NOAA Research Supporting Hydrometeorological Science & Services: The NOAA Hydrometeorology Testbed (HMT)

Timothy SchneiderNOAA Earth System Research Laboratory

Boulder, CO

December 10, 2008 1

Contact: Timothy.Schneider@noaa.gov

http://hmt.noaa.gov/

OutlineOutline

I. An Introduction to HMTII. An HMT Requirement

i. Observationsii. Modeling

III. Research to Operations Projectsi. An Underlying Principle: Atmospheric Rivers

IV. A Broader Context: Integrated Water Resource Services

December 10, 2008 2http://hmt.noaa.gov/

I. An Introduction to HMTI. An Introduction to HMT

December 10, 2008 http://hmt.noaa.gov/ 3

A critical element: engaging local, state and federal stakeholders…

Will evolve as HMT evolves:• Smaller panel with national scope• Form Regional Implementation Teams

December 10, 2008 4

Building Partnerships

http://hmt.noaa.gov/

Marty RalphNOAA/HMT

HMT: A National Testbed Strategy with HMT: A National Testbed Strategy with Regional ImplementationRegional Implementation

• Dabberdt et. al., 2005: Multifunctional Mesoscale Observing Networks, BAMS, pp. 961-982• Ralph et. al., 2005: Improving Short-Term (0–48 h) Cool-Season Quantitative Precipitation Forecasting Recommendations from a USWRP Workshop, BAMS, pp. 1619-1632

Major Activity Areas

• Quantitative Precipitation Estimation (QPE)• Quantitative Precipitation Forecasts (QPF)• Snow level and snow pack• Hydrologic Applications & Surface Processes• Verification & Decision Support Tools Enhancing & Accelerating Research to Operations Building partnerships

HMT WEST - Cool Season

HMT Southeast – All Season, Including Hurricane Landfall

HMT “Next” (TBD)

December 10, 2008 5http://hmt.noaa.gov/

Leve

l of e

ffort

/ a

ctivi

ty

Fiel

d W

ork

Anal

ysis

AmericanRiver

Tar-NeuseRiver Basins

Key U.S. Watershed

Fiscal Year

HMT-West

HMT-West Legacy

HMT-Southeast

HMT-SE Legacy

HMT-Next

Regions of Focus

NOAA Hydrology Program (Water Resources Data Assimilation)

NOAA Science and Technology Infusion Program (Hydrometeorology Testbed)

98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13

West Coast/Russian River

Caljet Pacjet

Proof of concept

HMT Prototype^̂̂̂

Hydrometeorology Testbed Timeline

CalWater

EFREP

December 10, 2008 6http://hmt.noaa.gov/

II. An HMT RequirementII. An HMT Requirement

December 10, 2008 http://hmt.noaa.gov/ 7

A Water Resource Driver(One Example)

Societal Need: Reservoir operators and water managers need to balance demands for flood control and water supply

They can achieve this through forecast-based operational techniques are being developed by reservoir operators

Stu Townsley:• “NOAA products are essential to Corps flood risk

reduction mission”• “The Corps supports forecast product enhancement”

Water Management Section, Sacramento District, US Army Water Management Section, Sacramento District, US Army Corps of EngineersCorps of Engineers

December 10, 2008 8http://hmt.noaa.gov/

December 10, 2008 9http://hmt.noaa.gov/

NowCast: QPE->24 hrs.(?)

QPF: 0-3 days

QPF: 3-5 days

QPF: 5-14 daysReforecasting (Hamill & Whitaker):• probabilistic• downscaling• analogues forvarious fields

Observations

Subseasonal Forecasting (Weickman & Berry):• global synoptic dynamic model (GSDM)

Assi

mila

toin

&

Dow

nsca

ling

(e.g

. PRI

SM)

HMT Requirement: Produce accurate 0-120 hour+ forecasts of extreme precipitation (NWS) to enable forecast-based water resource management

December 10, 2008 10http://hmt.noaa.gov/

Deterministic Forecasts (McGinley et. al.):• high resolution• ensembles

III. Research to Operations ProjectsIII. Research to Operations Projects

December 10, 2008 http://hmt.noaa.gov/ 11

A New Paradigm for Understanding West Coast Storms: Atmospheric Rivers

December 10, 2008 12http://hmt.noaa.gov/

Research to Operations ProjectsHigh-resolution ensemble QPF/PQPF

• Ensemble means• Probabilistic fields

Moisture Flux Tools• GPSMet Grids• Moisture Flux Verification Tool

Snow Information Tools• Gridded snow level from Q2/NMQ• Point verification; model bias correction

Atmospheric River (AR) ToolsPacific Atmospheric River Threat Indicator, “PARTI”

• Water Vapor Flux Anomaly• Observational-based AR intensity• Reforecasting products (NWP)

*HMT Mini-Workshop on FFMP; Boulder, CO, November 2008December 10, 2008 13http://hmt.noaa.gov/

