Test of background diffusion in the GFS

1
(No Transcript)
2
(No Transcript)
3
(No Transcript)
4
Test of background diffusion in the GFS Fanglin
Yang and Jongil Han 13g Kate Howard included in
package 13i with thermal roughness, Change in
upper level water vapor Glenn.White_at_noaa.gov
5
Background Diffusivity in NCEP GFS
For above PBL or stable PBL
Background diffusivity represents a subgrid
turbulence mixing that is not explained by grid
mean fields. It may be proportional to degree of
surface heterogeneity, grid size, and other
missing physics such as cumulus-induced gravity
wave drag.
6
Old background diffusivity Current background
diffusivity
K03.0m2s-1 everywhere
7
pru12r
pre13d
8
pru12r
pre13d
9
Aviation Weather Center evaluation of new
GFS --jets 5-10 kt weaker in mid-latitudes --jets
narrower with more realistic structure --tropical
easterly jet over Indian Ocean weaker by 10-20
kt, smaller in lat and lon extent --CAT weaker
--fronts, ITCZ, SPCZ much better --convection
better --recommended implementation despite jet,
CAT issues International desk, HPC --new GFS
underestimates trade wind surges in Caribbean,
concern For civil aviation safety CPC
stratospheric group, CSFRR --QBO too weak
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
International Desk

• GFS Parallel underestimates trade wind surges in
  the Caribbean (3 cases). A concern for civil
  aviation safety.

Comparison of observed 700 mb wind vs 72 hr
forecast for Barbados GFS Parallel initial
analysis 12z 01 Jul 2010
HPC

• Based on Barbados sounding, GFS Parallel forecast
  700 mb winds under forecast by 10 kt while
  Operational was correct.
• Have seen errors as large as 20 kt during
  evaluation

22
23
(No Transcript)
24
(No Transcript)
25
Time averaged Magnitude of winds At 15N
120W-0 Note weaker winds at 700 Near 35W in ECMWF
and new GFS than in operational GFS
26
Time averaged magnitude of Winds at 15N in 3 day
Forecasts. Note ECMWF Winds now stronger than
new GFS At 700 and 35W
27
Note weaker analyzed winds at 700 from 55W to 20W
in ECMWF and new GFS compared to operational GFS
28
(No Transcript)
29
for P/Psgt0.2 for P/Pslt0.2
A proposed background diffusivity profile
30
_____ pre13d --------- pre13g
31
_____ T382 old gfs --------- pre13d
32
_____ pre13d --------- pre13g
33
_____ T382 old gfs --------- pre13d
34
_____ pre13d --------- pre13g
35
50hPa
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
T584 new gfsincreased background
diffusion --reduced rms vector wind error
--weakened QBO --weak jet winds equator
weakening different than midlatitudes, evident in
analyses --weak easterly wind surges Reducing
diffusion above 200 hPa --restores
QBO --increases rms wind error --hurricane
tracks worse Atlantic, e. Pacific --hurricane
intensity improved --no impact 500 hPa anomaly
correlations Is lower troposphere damping due to
diffusion or to convection? Is damping at 200 at
equator due to diffusion or to convection? Need
a normalized RMS vector wind error to avoid
rewarding model for damping winds Need more
plots of bias against observations What do
forecasters wantless rms error or more accurate
features even if somewhat displaced?