A Statistical Analysis and Synoptic A Statistical Analysis and Synoptic Climatology of Heat Waves over Climatology of Heat Waves over

the Northeast United Statesthe Northeast United States

Scott C. RunyonScott C. Runyon

andand

Lance F. BosartLance F. Bosart

Department of Earth and Atmospheric Sciences, Department of Earth and Atmospheric Sciences,

University at Albany, State University of New YorkUniversity at Albany, State University of New York

  

Overview/MotivationOverview/Motivation

• Heat waves are a major contributor to Heat waves are a major contributor to weatherweather--related fatalitiesrelated fatalities

• From 1986 to 2004 across the U.S.:From 1986 to 2004 across the U.S.:

Source: www.nws.noaa.gov/om/severe_weather/65yrstats.pdfSource: www.nws.noaa.gov/om/severe_weather/65yrstats.pdf

Severe weather eventSevere weather event # of fatalities# of fatalitiesHurricanes*Hurricanes* 323323

TornadoesTornadoes 960960

LightningLightning 10811081

FloodsFloods 15751575

Extreme heat Extreme heat 25962596

Overview/Motivation (cont.)Overview/Motivation (cont.)

• Improved forecasting may result from a Improved forecasting may result from a better understanding of conditions better understanding of conditions antecedent to and during heat wavesantecedent to and during heat waves

• These improved forecasts may become These improved forecasts may become critical given the possibility of an critical given the possibility of an increase in frequency & intensity of heat increase in frequency & intensity of heat waveswaves

Overview/Motivation (cont.)Overview/Motivation (cont.)

• Previous work has largely focused only on Previous work has largely focused only on individual events (e.g., McQueen et al. 1956, individual events (e.g., McQueen et al. 1956, 1957; Livezey and Tinker 1996) or on 1957; Livezey and Tinker 1996) or on extended “heat waveextended “heat wave––droughts” (e.g., Namias droughts” (e.g., Namias 1982, 1991)1982, 1991)

• Published synoptic climatologies have been Published synoptic climatologies have been limited in scope to “Midwest” or Great Plains limited in scope to “Midwest” or Great Plains (e.g., Namias 1955, 1983; Chang & Wallace (e.g., Namias 1955, 1983; Chang & Wallace 1984)1984)

GoalsGoals

• Resolve annual and decadal trends in Resolve annual and decadal trends in heat wave frequency nationwideheat wave frequency nationwide

• Investigate conditions that lead to onset Investigate conditions that lead to onset of Northeast heat wavesof Northeast heat waves

• Understand synoptic evolution of and Understand synoptic evolution of and dynamic contributions to Northeast heat dynamic contributions to Northeast heat waveswaves

OverviewOverview

• MethodologyMethodology– InitialInitial– RevisedRevised

• ResultsResults– Northeast vs. regional heat wave statisticsNortheast vs. regional heat wave statistics– Northeast summer heat wave compositesNortheast summer heat wave composites

• Antecedent conditions Antecedent conditions • Evolution of heat wavesEvolution of heat waves

• ConclusionsConclusions

MethodologyMethodology

• Temperature data were gathered from the Temperature data were gathered from the National Climatic Data Center’s (NCDC) high National Climatic Data Center’s (NCDC) high resolution surface datasetresolution surface dataset

• Daily high temperatures were extracted for Daily high temperatures were extracted for 54 surface stations over a 54-year period 54 surface stations over a 54-year period (1948(1948––2001)2001)

• Stations selected on basis of both dataset Stations selected on basis of both dataset continuity and coverage within the NCDC’s continuity and coverage within the NCDC’s Standard Regions for temperature and Standard Regions for temperature and precipitationprecipitation

Methodology (cont.)Methodology (cont.)

• AnAn anomalously hot day anomalously hot day was initially was initially defined as a day having a high defined as a day having a high temperature temperature ≥ ≥ 2 standard deviations (2 standard deviations (σσ) ) above the normal high temperature*above the normal high temperature** 31-day running mean* 31-day running mean

• Result:Result:– Large variability in number of anomalous Large variability in number of anomalous

warm events between stationswarm events between stations– Extreme ratios of anomalously Extreme ratios of anomalously hot hot daysdays to to

anomalously anomalously coldcold days at many stations days at many stations

Methodology (cont.)Methodology (cont.)

• This initial method led to a discovery:This initial method led to a discovery:High temperatures at most stations are not High temperatures at most stations are not

normally distributednormally distributed (“skewed”)(“skewed”)

Negatively SkewedNegatively Skewed Positively SkewedPositively Skewed

DEN

ERI 71.228.8BOS 69.330.7

LGA 62.337.7

ALB 55.544.5CAR 59.240.8

Methodology (cont.)Methodology (cont.)

