-------
and V vectors between modeled (M) and observed (0) values, is used to assess wind vector
performance (Equation 1). Performance is best when these metrics approach 0.
(1) Wind displacement (km) = (Um - Uo + Vm - Vo)*(l km/1000 m)*(3600 s/hr)*(l hr)
Rainfall performance is examined spatially using side-by-side comparisons of monthly total
rainfall plots and statistically using monthly domain-wide biases. The WRF model outputs
predictions approximately 15 meters above the surface while observations are at 10 meters.
WRF generates output at near instantaneous values (90 second time step) as opposed to longer
averaging times taken at monitor stations. This should be considered when interpreting model
performance metrics.
3.1 Model Performance for Winds
WRF-predicted wind speed estimates are compared to surface-based measurements made in
the ds472 network described earlier and shown below for the 36NOAM (Figure 3.1.1) and 12US
(Figure 3.1.2) domains. Regional4 analysis of statistical metrics for wind speed performance by
quarter5 is shown in Table 3.1.1 for the 12US domain only. Monthly spatial biases across all
hours are shown for the 36NOAM (Figures 3.1.3-3.1.6) and 12US (Figures 3.1.7-3.1.10) domains.
Monthly spatial biases across daytime hours only are shown for the 36NOAM (3.1.11-3.1.14)
and 12US (3.1.15-3.1.18) domains. The hourly distribution of the observed and predicted wind
speeds by each Climate Region and quarter is shown in Figure 3.1.19 for the 12US domain.
There is a noticeable overprediction of wind speeds in the eastern US across all hours and
daytime hours only in both the 36NOAM and 12US simulations. That overprediction is coupled
with a general underprediction of wind speeds in the western US, except in areas of
exceedingly complex terrain (e.g., Rocky Mountains). There appears to be no significant
difference when analyzing the biases during daytime hours or all hours of the day.
Statistically, the mean bias across most regions and seasons is generally within +/- 0.5 m/s. The
underprediction in the western portions of the country noted above is less than -0.75 m/s on
average across all hours of the year.
4 Regions used are the NOAA Climate Regions outlined here:
https://www.ncei.noaa.gov/access/monitoring/reference-maps/us-climate-regions
5 Quarters are Q1 (January, February, March), Q2 (April, May, June), Q3 (July, August, September), and Q4
(October, November, December).
10
-------
Wind Speed Bias
Wind Speed Bias
-B-
-s& &se-
-s&
~T~
M
O
Wind Speed Error
1
a
g
i
S
i
e
a e e e
i
i
e
B
1
1
9
s
i
j
i
F
1
M
i
A
i i i i
M J J A
mon
Wind Speed Fractional Bias
i
s
1
O
i
N
1
D
i
i
i i i ฆ
i
i
1 | 1
i i r 1 i=i i 1
'iii
11
i
i 1
i
j
1
F
1
M
1
A
1 1 1 1
M J J A
mon
Wind Speed Fractional Error
i
s
1
o
1
N
1
D
&
i
^ ^ e e
1
i
1 1 1 1 1 1 1 1 1 1 1 r
J FMAMJ JASOND
mon
Figure 3.1.1. Distribution of hourly bias by hour and hourly bias, error, fractional bias, and
fractional error for wind speed by month for 36NOAM domain.
11
-------
Wind Speed Bias
f -
I I I I
I I I I
"i 1 1 r~
9 10 11 12
~r~
13
~i 1 1 r~
14 15 16 17
~~i 1 1 1 r~
19 20 21 22 23
Hour of day (GMT)
Wind Speed Bias
1
1
1 1
1
1
1
f -
j j
mon
Wind Speed Error
Wind Speed Fractional Bias
swmsssss ssss
~i-
M
"I-
M
~1~~
O
8
CM
* ง J
Wind Speed Fractional Error
^ ^ ^ ^
~r
M
~r~
o
Figure 3.1.2. Distribution of hourly bias by hour and hourly bias, error, fractional bias, a
fractional error for wind speed by month for 12US domain.
