HRRR Time-Lagged Ensemble (HRRR-TLE) Update Trevor Alcott, Isidora Jankov, Curtis Alexander Steve Weygandt, Stan Benjamin ESRL Global Systems Division, Boulder, CO http://rapidrefresh.noaa.gov/hrrrtle/
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
HRRRTime-LaggedEnsemble(HRRR-TLE)Update
TrevorAlcott,Isidora Jankov,CurtisAlexanderSteveWeygandt,StanBenjaminESRLGlobalSystemsDivision,Boulder,CO
http://rapidrefresh.noaa.gov/hrrrtle/
Initial&LateralBoundaryConditions
Initial & Lateral Boundary Conditions
Expanded RAP
13-km Rapid Refresh (RAP) –21h (NCEP in July 2016)
39h (testing now at ESRL)
3-km High-Resolution Rapid Refresh (HRRR) –18h (NCEP in July 2016)
36h (testing now at ESRL)
750-m HRRR nest Wind Forecast
Improvement Project Experiment (ongoing)
Prototype 3-km storm-scale HRRR ensemble (HRRRE)Testing Spring 2016
3-km HRRR-Alaska
Testing now at ESRL
3-km HRRR Time-Lagged Ensemble (HRRR-TLE)
RAP/HRRRSuite:Hourly-UpdatingForecastModels
01z +0 +1
+0 +1 +2
+0 +1 +2 +3
Time-LaggedHRRRx2013ColoradoFloodCompositeReflectivity(dBZ)
valid02UTC12Sep
00z
23z
02zMRMS
Exp 3-MemberTime-LaggedEnsemblewith2-hourLatency
00ZRun(notready)
23ZRun(notready)
22Z(mostrecentavailable)
21ZRun
20ZRun
PeriodEnds06 Z
PeriodBegins00Z
MoremembersMorespread
LowerskillShorterleadtime
MoremodelsMorespreadUsefulspread?
Memberweighting
Givesnewestdatathehighestimpact>3runshaslittleadditionalimpactevenat100%weight
Time-LaggedEnsembletrade-offs
Prob
ability
PrecipitationAmount
Model
Observations
AddressingBiasandUnderdispersion
LowBias
HighBias
Spring2015
“Quantile Mapping”BiasCorrection
1.00”≅ 99th%ileofanalysisclimatology
99th%ile ofmodelclimatology
≅ 1.23”
=2weeks=1month=2months
40km≅ 13-grid-point radius
553pointsX3runs=1659ensemble“members”
Increasingspreadwithaspatialfilter
3vs.6Time-LaggedMembers
Results:Probabilityof0.5”Precipitationin6hoursMay-Aug2015
Over-confident
Under-confident
Forecastsvalid22-23z
10 - 11 hr forecast
9- 10 hr forecast
8 - 9 hr forecast
Forecastsvalid23-00z
11 - 12 hr forecast
10- 11 hr forecast
9 - 10 hr forecast
HRRR 11z Init
HRRR 12z Init
HRRR 13z Init
Probability (%)
Combiningspatialandtemporalfilters
RecentChangestoHRRR-TLE
• 24-hforecastlength
• Newsevere,flashflood,aviationproducts,allexperimental(ofcourse)
• Conversionofallfieldsto40-kmneighborhoodprobabilities
• TestedinWPCwinterweatherandSPC/HWTspringexperiments,andWPCFFaIR
Neighborhood/SpatialFilterAlgorithmDemonstration
50forecast/analysispairs
biascorrection
RawmemberQPF
Bias-correctedQPF
Neighborhood/SpatialFilterAlgorithmDemonstrationBias-correctedQPF
0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 1 0 0 0 0 0 00 0 0 1 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 0 0 1 0 0 00 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0
Identifygridpointswherebias-correctedQPFexceedsthe
thresholdofinterestandmarktheseasa“hit”(1)
Neighborhood/SpatialFilterAlgorithmDemonstrationBias-correctedQPF
0 0 0 0 0 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 0 0 0 0 0
Alsosetneighboringgridpointsto1.Simplified
exampleusesa1-grid-pointradius,butrealproductsuse40-km(about13gridpoints)
Neighborhood/SpatialFilterAlgorithmDemonstrationBias-correctedQPF
0 0 0 0 0 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 0 0 0 0 0
Nowapplythespatialfilter(typicallyamuchlargerradius
thantheneighborhood)
Neighborhood/SpatialFilterAlgorithmDemonstrationBias-correctedQPF
0 0 0 0 0 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 1 1 1 0 0 0 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 1 1 1 0 00 0 0 0 0 0 0 0 0 0
Inthisexample,48%ofpointsinthespatialfilterradiusare“hits”.Repeatthisforeach
member,andthefinalprobabilityistheaverageof
48%,A%,B%...