Research to Operations ProjectsFFMP Alpha Tests

• Water vapor flux tools: Guidance• QPE bias correction• QPF Grids

EFREP: An HMT Legacy• Network of Snow-level Radars• Enhanced GPSMet Network• Enhanced Soil Moisture Network• Dedicated High-resolution ensemble modeling system• Decision Support Tools and Advanced Display System

(ALPS)

*HMT Mini-Workshop on FFMP; Boulder, CO, November 2008

December 10, 2008 14http://hmt.noaa.gov/

IV. A Broader Context: Integrated Water Resource Services

IV. A Broader Context: Integrated Water Resource Services

December 10, 2008 http://hmt.noaa.gov/ 15

NOAA Integrated Water Resource Services

Priority Area Task Team: FY 2009 Activities

Too Little Water: National Integrated Drought Information System Too Much Water: Hydrometeorology Testbed

• West• Carry out winter season field operations for the Carson and American River basins• Deploy advanced observational technology to support NOAA/USGS Debris Flow

Warning pilots for San Diego, Los Angeles, and Monterey Bay areas• Soil moisture demonstration for monsoon season flash flood support in San Pedro

Basin in Southern AZ• Southeast

• Conduct two workshops to identify operational needs and research priorities• Develop project implementation plan for Tar/Neuse watersheds

Water Quality (Delaware River Basin/Estuary & Great Lakes Coast Estuary River Information System (CERIS) & Coastal and Inland

Flooding Observation and Warning (CI-FLOW)

December 10, 2008 16http://hmt.noaa.gov/

Towards An Integrated Approach

NOAA Integrated Water Resource Services

Hydrometeorology Testbed (HMT) Southeast• Deploy and install advanced observing systems• Conduct intensive observational periods• Initiate research projects• Transition tools/findings into NWS operations

Coastal-Estuary-River Information System (CERIS) Implementation (Tar/Neuse/Pamlico)

• CI-FLOW Demonstration (Sea Grant Sponsored R&D Effort)• Provide routine access to HMT Southeast data• Migrate coupled water forecasting system to NOAA operations• Develop, deliver, evaluate and enhance new products and services

THORPEX• Medium Range QPF research supporting HMT• Ensemble numerical modeling for QPF• Strengthens University partner engagement

December 10, 2008 17http://hmt.noaa.gov/

Thank You!

December 10, 2008 18

Contact: Timothy.Schneider@noaa.gov

http://hmt.noaa.gov/

Backup Slides

Demonstration GPRAsin Testbeds

“Infusion”Performance Measures

“Science”Performance Measures

“Technology”Performance Measures

Linking Science, Technology & Infusion Performance Measures to NOAA GPRA Measures

ServiceGPRA

(e.g., QPF)Performance measures for “service”

programs (e.g., LFW)

Performance measures for ST&I (an enabling program)

Today’s predictive services exist on a foundation of earlier innovation in science and technology

December 10, 2008 19http://hmt.noaa.gov/

HMT introduced a real-time, wind profiler-derived, snow-level product that is updated hourly on the internet.

This new capability prompted NWS staff at the CNRFC to ask ESRL to quantify operational snow level forecast performance.

Lundquist et al. (2008) in J. Hydrometeor. documented the relationship to snow at ground level

15% of the freezing level forecast errors were greater than +/- 1,000 ft.

When predicted snow level is below what is observed, this translates to underestimates in stream flow, e.g., a 2,000 ft snow-level error can cause a factor of 3 runoff error (White et al., J. Tech. 2002)

On Developing a Performance Measure for Snow Level Forecasts

snow

rain

Results courtesy of Dr. Allen Whiteand Dan Gottas (ESRL/PSD)

20

HMT has worked with forecast users to identify critical needs for extreme event prediction.

The existing performance measure for QPF (1 inch “threat score”) does not address this need.

17 sites were used to assess QPF performance for events exceeding 1 inch, 3 inches, and 5 inches in 24 hours, at 1, 2 and 3-day lead times.

On Developing a Performance Measure for Forecasting Extreme Precipitation

Of 16 events with >5 inches in 24 hours, the QPF bias was low

-0.71 (1-day lead)-0.60 (2-day lead)-0.51 (3-day lead)

Of 16 events with >5 inches of rain in 24 h, 2 were predicted 1 day ahead

- 5 inch POD = 0.060.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

Day-1 Day-2 Day-3

QPF

Bia

s

NWS California-Nevada River Forecast Center

Prototype QPF performance measures for extreme precipitation events are being developed and baselines are being created by HMT-Probability of Detecting (Forecasting) a >5 inch event-Bias of QPF in events with >5 inches rain observed

21

The HMT ConceptTestbed as a Process

HMT is about “too much water”…• But is intertwined with broader water resource

management, climate & drought issuesObjectives:• To advance our understanding, monitoring and prediction

of processes leading to extreme precipitation and associated flooding

• To demonstrate and evaluate new research and technology through a national strategy employing regional testbeds

• To accelerate and enhance research to operations

December 10, 2008 22http://hmt.noaa.gov/

Marty RalphNOAA/HMT