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220101 AND 20220131
... ijaavktir .!
iv'.*. "Wr;;- 'v
% * yS' -V *2?vซฅ
J?' ป
*
.. ซ v,..
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220201 AND 20220228
vli *V * VT
?* *3hvV.*v ,
sV.'-T
0.25
0.5
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220301 AND 20220331
\ * /
** *
"ฆhi
-2
ฆ -1
-0.5
-0.25
0
0.25
0.5
1
Figure 3.1.3. Spatial distribution of wind speed bias (m/s) across all hours for the months of
January, February, and March (top to bottom) for the 36NOAM domain.
13
-------
Figure 3.1.4. Spatial distribution of wind speed bias (m/s) across all hours for the months of
April, May, and June (top to bottom) for the 36NOAM domain.
14
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220701 AND 20220731
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220801 AND 20220831
Figure 3.1.5. Spatial distribution of wind speed bias (m/s) across all hours for the months of
July, August, and September (top to bottom) for the 36NOAM domain.
15
-------
October, November, and December (top to bottom) for the 36NOAM domain.
16
-------
Mean bias of Wind Si
Speed (m/s)
Date: BETWEEN 20220101 AND 20220131
: . *
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220201 AND 20220228
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220301 AND 20220331
Figure 3.1.7. Spatial distribution of wind speed bias (m/s) across all hours for the months of
January, February, and March (top to bottom) for the 12US domain.
17
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220401 AND 20220430
t , i-w* ป* ป. ป .
fev- :vi,
ฆ* *fi
w#ซปป T %
;. :. j
.ฆr ' ^
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220601 AND 20220630
~7~
j, *.. -.ia.*."''. . H .
^ tu; ,,
Figure 3.1.8. Spatial distribution of wind speed bias (m/s) across all hours for the months of
April, May, and June (top to bottom) for the 12US domain.
18
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220701 AND 20220731
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220801 AND 20220831
Figure 3.1.9. Spatial distribution of wind speed bias (m/s) across all hours for the months of
July, August, and September (top to bottom) for the 12US domain.
19
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221001 AND 20221031
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221101 AND 20221130
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221201 AND 20221231
Figure 3.1.10. Spatial distribution of wind speed bias (m/s) across all hours for the months of
October, November, and December (top to bottom) for the 12LJS domain.
20
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220101 AND 20220131
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220201 AND 20220228
Figure 3,1.11. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of January, February, and March (top to bottom) for the 36NOAM domain.
21
-------
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220501 AND 20220531
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220601 AND 20220630
Figure 3.1.12. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of April, May, and June (top to bottom) for the 36NOAM domain.
22
-------
Mean oias or win a speea (trvs) uaie: ee i vyeen ^uzzuaui nnu zuz^uaou
\ V" ' . '
ฆฆฆฆ ฆ >f-
Figure 3,1.13. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of July, August, and September (top to bottom) for the 36NOAM domain.
23
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221001 AND 20221031
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221101 AND 20221130
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221201 AND 20221231
Figure 3.1.14. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of October, November, and December (top to bottom) for the 36NOAM domain.
24
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220101 AND 20220131
' 4
\ . sbi.-:-,% .
- -' -"y -
1
J ' !
'"-i
l* ^
>>-ซ* ' > i r
-3
-2
-1
-0.5
-0.25
Figure 3,1.15. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of January, February, and March (top to bottom) for the 12US domain.
25
-------
Mean bias of Wind Speed (m/s)
>: BETWEEN 20220401 AND 20220430
*4 *. W
j' Vs - y .
# m! J>MVL r ...
Mean bias of Wind Speed (m/s) Date: BETWEEN 20220601 AND 20220630
: . *
* ^ \
0.25
0.5
2
3
Figure 3.1.16. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of April, May, and June (top to bottom) for the 12US domain
26
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Figure 3,1.17. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of July, August, and September (top to bottom) for the 12US domain.