Python“fft_convolve”function
http://rapidrefresh.noaa.gov/hrrrtle/
PrecipitationProducts(exp)• 1-h,3-h,6-hPQPF• 6-hprobabilityofprecip exceeding100-yearARI• 3-hprobabilisticrunoff• 1-h,6-hprobabilisticsnowfall
HRRR 23z 13hr pcp fcst HRRR 00z 12 hr pcp fcst HRRR 01z 11 hr pcp fcst
ConsecutiveHRRRrunsproduced>6”precip in~12hrs06hrQPE>6”
Houston,TXFlashFlooding12UTC18April2016
HRRR-TLEforecasts>60%probabilityof6-hrQPFexceeding100-yearaveragereturninterval(ARI)inHoustonarea
ConvectiveProducts(exp)• 4-hprobabilitieswithin40-kmofapoint:
– Wind>50kt /Hail>1in/Tornado
• Hourlythunderstormprobability
AviationProducts(exp)• Hourlyprobabilityof:
– Visibility<5,3,1mi– Ceiling<3000,1000,500ft– IFRconditions– Echotops>25000,30000,35000ft
FireWeatherProducts(exp)• Hourlyprobabilityof:
– Drythunderstorm– 2-mRH<5,10,and/or15%– 10- or80-mwindspeedorgust>20,25and/or30mph– Combinationofaboveconditions
?
Hazard ModelField TruthHeavy rain QPF Stage-IV/MRMS
Heavysnow Explicit snowfrommodelmicrophysics
National snowfallanalysis
Thunderstorm Lightning threat3(graupelfluxandcolumnice)
NLDN
Severewind 10-m hourlymaxwind mesonet
Large hail Verticallyintegratedgraupel
MESH
Tornado Updrafthelicity,0-1-kmshear,SBCAPE/MUCAPE
stormreportsandMRMSrotationtracks
Visibility/Ceiling/EchoTop Visibility,cloudbase,echotopheight(18dBZ)
ASOS,MRMS
Red Flag- Fire 2-mRH,10-mwind URMA and/ormesonet
HRRR-TLE: Project TimelineOrganization/Experiment Hazards Platform Timeline
WPCWWE PQPF,Snowfall,Snow Rate NAWIPSandwebsite January 2016
NSSL/SPCEFP/EWP Tornadoes,Hail, Wind NAWIPSandAWIPSII May 2016
WPCFFaIR RefinedPQPFandFFguidance NAWIPS June2016
AWCSummerExperiment Initialaviationhazards:ceiling,visibility,convection
NAWIPS August 2016
WPCWWE RefinedwinterhazardsandPQPF NAWIPS January2017
AWCWinterExperiment Ceilingandvisibility NAWIPS February2017
NSSL/SPCEFP/EWP Refinedsevereweather guidance NAWIPSandAWPSII May 2017
WPCFFaIR RefinedFF guidance NAWIPS July 2017
AWCSummer Experiment/OPG Refinedaviationhazards NAWIPSandAWPSII August2017
Initiate NCO‘on-boarding” All IDP Late 2017or2018
ProductDevelopmentTimeline
EngageNationalCenterTestbeds
HRRRE-TLE
28Jun2016 17thWRFWorkshop 27
HRRR-TLEisakeycomponentofthisUSWRP-fundedproject:RefinementandEvaluationofAutomatedHigh-ResolutionEnsemble-BasedHazardDetectionToolsforTransitiontoNWSOperations
Related Documents
