Extreme precipitation events in the. southeast U.S.

Size: px

Start display at page:

Download "Extreme precipitation events in the. southeast U.S."

Cora Mary Shelton
6 years ago
Views:

Extreme precipitation events in the Southeast US A preliminary

moisture sources and transport TN flooding May 2010 Asheville,

2010 Asheville, NC 2004 Kelly Mahoney Ben Moore Ellen Sukovich

Motivation: Forecasting extreme precipitation in the southeast

seasons Large variability in types of floodproducing weather

impact events highlight additional challenges:

between decision making bodies Atlanta, GA September 2009

1 Extreme precipitation events in the Southeast US A preliminary investigation of operational forecast challenges related to moisture sources and transport TN flooding May 2010 Asheville, NC 2004 State Climate Office of North Carolina TN flooding May 2010 Asheville, NC 2004 Kelly Mahoney Ben Moore Ellen Sukovich Rob Cifelli CIRES/NOAA Earth Systems Research Lab, Boulder, CO Motivation: Forecasting extreme precipitation in the southeast U.S. Southeast U.S. experiences extreme rainfall during all seasons Large variability in types of floodproducing weather systems Both coastal and mountainous terrain Recent high impact events highlight additional challenges: urbanization/changing watershed hydrology, communication between decision making bodies Atlanta, GA September 2009 Asheville, NC September 2004 Melissa Tuttle Carr Nashville, TN May 2010 AP Photo/Mark Humphrey 1

NOAA s Hydrometeorology Testbed (HMT) HMT conducts research on precipitation and weather conditions that can lead to flooding Fosters transition of scientific advances and new tools into forecasting

Pilot Study in western North Carolina (Spring 2013 Fall 2014) 2.

2 NOAA s Hydrometeorology Testbed (HMT) HMT conducts research on precipitation and weather conditions that can lead to flooding Fosters transition of scientific advances and new tools into forecasting operations Accelerates development, prototyping of advanced hydrometeorological observations, models, dl and physical process understanding di HMT Southeast: 2 parts 1. Pilot Study in western North Carolina (Spring 2013 Fall 2014) 2. Operationally oriented research on extreme precipitation and forecast challenge identification HMT s regional projects HMT Southeast Pilot Study map Mobile atmospheric river observatory (as used in HMT West and HMT Northwest) HMT Southeast: Research Questions What is the climatology of extreme precipitation events in the southeast U.S.? How do QPF errors relate to the largest observed precipitation it ti events? t? What are the primary moisture sources and moisture transport mechanisms for extreme rainfall in the southeast U.S.? Research approach: 1. Climatology of observed extreme events and associated environments 2. Climatology of QPF successes, errors 3. Case studies of observed extreme events: Diagnostic observational analysis, numerical model based experiments 2

1. Climatology of extreme precipitation events in the Southeast: Data and methods Data sources: Daily (12Z 12Z) precipitation accumulations for 2002 2011 NCEP Stage IV QPE product at 4 km resolution

characteristics; flexible approach 99 th percentile of 24 h precipitation (mm) Method of event identification: 99 th percentile of 24 h precipitation calculated at each grid point for all days with

3 1. Climatology of extreme precipitation events in the Southeast: Data and methods Data sources: Daily (12Z 12Z) precipitation accumulations for NCEP Stage IV QPE product at 4 km resolution NARR and CFSR used to examine environmental characteristics of events Previous climatologies exist: many use gauge/station data Gridded dataset facilitates better coverage, diagnosis of event spatial characteristics; flexible approach 99 th percentile of 24 h precipitation (mm) Method of event identification: 99 th percentile of 24 h precipitation calculated at each grid point for all days with precipitation >0 Extreme precipitation events : 24 h periods for which the 99 th percentile threshold was exceeded at >1000 grid points (~16000 km 2 ) within southeast domain Tropical/non tropical classification using National Hurricane Center Best Track information + national radar mosaic imagery This presentation focuses primarily on non tropical events Climatology of extreme precipitation events in the Southeast Locations of maximum 24 h precipitation for each extreme precipitation event = Non tropical (N=140) = Tropical (N=44) Monthly frequency distribution of non tropical and tropical events Salient climatological characteristics: Non tropical events most common in interior southeast; tropical events most common along east coast Extreme precipitation events occur in all months in southeast; least common in summer months Non tropical events most frequent in May; tropical events most frequent in Sept West of Appalachians and in Gulf Coast states, non tropical events most frequent in DJF and MAM East of Appalachians, non tropical events most frequent in SON 3

Approach: Classification of non tropical events Key question: How does

Quantify water vapor transport: vertically integrated water vapor

For each event, average the 24 h time integrated IVT from NARR within 5

Examine correlation between maximum precipitation amount and IVT value

between max precipitation and IVT How are large precipitation events with

Compare largest events in the strongest, weakest IVT terciles Integrated

events Areal coverage Significant at 99.