27
-------
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221001 AND 20221031
Mean bias of Wind Speed (m/s) Date: BETWEEN 20221101 AND 20221130
-3
-2
-1
-0.5
ซ -0.25
0
0.25
0.5
1
2
3
Figure 3,1.18. Spatial distribution of wind speed bias (m/s) across daytime hours for the months
of October, November, and December (top to bottom) for the 12US domain.
28
-------
"r
3
3
-------
u
5 -EV*
!T!5V.,
.5 -EV*
i
' rnsimmnnmm^ MlMP twป|iซii*iniiปrซ?ป*ii
WWUป
Figure 3.1.19. Hourly average distribution of observed and predicted wind speeds (m/s) for the
12US domain in the Northeast, Northwest, Northern Rockies & Plains, Ohio Valley, South,
Southeast, Southwest, Upper Midwest, and West Climate Regions (respectively, top to bottom)
for each quarter.
30
-------
Climate Region
Season
Mean Obs
Mean Mod
MAE
MB
NMB
NME
RMSE
Northeast
Q1
4.54
4.25
1.4
0.16
3.53
30.77
2.01
Q2
4.11
3.78
1.24
0.09
2.09
30.08
1.78
Q3
3.5
3.13
1.11
0.02
0.54
31.58
1.63
Q4
4.15
3.83
1.3
0.18
4.4
31.28
1.91
N. Rockies & Plains
Q1
5.66
4.81
1.54
-0.63
-11.19
27.21
2.08
Q2
5.81
4.92
1.53
-0.61
-10.45
26.38
2.07
Q3
4.25
3.78
1.25
-0.24
-5.65
29.37
1.72
Q4
5.31
4.44
1.42
-0.56
-10.62
26.85
1.93
Northwest
Q1
3.8
3.05
1.39
-0.36
-9.6
36.69
1.88
Q2
4.27
3.58
1.42
-0.25
-5.87
33.29
1.91
Q3
3.57
2.94
1.21
-0.31
-8.79
33.78
1.63
Q4
3.82
3.04
1.42
-0.32
-8.4
37.22
1.93
Ohio Valley
Q1
4.59
4.33
1.1
0.07
1.59
24.01
1.45
Q2
4.19
3.92
1.11
0.1
2.27
26.41
1.46
Q3
3.21
2.96
0.94
0.13
3.92
29.28
1.24
Q4
4.28
3.99
1.05
0.14
3.27
24.46
1.38
South
Q1
4.84
4.27
1.21
-0.18
-3.67
24.98
1.63
Q2
5.01
4.54
1.31
-0.14
-2.75
26.14
1.76
Q3
3.79
3.38
1.07
-0.06
-1.71
28.22
1.43
Q4
4.29
3.75
1.14
-0.15
-3.41
26.5
1.54
Southeast
Q1
3.92
3.77
1.2
0.36
9.26
30.48
1.6
Q2
3.58
3.36
1.12
0.22
6.25
31.35
1.49
Q3
3.15
2.8
1.05
0.12
3.7
33.34
1.43
Q4
3.44
3.28
1.09
0.37
10.82
31.83
1.46
Southwest
Q1
4.39
3.54
1.56
-0.54
-12.24
35.41
2.14
Q2
5.24
4.38
1.75
-0.52
-9.97
33.37
2.37
Q3
3.83
2.96
1.48
-0.67
-17.58
38.65
2.03
Q4
4.26
3.32
1.55
-0.61
-14.24
36.51
2.17
Upper Midwest
Q1
4.87
4.68
1.2
0.15
3.05
24.72
1.59
Q2
4.79
4.63
1.24
0.18
3.7
25.92
1.63
Q3
3.61
3.6
1.08
0.38
10.47
29.89
1.43
Q4
4.66
4.59
1.22
0.3
6.49
26.18
1.63
West
Q1
3.86
3.1
1.37
-0.33
-8.55
35.56
1.86
Q2
4.82
3.9
1.54
-0.45
-9.36
31.89
2.07
Q3
3.97
3.11
1.33
-0.53
-13.28
33.4
1.82
Q4
3.55
2.82
1.26
-0.34
-9.46
35.43
1.73
Table 3.1.1. Mean observed, mean modeled, mean absolute error
normalized mean bias (NMB), normalized mean error (NME), and
(RMSE) for wind speed (m/s).
