1. ATM 401/501 Status of Forecasting: Spring 2013
1. Forecasting at NCEP
Environmental Modeling Center
Ocean Prediction Center
National Hurricane Center
Hydrometeorological Prediction Center
Storm Prediction Center

2. ATM 401/501 Status of Forecasting: Spring 2013
See also:
Hitchens, Nathan M., Harold E. Brooks, 2012: Evaluation of the Storm Prediction Center’s Day 1 Convective Outlooks. Wea. Forecasting, 27, 1580– (PDF Link)

10. Super Tuesday Tornado Outbreak
February 5-6, 2008
63 tornadoes, 57 fatalities
Deadliest event since ’85
Outlook issued 6 days prior
POD 100% for tornadoes occurring in SPC watches
Average warning lead time 17 min
DR. UCCELLINI
2008 is already tied as the 23rd deadliest tornado year since record keeping began in 1950, and it is only mid-February.
America is vulnerable to severe weather.
90% of all presidential disaster declarations are weather related. We still have an average 500 deaths per year and $14 billion damage.
We average 1,300 tornadoes a year in this country (10 year average).
2008 is the 60th anniversary of the first tornado forecast. March 25, 1948, two Air Force Officers, predicted that tornadoes would hit Tinker Air Force Base, Okla., where they were weather officers. They issued the first successful tornado forecast in America, saving many lives.
So far this year we are far above average with 291 tornadoes and 67 deaths.
Despite our best efforts, the past few years have seen a steady increase in the number of tornado deaths in the U.S.
2008: 67 fatalities so far this year (number of tornadoes: 291; killer tornadoes: 16)
2007: 81 (number of tornadoes: 1,074; killer tornadoes: 26)
2006: 67 (number of tornadoes: 1,106; killer tornadoes: 25)
2005: 38 (number of tornadoes: 1,264; killer tornadoes: 13)
Event Synopsis:
The February 5-6 tornado outbreak in the South was the deadliest February tornado outbreak since Feb (Southern Mississippi Valley tornado outbreak with 121 fatalities)
Deadliest single day tornado event since May 31, (Ohio, Pennsylvania tornado outbreak with 76 fatalities)
This event was well forecast.
Storm Prediction Center warned a tornado outbreak was possible 6 days in advance.
SPC issued a high-risk warning early in the morning of Feb. 5, and a moderate-risk warning two days before. This captured the attention of broadcast media and emergency managers since we only issue a handful of the high-risk warnings each year.
The average lead time for the tornado watches was 2 days.
The average lead time for the warnings was 17 minutes.
In terms of the destruction, the numbers are devastating:
63 confirmed tornadoes (spawned by a long-lived squall line that developed out of a strong upper level storm system) across SR and into CR
11 killer tornadoes
59 people killed and more than 350 injured in five states (Alabama, Arkansas, Kentucky, Tennessee, and Mississippi)
Extensive damage across the affected areas
The service performed by our national centers, regional staff, and field staff was exemplary x

11. February 5-6, 2008 Tornado Outbreak
Hazards
Assessment x

12. February 5-6, 2008
Tornado Outbreak
Day 4-8 x

13. February 5-6, 2008 Tornado Outbreak
Day 3
Categorical
Probabilistic
February 5-6, 2008 Tornado Outbreak
Categorical
Probabilistic
Day 2 x

14. February 5-6, 2008 Tornado Outbreak
Day 1
Convective
High Winds
February 5-6, 2008 Tornado Outbreak
Hail
Tornado x

15. x

16. Another Example

17. June 17, 2010 Tornado Outbreak – Upper Midwest
Record tornado day in Minnesota (48 tornadoes)
17 EF2+ tornadoes including four rated EF4 (in MN and ND)
3 killed, more than 40 injured, and widespread damage
Strongly forced situation with well-defined frontal boundary
Tornado Damage in Holmes, ND
From Steve Weiss’ Talk at 2010 NCEP Review

18. Deterministic 00z WRF Model Forecasts valid 18z-04z
EMC 4 km WRF-NMM
NSSL 4 km WRF-ARW
From Steve Weiss’ Talk at 2010 NCEP Review

19. WRF Simulated Reflectivity 18 hr Forecast valid 18z 17 June 2010
NMM NSSL4
Radar

20. WRF Simulated Reflectivity 19 hr Forecast valid 19z 17 June 2010
NMM NSSL4
Radar

21. WRF Simulated Reflectivity 20 hr Forecast valid 20z 17 June 2010
NMM NSSL4
Radar

22. WRF Simulated Reflectivity 21 hr Forecast valid 21z 17 June 2010
NMM NSSL4
Otter Tail Cnty EF Killed
Radar

23. WRF Simulated Reflectivity 22 hr Forecast valid 22z 17 June 2010
NMM NSSL4
Radar

24. WRF Simulated Reflectivity 23 hr Forecast valid 23z 17 June 2010
NMM NSSL4
Polk Cnty EF3 1 Killed
Radar

25. WRF Simulated Reflectivity 24 hr Forecast valid 00z 18 June 2010
NMM NSSL4
Radar
Freeborn Cnty EF4 1 Killed

26. WRF Simulated Reflectivity 25 hr Forecast valid 01z 18 June 2010
NMM NSSL4
Radar

27. WRF Simulated Reflectivity 26 hr Forecast valid 02z 18 June 2010
NMM NSSL4
Radar

28. WRF Simulated Reflectivity 27 hr Forecast valid 03z 18 June 2010
NMM NSSL4
Radar

29. WRF Simulated Reflectivity 28 hr Forecast valid 04z 18 June 2010
NMM NSSL4
NSSL WRF developed excessive cold pool and surged leading edge of convection too far southeast across Iowa. Initially “slow” NMM was better with location but intensity was too weak near Mississippi River.
Radar

30. SSEF Tests of Double-Moment Microphysics 27 hr Forecast valid 03z 18 June 2010
Thompson WDM6
All members identical to SSEF 4 km ARW control run (upper left) except for different double-moment microphysics schemes. All moved convection southeastward too rapidly.
Morrison
Radar

31. x

32. x