water Weak IVT, Strong IVT, large precip large precip CAPE Significant at

700-hPa layer-averaged Q- vector convergence Significant at 98.

large precipitation events: More localized/small scale Larger CAPE Weaker

4 Approach: Classification of non tropical events Key question: How does magnitude of water vapor transport relate to precipitation amount? 1. Quantify water vapor transport: vertically integrated water vapor transport (IVT) 2. For each event, average the 24 h time integrated IVT from NARR within 5 lat 5 lon box centered on maximum precipitation p location 3. Examine correlation between maximum precipitation amount and IVT value tion (mm) 24 h max precipitat Weak IVT, large precip. N=19 R 2 =.002 Strong IVT, large precip. N=15 Results and additional questions: Nearly zero correlation observed between max precipitation and IVT How are large precipitation events with ihweak IVT distinguished i i d from those with strong IVT? Compare largest events in the strongest, weakest IVT terciles Integrated water vapor transport Extreme precipitation climatology: Key properties of events Areal coverage Significant at 99.9 th % level Weak IVT, large precip Strong IVT, large precip Precipitable water Weak IVT, Strong IVT, large precip large precip CAPE Significant at 93.9 th % level Weak IVT, Strong IVT, large precip large precip hPa layer-averaged Q- vector convergence Significant at 98.6 th % level Weak IVT, large precip Strong IVT, large precip Weak IVT, large precipitation events: More localized/small scale Larger CAPE Weaker QG forcing for ascent Strong IVT, large precipitation events: More widespread/larger scale Weaker CAPE Stronger QG forcing for ascent Analysis uses CFSR dataset for 6 h largest precipitation period using maximum value (of PW, CAPE, Q vect conv) within 5 lat 5 lonarea centered on event 4

N=15 Composite synoptic scale environment of strong IVT, large

m 1 s 1 ) PW (mm); 850 hpa Z (dam), θ (K) CAPE (J kg 1 ); 925 hpa

couplet, strong jet streak Intense corridor of strong IVT from

Strong low level winds and poleward extending moisture plume; low

unstable air moderate CAPE * Composites are event relative;

Computed using CFSR at beginning of 6h period of largest precip

(dam) Key characteristics Zonal upper level flow pattern;

streak anddownstreamdownstream of weak short wave trough Gradual

anticyclone Precipitation region on warm side of quasistationary

than Strong IVT composite * Composites are event relative; 5

5 N=15 Composite synoptic scale environment of strong IVT, large precipitation events 250 hpa Z (dam), wind speed (m s 1 ) IVT (kg m 1 s 1 ) PW (mm); 850 hpa Z (dam), θ (K) CAPE (J kg 1 ); 925 hpa Z (dam) Key characteristics Amplified upper level troughridge couplet, strong jet streak Intense corridor of strong IVT from low latitudes; connection to subtropics/tropics? Strong low level winds and poleward extending moisture plume; low level warm advection in precipitation region Flow of warm, moist, unstable air moderate CAPE * Composites are event relative; geography shown for spatial reference and distance scaling only. Computed using CFSR at beginning of 6h period of largest precip N=19 Composite synoptic scale environment of weak IVT, large precipitation events 250 hpa Z (dam), wind speed (m s 1 ) IVT (kg m 1 s 1 ) PW (mm); 850 hpa Z (dam), θ (K) CAPE (J kg 1 ); 925 hpa Z (dam) Key characteristics Zonal upper level flow pattern; precipitation region south of equatorward entrance of weak jet streak anddownstreamdownstream of weak short wave trough Gradual poleward low level flow of very moist air around subtropical anticyclone Precipitation region on warm side of quasistationary baroclinic zone in weak warm advection Much larger CAPE values than Strong IVT composite * Composites are event relative; geography shown for spatial reference and distance scaling only. Computed using CFSR at beginning of 6h period of largest precip 5

Research Questions What is the climatology of extreme precipitation events in the southeast U.S.? How do QPF errors relate to the largest observed precipitation events?

Climatologyofobservedextreme of observed extreme eventsandassociated associated environments 2. Climatology of QPF successes, errors 3.