(MAE), mean bias (MB),
root mean square error
31
-------
Wind vector displacement (km) is presented below for the 36NOAM (Figure 3.1.20) and 12US
(Figure 3.1.21) domains utilizing the ds472 observation network described earlier. These plots
show the entire distribution of hourly wind displacement by month and by hour of the day. The
average wind displacement for the WRF simulation is around 5-km for all months and hours of
the day. The interquartile ranges are roughly 2-10km. As the displacement is generally less than
the resolution of the model for either simulation, minimal impacts due to displacement of wind
vectors are expected.
32
-------
to _
CNt
O
CM
i/> _
E
JXL
O _
U")
o
0 1 2 3 4 5 6 7 8 9 11 13 15 17 19 21 23
Hour of day (GMT)
Wind Displacement
o
CM ~
ii"> _
ฃ
O _
i
ID
O -
J FMAMJ JASOND
mon
Figure 3.1.20. Distribution of hourly wind displacement by hour and month for the 36NOAM
domain.
33
Wind Displacement
"iir
"iiiiir
"iir
"iiiiiir
-------
Wind Displacement
O
CM
LD
E
O _
i i i
niiiiiiiii iiiiiiiiiiiiir~
0 1 2 3 4 5 6 7 8 9 11 13 15 17 19 21 23
Hour of day (GMT)
Wind Displacement
T
S
T
O
T
N
T
D
mon
Figure 3.1.21. Distribution of hourly wind displacement by hour and month for the 12US
domain.
34
-------
3.2 Model Performance for Temperature
Temperature estimates are compared to the ds472 observation network described earlier and
are presented below for the 36NOAM (Figure 3.2.1) and 12US (Figure 3.2.1) domains. Regional
analysis of statistical metrics for temperature performance by quarter is shown in Table 3.2.1
for the 12US domain only.
WRF performs very well in terms of predicting temperature, showing a bias that oscillates
around 0 degrees for most hours of the day and months of the year. Model error decreases
slightly during the Spring and Summer months. In general, the interquartile range of the bias of
+/-1 degree is persistent across all hours of the day and all months of the year.
Spatial distribution of monthly biases is presented across all hours for the 36NOAM (Figures
3.2.3-3.2.6) and 12US (Figures 3.2.7-3.2.10) domains. The spatial distribution of monthly biases
across daytime hours only is presented for the 36NOAM (Figures 3.2.11-3.2.14) and 12US
(Figures 3.2.15-3.2.18) domains. The hourly distribution of observed and predicted
temperatures by Climate Region and quarter are shown in Figure 3.2.19. WRF generally
underpredicts temperatures slightly across the eastern US during the Winter into the early
Spring with the underprediction persisting longest in the northeast. A more noticeable
overprediction is noted across the eastern US during the summer and fall months with an
average overprediction of 0.25 to 0.5 degrees.
In areas of the western US, performance for temperature is mixed, with persistent significant
overpredictions and underpredictions observed in varying locations. Across daytime hours,
there is a more noticeable underprediction in the 36NOAM simulation compared the 12US
simulation. However, the range of the biases in both simulations tends to be mainly between
+/- 0.5 degrees.
35
-------
Temperature Bias
1 1 1 1 1
3 ECS
"i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r
0 1 2 3 4 5 8 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour of day < GMT}
Temperature Bias
i
T
-
?
