PSD Staff List  »  Lesley L. Smith
Lesley L. Smith – Research Associate
Attribution and Predictability Assessments Team
Affiliation:
Phone: (303) 497-6172
E-mail: lesley.l.smith@noaa.gov
Mailing Address:
NOAA/ESRL Physical Sciences Division
R/PSD1
325 Broadway
Boulder, CO 80305

Fax: (303) 497-6020

All plots preliminary. Right click to see bigger versions.
Back to main Smith page.

New! Newer plots at top of page, older plots at bottom.

ENRRR Analysis 2018

Energy and Moisture Budgets

Older Budget-Related Plots Here

new GFS rectangles Q1 & Q2 3cases, sample results for 'flight' days:




Comparisons of vertically-integrated Qs and Precip

------------------------------------------------------------------------------------------------------------------------------

new TRMM rectangles

Samples (day-month-year)

all original June 2018 rectangle images here

 

Apparent Heat Source and Apparent Moisture Sinks

vertical integrations of Q1 and Q2:


Above plots created via Kim et al method, but averaging over flight path first


Above plots courtesy Byron Blomquist/Chris Fairall


Above plots created via Kim et al method

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------

formulas used 'Kim et al Method'
All fields averaged over the geographic flight region last

Re. Q1

Average over 5 flights:

Re. Q2

Average over 5 flights:

Differences Q1-Q2:

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

formulas used 'Yanai et al Method'

The following plots show Q1, Q2 computed at each grid point and averaged over the flight region last


Average over 5 flights:

Re. Q2

Average over 5 flights:

-----------------------------------------------------------------------------------------------------------------------------------------

Re. 2017 rectangles:





----------------------------------------------------------

Tests of continuity equation (Kim et al. method):

 

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

ENRRR Analysis 2017

Skinny Wrap/Coat Hanger flights here

Wrap flights

Data files:

latlon_D20160121_RF01.nc       latlon_20160126_RF03.nc       latlon_20160203_RF07.nc       latlon_20160212_RF08.nc       latlon_20160215_RF10.nc

profile.avg.rectangles.blobs.xlsx       div.avgprofile.rects.wraps.nc       vvelms.avgprofile.rects.wraps.nc

profile.avg.largerects.blobs.xlsx       div.avgprofile.bigrects.wraps.nc       vvelms.avgprofile.bigrects.wraps.nc

Above rectangle images courtesy Joe Barsugli.

                       

Above rain images courtesy Juliana Dias.

Below unless otherwise stated, profiles are for integrated wrap area, i.e. totals.

Flight 1: 20160121_RF01 profile.avg.20160121_RF01.xlsx     area ~ 693.

 

Flight 3: 20160126_RF03profile.avg.20160126_RF03.xlsx     area ~ 965.

 

Flight 7: 20160203_RF07profile.avg.20160203_RF07.xlsx     area ~ 1052.

 

Flight 8: 20160212_RF08profile.avg.20160212_RF08.xlsx     area ~ 376.

 

Flight 10: 20160215_RF10profile.avg.20160215_RF10.xlsx     area ~ 297.

 


----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Tests related to Rossby Wave Source analysis


some similar plots using ERA-Interim data here
Sample days 20151222 (left) and 20161022 (right)
variables in alphabetical order, color bars/scales differ


-----------------------------------------------------------------------------------------

Following data created with monthly means of daily anomalies


For more info see /Projects/EN3R/forJOE
-----------------------------------------------------------------------------------------

Sample monthly means of daily wind anomalies

Following calculations all done with monthly means of daily wind anomalies


following Sardeshmukh and Hoskins 1998
Rossby wave source terms are minus absolute vorticity times divergence and
minus div winds dot-product gradient absolute vorticity, respectively.