Use of HPC Confidence Use of HPC Confidence Interval Forecasts to Interval Forecasts to Produce a Hydrologic Produce a Hydrologic Ensemble of River Ensemble of River Forecasts Forecasts John Halquist John Halquist NOAA / NWS / NCRFC NOAA / NWS / NCRFC Chanhassen, MN Chanhassen, MN
17
Embed
Use of HPC Confidence Interval Forecasts to Produce a Hydrologic Ensemble of River Forecasts
Use of HPC Confidence Interval Forecasts to Produce a Hydrologic Ensemble of River Forecasts. John Halquist NOAA / NWS / NCRFC Chanhassen, MN. The problem:. River Forecasts (deterministic) are typically based on one possible hydrometeorologic scenario - PowerPoint PPT Presentation
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
Use of HPC Confidence Use of HPC Confidence Interval Forecasts to Interval Forecasts to Produce a Hydrologic Produce a Hydrologic
Ensemble of River Ensemble of River ForecastsForecasts
The problem:The problem: River Forecasts (deterministic) are typically River Forecasts (deterministic) are typically
based on one possible hydrometeorologic based on one possible hydrometeorologic scenarioscenario
Changing or dynamic weather patterns result Changing or dynamic weather patterns result in different rainfall amounts or distribution than in different rainfall amounts or distribution than forecastforecast
Users want access to more potential scenariosUsers want access to more potential scenarios Some want more future rain, while others want lessSome want more future rain, while others want less
A solution:A solution: Use additional possible Quantitative Use additional possible Quantitative
Precipitation Forecasts (QPF) to produce Precipitation Forecasts (QPF) to produce an ensemble of possible River Forecastsan ensemble of possible River Forecasts
Maintain connection to realistic Maintain connection to realistic meteorologic conditionsmeteorologic conditions
Utilize Short Range Ensemble Forecasts Utilize Short Range Ensemble Forecasts (SREFs) and Hydrometeorological (SREFs) and Hydrometeorological Prediction Center (HPC) expertisePrediction Center (HPC) expertise
River Forecast EnsembleRiver Forecast Ensemble Mean areal QPF are computed from the Mean areal QPF are computed from the
HPC 95% CI QPF – Maximum and MinimumHPC 95% CI QPF – Maximum and Minimum Independent hydrologic model simulations Independent hydrologic model simulations
are produced using these QPF as inputare produced using these QPF as input Resultant hydrologic time-series are Resultant hydrologic time-series are
provided in Scalable Vector Graphics (SVG) provided in Scalable Vector Graphics (SVG) Allows for interactive query of data Allows for interactive query of data
represented in plotsrepresented in plots
Sample ensemble displaySample ensemble display
Observed data
Hydrograph w/HPC 95% CI Minimum QPF
Hydrograph w/HPC 95% CI Maximum QPF
Hydrograph w/ HPC deterministic QPF
Mean Areal QPF (local)
Example eventExample eventForecast, Model and Observed hydrographs for CIDI4 - June 25, 2005
AnalysisAnalysis Performed for April Performed for April
– September 2005– September 2005 Locations: Locations:
Cedar River, IowaCedar River, Iowa 11,593 forecast vs 11,593 forecast vs
observation pairsobservation pairs 49,211 modeled vs 49,211 modeled vs
observation pairsobservation pairs
AnalysisAnalysis Measures:Measures:
BIASBIAS Probability of Detection (POD)Probability of Detection (POD) False Alarm Ratio (FAR)False Alarm Ratio (FAR) Critical Success Index (CSI)Critical Success Index (CSI) Percent Correct (PC)Percent Correct (PC) Mean Error (ME)Mean Error (ME) Mean Absolute Error (MAE)Mean Absolute Error (MAE) Root Mean Square Error (RMSE)Root Mean Square Error (RMSE)