^ ^ ^ ^
T
T
T
-
-
-
-
_i_
-I-
1
J
i
F
1
M
1
A
I l 1 l
M J J A
mon
Temperature Error
1
s
1
o
i
N
1
D
-y
>iii
T
7
i
_i_
"J-
1
J
1
F
1
M
1
A
i i i i
M J J A
mon
Temperature Fractional Bias
1
s
1
o
1
N
1
D
-=i
-=L- -=L-
1
J
1
F
1
M
1
A
I I 1 I
M J J A
mon
Temperature Fractional Error
1
s
1
o
1
N
1
D
Figure 3.2,1. Distribution of hourly bias by hour and hourly bias, error, fractional bias, and
fractional error for temperature by month for the 36NOAM domain.
36
-------
Temperature Bias
iO -
1 1
0 1
1 1
2 3
1 1
4 5
1 1
0 7
1 1
3 9
i i i i
1D 11 12 13
1 1
14 15
1 1
16 17
1 1
18 19
i i
20 21
I I
22 23
Hour of day (GMT)
Temperature Bias
-r-
-r-
'
,i,
i
. i .
ฆ 1 i
ii i ' .
l I
l I
i | i
ฆ ; J
*=-1J "-T-
ki1
JT1
1 1
i i
I 3
-I-
-L"
-1
-L-
"L"
I
J
1
F
l
M
I
A
1
w
1 1
J J
1
A
1
5
I
O
l
N
I
D
mon
Temperature Error
i
i
i
J
-r
-r-
-r-
,
i
1
ฆ
e
B
e
e
m a
ฆ
e
e
&
1
J
1
F
I
M
i
A
i
M
1 i
J J
1
A
1
S
i
o
i
N
1
D
mon
Temperature Fractional Bias
1
J
1
F
1
M
1
A
i
M
1 1
J J
1
A
1
s
i
o
I
N
1
D
mon
Temperature Fractional Error
-i-
"l 1 1 1 1 1 1 1 1 1 1 r
J FMAMJ JASOND
men
Figure 3.2.2. Distribution of hourly bias by hour and hourly bias, error, fractional bias, and
fractional error for temperature by month for the 12US domain.
37
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220101 AND 20220131
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220201 AND 20220228
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220301 AND 20220331
Figure 3.2,3. Spatial distribution of temperature bias (C) across all hours for the months of
January, February, and March (top to bottom) for the 36NOAM domain.
38
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220401 AND 20220430
A v. r ซ* f -
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220501 AND 20220531
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220601 AND 20220630
Figure 3.2.4. Spatial distribution of temperature bias (C) across all hours for the months of April,
May, and June (top to bottom) for the 36NOAM domain.
39
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220701 AND 20220731
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220801 AND 20220831
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220901 AND 20220930
Figure 3.2.5, Spatial distribution of temperature bias (C) across all hours for the months of July,
August, and September (top to bottom) for the 36NOAM domain.
40
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221001 AND 20221031
pr
'1 V>^\
"XV^ ' * ' " ' ?*ฆ> -0.5
H,\ f * * "a2i
'# .* i <1 0-25
---4* . &_
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221101 AND 20221130
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221201 AND 20221231
Figure 3.2.6. Spatial distribution of temperature bias (C) across ail hours for the months of
October, November, and December (top to bottom) for the 36NOAM domain.
41
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220101 AND 20220131
r *
ฆ 1 . *
V-
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220201 AND 20220228
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220301 AND 20220331
Figure 3.2.7. Spatial distribution of temperature bias (C) across all hours for the months of
January, February, and March (top to bottom) for the 12US domain.
42
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220401 AND 20220430
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220501 AND 20220531
:-.Y :
\ . .
m 'ฆ
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220601 AND 20220630
v "!#- '
T- ป
77?
- ; 'ฆ .-v..'' .
, V " ป -ฆ? ~
-J ml, r ฆ ป *
-iJf ง
Figure 3.2.8. Spatial distribution of temperature bias (C) across all hours for the months of April,
May, and June (top to bottom) for the 12US domain.
