Impact of Tropical Easterly Waves during the North American Monsoon (NAM) using a Mesoscale Model

1
Impact of Tropical Easterly Waves during the
North American Monsoon (NAM) using a Mesoscale
Model

• Jennifer L. Adams
• CIMMS/University of Oklahoma
• Dr. David Stensrud
• NOAA/National Severe Storms Laboratory
• October 27, 2005

2
What is the NAM?

• Distinct shift in mid-level winds accompanied by
  an increase in rainfall
• Occurs over NW Mexico and SW United States
• Onset usually in July and decays in September
• Great deal of variability

3
NAM Moisture Source

• Moisture source current consensus
• low-level moisture ? Gulf of California (GoC)
• mid-level moisture ? Gulf of Mexico (GoM)
• Transport of low-level moisture by gulf surges
  (one way)
• Induced by passage of tropical easterly waves
  (TEWs) over GoC and/or outflow boundaries/gust
  fronts

4
Gulf surges

• Hales (1972) and Brenner (1974)
• Cooler temps, increased dewpoints, pressure rise,
  southerly wind
• Increase in convection
• Shallow vertical extent
• Loss of definition upon entering desert SW

Adams and Comrie (1997)
5
Motivation

• NAM predictability very low
• TEWs influential to strength of NAM
• Inverse relationship between NAM and U.S. central
  plains rainfall

6
Goals

• Explore impact of TEWs on the NAM
• gulf surges
• NAM region rainfall
• Control run of MM5 compared to simulation where
  TEWs are removed

7
Model Description

• Pennsylvania State University/National Center for
  Atmospheric Research Mesoscale Model (MM5)
• Model domain (350x180x23) at 25 km grid spacing

Puerto Penasco
8
MM5 Parameterization Schemes

• Kain-Fritsch convective scheme (Kain and Fritsch
  1990)
• MRF PBL scheme (Hong and Pan 1996)
• Simple water and ice microphysics (Dudhia 1989)
• Global terrain dataset 10 minute resolution (25
  USGS land use categories)
• Rapid Radiative Transfer Model for radiation
• 5-layer soil model (Dudhia 1996)
• Model initialization NCEP/NCAR reanalysis data
• supply boundary conditions every 6-h
• GoC SSTs set to constant 29.0ºC

9
Methodology

• Four one-month periods
• July 1990, July 1992, August 1988, August 1986
• ECMWF reanalysis data and CPC precipitation
  analysis
• Varying number of TEWs and rainfall amounts

10
ECMWF Hövmoller Diagrams (850 mb)
July 1990
July 1992
August 1986
August 1988
11
Methodology

• Harmonic analysis to remove TEWs from boundary
  conditions
• Reed et al. (1977) TEWs average wavelength 2500
  km, propagation speed of 8 m/s, and average
  period of 3.5 days
• TEWs with periods of approx. 3.5-7.5 days
  identified amplitudes replaced with value of
  zero south of 30ºN
• T, q, u, v, ght, and slp

12
(No Transcript)
13
Harmonics
Harmonic Period (days)
A 31
B 15.5
C 10.33
D 7.75
E 6.20
F 5.17
G 4.43
H 3.88
I 3.44
J 3.10
14
Harmonic Amplitudes (40W)
August 1988
Harmonic E
Harmonic E
TEW
No-TEW
15
MM5 Hövmoller Diagrams (700 mb)
TEW
No TEW
16
Results

• 18 surges over 4 months examined
• 17 induced by TEW/tropical storm
• Varying degrees of strength and frequency

Month of surges
August 1986 6
August 1988 4
July 1990 5
July 1992 3
17
Surge Criteria

• Used time-series data at Puerto Penasco, Mexico
  as a first pass to ID surge events
• Surges occur when
• winds shift to southerly
• maximum daily dewpoint exceeding 65ºF for at
  least 2 days
• peak wind speeds greater than 5 m/s
• decrease in daily max temp of greater than 5ºF
  from the previous day

18
August 1986 Time-series
19
August 1986 Time-series
20
August 1986 Results

• 6 surges in the control run
• all show up in the time-series data at Puerto
  Penasco
• 5 induced by TEWs and 1 initiated by a tropical
  storm (Howard?)
• 2 TEWs possibly contained in the model initial
  conditions

21
TEW passage (18Z Aug 26)
No TEW
TEW
22
Pre-surge (18Z Aug 26)
TEW
No TEW
23
Surge onset (06Z Aug 27)
TEW
No TEW
24
Surge (12Z Aug 28)
TEW
No TEW
25
Post-surge (12Z Aug 29)
TEW
No TEW
26
Surge summary

• TEW passage 12 hours prior to surge onset
• Entire GoC shifts to southerly winds
• 10 of 18 surges (most common)
• Surge virtually absent from no-TEW simulation

27
TEWs and NAM rainfall

• Absence of TEWs has impact on precipitation
  amounts over the NAM region
• Many areas receive more rainfall when TEWs are
  present
• Influences overall extent of NAM precipitation

28
August 1988 Rainfall Differences (TEW-no TEW)
29
Central Plains Rainfall Differences (TEW-no TEW)
August 1988
30
Meridional Moisture Flux
31
Rainfall Differences (00Z Aug 17-06Z Aug 23)
August 1988
32
Rainfall Differences -- 12Z Aug 19 - 00Z Aug 25
August 1986
33
Meridional Moisture Flux
34
Precipitable Water
35
Mid-latitude forcing -- 12Z Aug 20, 1986
TEW
No TEW
36
Adding TEWs

• July 1992 --gt weak monsoon season
• July 1990 --gt strong monsoon season
• Removed waves from July 1992 boundary conditions
• Inserted July 1990 TEWs into July 1992 boundary
  conditions

37
MM5 Hövmoller Diagrams
Hybrid
July 1992
38
12Z July 19 Hybrid run TEW passage
39
Surge (00Z July 20)
Hybrid
July 1992
40
Hybrid - July 1992 TEW Run
41
Conclusions

• Harmonic analysis successfully removes TEWs from
  the model boundary conditions
• MM5 reproduces surges over the GoC
• full gulf, partial gulf, and SMO
• NAM shows great deal of interannual variability
• Surges impacted by absence of TEWs

42
Conclusions

• Reduction of surge events in the no-TEW run
  reduced rainfall amounts over the NAM region
• Absence of TEWs increases precipitation over the
  central United States
• CAPE
• mid-latitude forcing
• Adding waves enhances NAM
• more distinct surge events
• increase in rainfall over core monsoon region

43
Harmonic Analysis

• Since the model data used to create the boundary
  conditions are equally spaced in time and contain
  no missing values, the model data can be
  represented exactly as a series of n points in
  time by summing a series of n/2 harmonic
  functions.