulence – Boundary Layer Mixing
4/11/2013 24
Next Steps
• Analyze model soundings to better compare surface winds to upper air winds – Atmospheric stability may also play a role in gusts
• Closer look at effects of surface roughness – Variation with location
• Update Smart Tool within the NWS Integrated
Forecast Preparation System – Automatically apply trends to forecast grids
4/11/2013 25
Acknowledgements
• Frank Alsheimer
• Jonathan Lamb and Blair Holloway
• The rest of the NWS Charleston office
• Hollings Program Coordinators
4/11/2013 26
References/Resources
• Cook, R., Gruenbacher, B., 2008. Assessment of Methodologies to Forecast Wind Gust Speed. NWS Wichita. http://www.crh.noaa.gov/ict/?n=windgust
• Davis, F.K., Newstein H., 1968. The Variation of Gust Factors with Mean Wind Speed and with Height. J. Appl. Meteor., 7, 372–378.
• Vickery, P., Skerlj, P., 2005. Hurricane Gust Factors Revisited. Journal of Structural Engineering. Vol. 131, No. 5, pp. 825-832
• ASOS Users’ Guide, 1998. • Wieringa, J., 1972. Gust Factors Over Open Water and Built-Up Country.
Boundary-Layer Meteorology. Volume 3, Issue 4, pp.424-441 • Durst, C.S., 1960. Winds Speeds Over Short Periods of Time. The
Meteorological Magazine. Volume 89, No. 1,056. • Krayer, W., Marshall, R., 1992. Gust Factors Applied to Hurricane Winds.
Bulletin American Meteorological Society, Volume 73, No. 5 • National Data Buoy Center., 2012.
4/11/2013 27
Methodology
• Gathered 12z sounding data for Charleston Air Force Base/Intl. Airport (CHS)
– CHS is one of the surface data sites
• Saved CHS data for days with both upper air and surface data
4/11/2013 28
Upper Air Analysis
• Compared winds at 850mb, 925mb, and 1000mb to surface winds
4/11/2013 29
R² = 0.7581
0
5
10
15
20
25
30
35
1 6 11 16 21 26 31 36 41 46 51
Max
imu
m S
ust
ain
ed
Su
rfac
e W
ind
(k
no
ts)
925mb Wind (knots)
Maximum Daily Max. Sustained Wind Speed vs. 12z 925mb Wind
• Strongest relationship at 925mb
– Inversion at 1000mb, 850mb too high
• Are 925mb winds a reliable indicator of surface conditions around the same time of sounding?
• No stand out trends for Gusts or Gust Factors
Upper Air Analysis
R2
1000mb .12
925mb .76
850mb .68
4/11/2013 30
R2
1000mb .05
925mb .76
850mb .62
Max. Sfc Wind Ave. Sfc Wind
R2
1000mb .78
925mb .89
850mb .68
12z Sfc Wind
Addition Method
• Add a number to the sustained wind forecast to get gust forecast value
4/11/2013 31
Sustained Wind Add on
<10 knots 5.4
10-15 knots 3
15-20 knots 3.4
20-25 knots 4.6
25-30 knots 5.9
30+ knots 7.3
Sustained Wind Add On
<10 knots 8.2
10-15 knots 7.3
15-20 knots 8
20-25 knots 9
25-30 knots 10.8
30+ knots 12.8
Marine/Shoreline Land