43
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220701 AND 20220731
A' '
#ป r ^
% Jซ! m\9
? - J - - .
.r-\-k-^*
p<^.JnrJ "r.v^w
%||^w
\i\ \>. iป>ปป ay?'
. \\V. v/VS-'v^----i *
a \
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220801 AND 20220831
W:'
-nV
Jfi I "
V/ * Ap . i 1
* > *-; .v "
4
;
t * . ,r
.i v. <#r#
- ' !tg i , ;
ft \
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220901 AND 20220930
Figure 3.2.9. Spatial distribution of temperature bias (C) across all hours for the months of July,
August, and September (top to bottom) for the 12US domain.
44
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221001 AND 20221031
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221101 AND 20221130
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221201 AND 20221231
Figure 3.2.10. Spatial distribution of temperature bias (C) across all hours for the months of
October, November, and December (top to bottom) for the 12US domain.
45
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220101 AND 20220131
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220301 AND 20220331
Figure 3.2.11. Spatial distribution of temperature bias (C) across daytime hours for the months
of January, February, and March (top to bottom) for the 36NOAM domain.
46
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220401 AND 20220430
%
d
J?
l , /. * t* . ซ>
W* ' ซ r 'fr
ซ * * . ป- * : * ป_ ~"* ** "r f
' * . - -- J*
i-_ *-C- .:V.k4>#
t.-r* r.;* viSilr
-..4V v'* .
Lซ
vv
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220501 AND 20220531
r.-
: * - v. V v
.'iฃ
#~f iSฃsfj
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220601 AND 20220630
Figure 3.2.12. Spatial distribution of temperature bias (C) across daytime hours for the months
of April, May, and June (top to bottom) for the 36NOAM domain.
47
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220701 AND 20220731
TT"
- *
0-
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220801 AND 20220831
ซ H
/ป . '
I ** *
r<
If
ฆ, . . -J *>Y ^ ^ ^ . Ai
*ฆ * ปJV to ' 4
K ^ - ' v. Kv
: - % .. vy& -Vt
V\'ป.
Pi--J* L
if-- sy
* ? r
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220901 AND 20220930
Figure 3.2.13. Spatial distribution of temperature bias (C) across daytime hours for the months
of July, August, and September (top to bottom) for the 36NOAM domain.
48
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221001 AND 20221031
u * * n
* .l\*
v.* v. .
ฆ/j:*
- '
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221101 AND 20221130
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221201 AND 20221231
Figure 3.2.14. Spatial distribution of temperature bias (C) across daytime hours for the months
of October, November, and December (top to bottom) for the 36NOAM domain.
49
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220101 AND 20220131
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220301 AND 20220331
Figure 3.2.15. Spatial distribution of temperature bias (C) across daytime hours for the months
of January, February, and March (top to bottom) for the 12US domain.
50
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220401 AND 20220430
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220501 AND 20220531
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220601 AND 20220630
v* 'V '
JU 0 ฆ *' *ป
ft' .: v'
: ; . . '
. i - iif 2 v '. * *
ฆ* **0
. *>
A \
VV
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Figure 3.2.16. Spatial distribution of temperature bias (C) across daytime hours for the months
of April, May, and June (top to bottom) for the 12US domain.
51
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220701 AND 20220731
~ ,f, ป N
. ~ r -I/
*.j. ,-f.
&
, . v y.v > ,Lt: ;jt
/<
ฆ - ? - , ฆ v
r t- - ^ t ฆ ^ - ,ซ
L . . * . ' J*
i , -
* v / t *y .
1
TV
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220801 AND 20220831
i
. Ef--: *
>* v* \v? , *, * *':?4 . \-. i J **
v. <-ฆ>
v\'
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of 2 m Temperature (C) Date: BETWEEN 20220901 AND 20220930
mean Dias orzm temperature (t..} uaie: et i wee
* *ป. * r\ *
* * .5
" ฃ o V,\
*: . . %
%y.TST>ซr
U *
I J *0
Figure 3.2.17. Spatial distribution of temperature bias (C) across daytime hours for the months
of July, August, and September (top to bottom) for the 12US domain.