6 Research Questions What is the climatology of extreme precipitation events in the southeast U.S.? How do QPF errors relate to the largest observed precipitation events? What are the primary moisture sources and moisture transport mechanisms for extreme rainfall in the southeast U.S.? Research approach: 1. Climatologyofobservedextreme of observed extreme eventsandassociated associated environments 2. Climatology of QPF successes, errors 3. Case studies of observed extreme events: Diagnostic observational analysis, numerical model based experiments Forecast skill for extreme precipitation events: : Forecast and Evaluation Data Example QPF from HPC Data Period: 1 January December 2011 Example QPE from Stage IV Forecasts: HPC CONUS Quantitative Precipitation Forecasts (QPF) 32 km resolution Forecasts valid 12Z to 12Z Day 1 (24 h) forecasts evaluated here f Verification/quantitative precipitation estimates (QPE) Stage IV data 32 km resolution (upscaled from 4 km) Accumulated precipitation from 12Z to 12Z 6

tropical Extreme Events Tropical events show slight

amounts Non tropical events: underforecasting higher

Weak IVT cases Weak IVT, Large Precip extreme events

cases: underforecast at higher precipitation amounts

7 QPE vs. QPF: Tropical vs. Non Tropical Events Tropical Extreme Events Non tropical Extreme Events Tropical events show slight tendency for overforecasting lower precipitation amounts Non tropical events: underforecasting higher precipitation amounts QPE vs. QPF: Strong IVT vs. Weak IVT cases Weak IVT, Large Precip extreme events Strong IVT, Large Precip extreme events Weak IVT cases: underforecast at higher precipitation amounts Strong IVT cases: also underforecast at higher precipitation amounts, but not by as much 7

Ongoing work and future directions Climatology: More compositing, sub regional event types, focus on event subsets associated with largest QPF errors QPF: QPF error

diagnostic assessments, NWP experiments R2O: Connecting with operational community Object based spatial QPF verification example: Atlanta floods: 20 September 2009 HPC Day 1

(2012) Challenging forecast, high impact event WRF simulated water vapor (purple) and rain (cyan) mixing ratio surfaces Observed vs.

8 Ongoing work and future directions Climatology: More compositing, sub regional event types, focus on event subsets associated with largest QPF errors QPF: QPF error assessments using traditional verification metrics and spatial verification, expand to additional forecast datasets (NWP and human generated) Case studies, process studies: diagnostic assessments, NWP experiments R2O: Connecting with operational community Object based spatial QPF verification example: Atlanta floods: 20 September 2009 HPC Day 1 QPF NPVU 24 h QPE Case study example: WRF model diagnostic/sensitivity studies of 2010 Tennessee floods Moore et al. (2012) Challenging forecast, high impact event WRF simulated water vapor (purple) and rain (cyan) mixing ratio surfaces Observed vs. WRFsimulated radar reflectivity Summary Southeast U.S. experiences extreme rainfall over all seasons Characterizing events, QPF errors in terms of moisture sources, transport properties helps to identify forecast challenges Weak IVT, Large Precip Strong IVT, Large Precip Weak IVT, large precipitation events: More localized/small scale Larger CAPE Weaker QG forcing for ascent Strong IVT, large precipitation events: More widespread/larger scale events Weaker CAPE Stronger QG forcing for ascent Underpredictions of large precip amounts slightly greater in cases with weaker IVT Much work remains; regional partnerships, collaboration key 8

9 Acknowledgements U.S. Weather Research Program, NOAA ESRL s Physical Sciences Division for financial support Hydrometeorological Prediction Center NWS Eastern Region Headquarters HMT SE Planning Workshop participants NASA s Global Precipitation Measurement mission Ana Barros (Duke University) Pilot Study instrumentation collaboration Extra slides 9

HMT Southeast Pilot Study ( HMT SEPS ) Planned for May 2013 September 2014 in western

and eastern NC) NOAA will bring instrumentation and also leverage additional assets from

components include: 3 profiling radars (including 449 MHz for wind profiling with RASS

A nearby GPS receiver will provide integrated water vapor measurements.

NASA may also support deployment of additional NOAA resources for HMT SEPS HMT SEPS main

10 HMT Southeast Pilot Study ( HMT SEPS ) Planned for May 2013 September 2014 in western North Carolina Largely focused on QPE in western NC (but some instrumentation in central and eastern NC) NOAA will bring instrumentation and also leverage additional assets from NASA ground validation campaign Exact deployment plan still in development; likely components include: 3 profiling radars (including 449 MHz for wind profiling with RASS and S band for precipitation vertical structure). An upgrade to the Charlotte profiler. A nearby GPS receiver will provide integrated water vapor measurements. The NASA GPM program will deploy instrumentation as part of COPrHEX. NASA may also support deployment of additional NOAA resources for HMT SEPS HMT SEPS main objectives: Support NWS adaptation to polarimetric radar algorithms Evaluate and improve QPE systems (e.g., MPE, MRMS, Stage IV) 10

11 Atlanta: 20 September 2009 HPC Day 1 24 h QPF NPVU 24 h QPE 11

Cooperative Institute for Research in Environmental Sciences (CIRES) CU-Boulder 2. National Oceanic and Atmospheric Administration

Cooperative Institute for Research in Environmental Sciences (CIRES) CU-Boulder 2. National Oceanic and Atmospheric Administration Moisture transport during the inland penetrating atmospheric river of early November 006 in the Pacific Northwest: A high-resolution model-based study Michael J. Mueller 1 and Kelly Mahoney 1 Cooperative