• Temperature data is widely Temperature data is widely assumed to be normally distributedassumed to be normally distributed

• Los Angeles, CA and Denver, CO:Los Angeles, CA and Denver, CO:– Most positively and negatively skewed Most positively and negatively skewed

datasets, respectivelydatasets, respectively– Using 2Using 2σ method 989 (49) σ method 989 (49)

anomalously hot days were found in anomalously hot days were found in Los Angeles (Denver) for all seasonsLos Angeles (Denver) for all seasons

Most Positively Skewed Station:Most Positively Skewed Station:Los Angeles, CALos Angeles, CA

Methodology (cont.)Methodology (cont.)

Los Angeles, CA: Daily High Temperatures

June 1 – August 31, 1948 – 2001

MeanMean

Mean +2Mean +2σσ

Mean Mean −−22σσ

Methodology (cont.)Methodology (cont.)

Los Angeles, CA: Composite Mean Sea Los Angeles, CA: Composite Mean Sea Level Pressure (hPa) – Ten Most Level Pressure (hPa) – Ten Most Anomalous Warm Summer DaysAnomalous Warm Summer Days

Most Negatively Skewed Station:Most Negatively Skewed Station:Denver, CODenver, CO

Methodology (cont.)Methodology (cont.)

Methodology (cont.)Methodology (cont.)

June 1 – August 31, 1948 – 2001

MeanMean

Mean +2Mean +2σσ

Mean Mean −−22σσ

Denver, CO: Daily High Temperatures

Denver, CO: Composite Mean Sea Level Denver, CO: Composite Mean Sea Level Pressure (hPa) – Ten Most Anomalous Pressure (hPa) – Ten Most Anomalous

Cold Summer DaysCold Summer Days

• Los Angeles:Los Angeles:– Large temperature anomalies more likely positive Large temperature anomalies more likely positive

and associated with Santa Ana (offshore) eventsand associated with Santa Ana (offshore) events– Typically, marine layer moderates daily high Typically, marine layer moderates daily high

temperaturestemperatures

• Denver:Denver:– Large temperature anomalies more likely negative Large temperature anomalies more likely negative

and associated with upslope eventsand associated with upslope events– Typically, mean westerly, downsloping flow Typically, mean westerly, downsloping flow

influences daily high temperaturesinfluences daily high temperatures

• Non-normal temperature distributions are Non-normal temperature distributions are common and have distinct physical common and have distinct physical explanationsexplanations

Methodology (cont.)Methodology (cont.)

Methodology (cont.)Methodology (cont.)

• Anomalous hot dayAnomalous hot day: A day having a high : A day having a high temperature greater than the daily temperature greater than the daily climatological 97.5 percentile thresholdclimatological 97.5 percentile threshold

• New method eased distribution problems [Los New method eased distribution problems [Los Angeles 669, Denver 736]Angeles 669, Denver 736]

• Heat wave:Heat wave: Three or more consecutive Three or more consecutive anomalously hot daysanomalously hot days

• Regional heat waveRegional heat wave: When heat waves : When heat waves occurring at two or more stations within a occurring at two or more stations within a region had at least one overlapping day region had at least one overlapping day

Statistical AnalysisStatistical Analysis

Statistical Analysis (cont.)Statistical Analysis (cont.)

Statistical Analysis (cont.)Statistical Analysis (cont.)

40 stations40 stations

Northeast Northeast Heat WaveHeat WaveCompositesComposites

• Created using NCEP/NCAR Reanalysis Created using NCEP/NCAR Reanalysis Dataset (0000 & 1200 UTC)Dataset (0000 & 1200 UTC)

Loop 1: Antecedent ConditionsLoop 1: Antecedent Conditions

• Composites are three-day averages Composites are three-day averages centered on 7, 5, 3, & 1 day(s) prior to heat centered on 7, 5, 3, & 1 day(s) prior to heat wave, 2wave, 2ndnd day into heat wave day into heat wave

Antecedent ConditionsAntecedent Conditions(7 Days Prior)(7 Days Prior) n=17n=17

200 hPa heights & 200 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam)anomalies (dam)

Antecedent ConditionsAntecedent Conditions(5 Days Prior)(5 Days Prior) n=17n=17

200 hPa heights & 200 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam)anomalies (dam)

Antecedent ConditionsAntecedent Conditions(3 Days Prior)(3 Days Prior) n=17n=17

200 hPa heights & 200 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam)anomalies (dam)

Antecedent ConditionsAntecedent Conditions(1 Day Prior)(1 Day Prior) n=17n=17

200 hPa heights & 200 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam)anomalies (dam)

Antecedent ConditionsAntecedent Conditions(Onset)(Onset) n=17n=17

200 hPa heights & 200 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam)anomalies (dam)

700 hPa heights & 700 hPa heights & anomalies (dam) anomalies (dam) 1 day before onset of 1 day before onset of

heat waveheat wave

700 hPa height anomaly 700 hPa height anomaly correlations (from correlations (from

