collaborating on the development of warn-on-forecast
DESCRIPTION
Collaborating on the Development of Warn-On-Forecast. Mike Foster / David Andra WFO Norman OK Feb. 18, 2010. Norman Forecast Office. Storm Prediction Center. National Severe Storms Laboratory. Collaboration. Radar Operations Center. OU. ESRL/GSD. Warning Decision Training Branch. - PowerPoint PPT PresentationTRANSCRIPT
Collaborating on the Development of Warn-On-Forecast
Collaborating on the Development of Warn-On-Forecast
Mike Foster / David Andra
WFO Norman OK
Feb. 18, 2010
Mike Foster / David Andra
WFO Norman OK
Feb. 18, 2010
CollaborationCollaboration
Science/Technology OperationsScience/Technology Operations
JDOP
Pre-STORM
NEXRAD IOT&E IIMAPSCOPSQED
VORTEX
1980 1985 1990 1995 2000 2005 | | | | | | 20xx >
Early algorithms
DOPLIGHT
Pre-AWIPS
MARD Risk Reduction
WDSS OK Mesonet
AWIPSMAR NSSL/SPC
WDSS IIJPOLIHOP
JointHWT
PARCASAVFloREFRACTVORTEX2ProbWarnWarn-On-Forecast
Storm Prediction Center
National Severe Storms LaboratoryRadar Operations Center
Norman Forecast Office
ESRL/GSDOUWarning Decision Training Branch
2008 EWP Participants2008 EWP Participants
• Visiting Forecasters/Evaluators (26)
– NWS Regions: Alaska, Central, Western, Southern, Western
– Environment Canada
– Hydrometeorological Service of Serbia
• Cognizant Scientists (30)
– NSSL, OU, CIMMS, WDTB, WFO-OUN, U. Mass, U. VA
– WAS*IS: Dr. Eve Gruntfest
• Weekly Coordinators
• IT Coordinator
– Plus other IT help from NWC
• Operations Coordinator
• EWP Team Leaders
• Visiting Forecasters/Evaluators (26)
– NWS Regions: Alaska, Central, Western, Southern, Western
– Environment Canada
– Hydrometeorological Service of Serbia
• Cognizant Scientists (30)
– NSSL, OU, CIMMS, WDTB, WFO-OUN, U. Mass, U. VA
– WAS*IS: Dr. Eve Gruntfest
• Weekly Coordinators
• IT Coordinator
– Plus other IT help from NWC
• Operations Coordinator
• EWP Team Leaders
OUN WRFOUN WRF
• Convective initiation and mode
• 4 km grid – ARW core
• Initialize every hour – eight hour forecast
• LAPS initialization – hot start radar
• OUN WRF 20 km - boundary conditions
• Convective initiation and mode
• 4 km grid – ARW core
• Initialize every hour – eight hour forecast
• LAPS initialization – hot start radar
• OUN WRF 20 km - boundary conditions
LOCAL WRF MODEL18Z April 7 2008Forecast for 23Z
450 PM
Warn on Forecast in 2020:
What might it look like?
Warn on Forecast in 2020:
What might it look like?
Forecast looks on track, storm circulation (hook echo) is tracking along centerline of highest tornadic
probabilities
Radar and Initial Forecast at 2100 CST Radar at 2130 CST: Accurate Forecast
MostLikelyTornadoPath
T=2120 CST
T=2150
T=2130T=2140
70%
50%
30%
T=2200 CST
Developing thunderstorm
2 of 5
NSSL Warn on Forecast Briefing March 5, 2007
MostLikelyTornadoPath
T=2120 CST
T=2150
T=2130T=2140
T=2200 CST
An ensemble of storm-scale NWP models predict the path of a potentially tornadic supercell during the next 1 hour. The ensemble is used to create a probabilistic tornado hazard forecast grid.
70%
50%
30%
Courtesy Lou Wicker, NSSL
Alternative Warning and Forecast Formats
Alternative Warning and Forecast Formats
• High temporal- and spatial-resolution probabilistic hazard grids
• High temporal- and spatial-resolution probabilistic hazard grids
WoF ChallengesWoF Challenges
• best approaches to radar data qc and assimilation into models
• determining how WoF information will be used in NWS operations and communicated to the public
• storm-scale process understanding, predictability, and parameterization improvements
• WoF project priorities, timing, evaluation and implementation strategies
• best approaches to radar data qc and assimilation into models
• determining how WoF information will be used in NWS operations and communicated to the public
• storm-scale process understanding, predictability, and parameterization improvements
• WoF project priorities, timing, evaluation and implementation strategies
Some assumptions…Some assumptions…
• WoF will evolve into operations
– Not a turn key system
– May be regional or local differences in implementation
• There will remain a detection component
– An alert notification will inform of high impact phenomena
• Science and technology will continue to expand the number of observations, analyses and forecasts
• WoF will evolve into operations
– Not a turn key system
– May be regional or local differences in implementation
• There will remain a detection component
– An alert notification will inform of high impact phenomena
• Science and technology will continue to expand the number of observations, analyses and forecasts
…and some more……and some more…• Data volume and complexity will grow
– Dual pol, Phased array , CASA
– Ensemble regional, local and storm scale models
• New forecast and warning formats and dissemination– Hazard probability envelopes
• Graphical
• Text
– More frequent updates possible
– Techniques will require calibration
– Techniques will require bias removal
• Data volume and complexity will grow– Dual pol, Phased array , CASA
– Ensemble regional, local and storm scale models
• New forecast and warning formats and dissemination– Hazard probability envelopes
• Graphical
• Text
– More frequent updates possible
– Techniques will require calibration
– Techniques will require bias removal
…and more…and more
• Role of the human expert will change
– Today’s warning forecaster processes inputs and makes warning decision
– The same process tomorrow will mean many inputs not considered
• Manage the process
– Review
– Assess
– Correct
• Many others…
• Role of the human expert will change
– Today’s warning forecaster processes inputs and makes warning decision
– The same process tomorrow will mean many inputs not considered
• Manage the process
– Review
– Assess
– Correct
• Many others…
Things the WFO can do now…Things the WFO can do now…
Radar data quality control Radar data quality control • Review and assess radar data in real time
• Review and assess storm scale analysis in real time
• Review and assess radar data in real time
• Review and assess storm scale analysis in real time
Information use in operations and communication to users
Information use in operations and communication to users
• A proving ground for operations
– Dedicated WFO position(s) during event operations
– Post-event reviews of information and actions
– HWT
• A proving ground for operations
– Dedicated WFO position(s) during event operations
– Post-event reviews of information and actions
– HWT
Information use in operations and communication to users
Information use in operations and communication to users
• A proving ground for information
– Graphicasts
– Significant weather advisories
– Warning decision updates
• Interact with users groups; EMs, Schools, medical groups, local decision makers, SSWIM
• A proving ground for information
– Graphicasts
– Significant weather advisories
– Warning decision updates
• Interact with users groups; EMs, Schools, medical groups, local decision makers, SSWIM
Evaluation and implementation strategies
Evaluation and implementation strategies
• Evaluate suitability of storm scale forecasts
• Risk reduction of phased implementation
• Evaluate suitability of storm scale forecasts
• Risk reduction of phased implementation