Namias 1955)Namias 1955)

Loop 2: Synoptic Evolution of Loop 2: Synoptic Evolution of Heat WavesHeat Waves

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(4 Days Prior)(4 Days Prior) n=17n=17

200 hPa heights (dam), 200 hPa heights (dam), & winds (m s& winds (m s-1-1))

700 hPa heights (dam), 700 hPa heights (dam), Temp. (°C, red), & Temp. (°C, red), &

winds (m swinds (m s-1-1))

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(2 Days Prior)(2 Days Prior) n=17n=17

200 hPa heights (dam), 200 hPa heights (dam), & winds (m s& winds (m s-1-1))

700 hPa heights (dam), 700 hPa heights (dam), Temp. (°C, red), & Temp. (°C, red), &

winds (m swinds (m s-1-1))

Synoptic Evolution of Heat Waves Synoptic Evolution of Heat Waves (First Day)(First Day) n=17n=17

200 hPa heights (dam), 200 hPa heights (dam), & winds (m s& winds (m s-1-1))

700 hPa heights (dam), 700 hPa heights (dam), Temp. (°C, red), & Temp. (°C, red), &

winds (m swinds (m s-1-1))

Synoptic Evolution of Heat Waves Synoptic Evolution of Heat Waves (Midpoint)(Midpoint) n=17n=17

200 hPa heights (dam), 200 hPa heights (dam), & winds (m s& winds (m s-1-1))

700 hPa heights (dam), 700 hPa heights (dam), Temp. (°C, red), & Temp. (°C, red), &

winds (m swinds (m s-1-1))

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(Last Day)(Last Day) n=17n=17

200 hPa heights (dam), 200 hPa heights (dam), & winds (m s& winds (m s-1-1))

700 hPa heights (dam), 700 hPa heights (dam), Temp. (°C, red), & Temp. (°C, red), &

winds (m swinds (m s-1-1))

Loop 3: Air Mass TransportLoop 3: Air Mass Transport

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(4 Days Prior)(4 Days Prior) n=17n=17

850 hPa: 18°C isotherm850 hPa: 18°C isotherm

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(2 Days Prior)(2 Days Prior) n=17n=17

850 hPa: 18°C isotherm850 hPa: 18°C isotherm

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(First Day)(First Day) n=17n=17

850 hPa: 18°C isotherm850 hPa: 18°C isotherm

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(Midpoint)(Midpoint) n=17n=17

850 hPa: 18°C isotherm850 hPa: 18°C isotherm

Synoptic Evolution of Heat WavesSynoptic Evolution of Heat Waves(Last Day)(Last Day) n=17n=17

850 hPa: 18°C isotherm850 hPa: 18°C isotherm

ConclusionsConclusions

Conclusions – Statistical AnalysisConclusions – Statistical Analysis

• Negative trend in Northeast & Southeast Negative trend in Northeast & Southeast hot days and heat waves turns positivehot days and heat waves turns positive

• 1950’s, 1980’s, and 1990’s are peak in 1950’s, 1980’s, and 1990’s are peak in heat wave frequencyheat wave frequency

• Although weaker, trends seen in other Although weaker, trends seen in other eastern regions except Southerneastern regions except Southern

• Trends east of Rockies match those Trends east of Rockies match those found by DeGaetano and Allen (2002); found by DeGaetano and Allen (2002); Gaffen and Ross (1999)Gaffen and Ross (1999)

Conclusions – Antecedent ConditionsConclusions – Antecedent Conditions

• Correlation between strong 700 hPa Correlation between strong 700 hPa anticyclone in eastern Pacific and positive anticyclone in eastern Pacific and positive height anomalies over Northeast U.S. in height anomalies over Northeast U.S. in agreement with past teleconnection researchagreement with past teleconnection research

• Enhancement and eastward progression of Enhancement and eastward progression of ridge over North America coupled to ridge over North America coupled to amplification of ridge (trough) over Eastern amplification of ridge (trough) over Eastern Pacific (West Coast)Pacific (West Coast)

• Origin of wave energy appears to be Origin of wave energy appears to be equatorial Pacificequatorial Pacific

Conclusions – Heat Wave EvolutionConclusions – Heat Wave Evolution

• ““Kicking trough” builds ridge in the East, Kicking trough” builds ridge in the East, displacing warm air mass eastwarddisplacing warm air mass eastward

• Northeast located underneath both Northeast located underneath both anticyclonic shear side of jet exit and anticyclonic shear side of jet exit and downstream of ridge axis (favored area for downstream of ridge axis (favored area for subsidence) subsidence)

• Westward extension and intensification of Westward extension and intensification of Bermuda High allows for west-northwesterly Bermuda High allows for west-northwesterly (downsloping) flow into Northeast(downsloping) flow into Northeast

Statistical Analysis (cont.)Statistical Analysis (cont.)

14 stations14 stations

54 stations54 stations