52
-------
Mean bias of 2 m Temperature (C) Date: BETWEEN 20221001 AND 20221031
Figure 3.2.18. Spatial distribution of temperature bias (C) across daytime hours for the months
of October, November, and December (top to bottom) for the 12US domain.
53
-------
-------
Figure 3.2.19. Hourly average distribution of observed arid predicted temperatures (K) for the
12US domain in the Northeast, Northwest, Northern Rockies & Plains, Ohio Valley, South,
Southeast, Southwest, Upper Midwest, and West Climate Regions (respectively, top to bottom)
for each quarter.
55
-------
Climate Region
Season
Mean
Obs
Mean
Mod
MAE
MB
NMB
NME
RMSE
Northeast
Q1
273.15
272.96
1.71
-0.19
-0.07
0.63
2.27
Q2
288.21
288.25
1.39
0.04
0.01
0.48
1.88
Q3
294.68
294.96
1.3
0.28
0.1
0.44
1.8
Q4
279.9
280.02
1.6
0.11
0.04
0.57
2.12
N. Rockies & Plains
Q1
269.78
269.55
2.08
-0.24
-0.09
0.77
2.69
Q2
285.6
285.73
1.48
0.14
0.05
0.52
1.98
Q3
294.53
294.87
1.48
0.34
0.12
0.5
2.01
Q4
273.17
273.33
1.99
0.16
0.06
0.73
2.63
Northwest
Q1
276.42
276.44
1.85
0.02
0.01
0.67
2.48
Q2
284.58
284.64
1.43
0.06
0.02
0.5
1.91
Q3
293.59
293.86
1.69
0.27
0.09
0.58
2.26
Q4
277.92
278.27
1.86
0.35
0.13
0.67
2.49
Ohio Valley
Q1
274.91
274.75
1.49
-0.16
-0.06
0.54
1.94
Q2
290.97
291.12
1.36
0.14
0.05
0.47
1.77
Q3
295.7
296.1
1.18
0.39
0.13
0.4
1.59
Q4
280.14
280.06
1.47
-0.08
-0.03
0.53
1.9
South
Q1
281.95
282.01
1.74
0.06
0.02
0.62
2.22
Q2
296.65
296.79
1.27
0.14
0.05
0.43
1.72
Q3
300.69
300.95
1.12
0.25
0.08
0.37
1.57
Q4
286.23
286.35
1.58
0.12
0.04
0.55
2.08
Southeast
Q1
284.29
284.28
1.59
-0.01
0
0.56
2.06
Q2
295.05
295.2
1.28
0.15
0.05
0.43
1.71
Q3
298.35
298.67
1.19
0.32
0.11
0.4
1.62
Q4
286.86
286.95
1.53
0.09
0.03
0.53
2
Southwest
Q1
275.72
275.75
2.17
0.04
0.01
0.79
2.89
Q2
290.26
290.27
1.94
0.02
0.01
0.67
2.61
Q3
295.8
296.17
1.81
0.37
0.13
0.61
2.45
Q4
278.61
278.94
2.02
0.33
0.12
0.73
2.67
Upper Midwest
Q1
266.51
266.18
1.72
-0.33
-0.12
0.64
2.29
Q2
286.17
286.28
1.43
0.11
0.04
0.5
1.91
Q3
292.9
293.33
1.28
0.43
0.15
0.44
1.71
Q4
274.91
274.93
1.55
0.02
0.01
0.56
2.01
West
Q1
283.82
283.85
1.86
0.03
0.01
0.65
2.49
Q2
291.58
291.42
1.71
-0.17
-0.06
0.59
2.31
Q3
297.2
297.07
1.7
-0.13
-0.04
0.57
2.31
Q4
285.12
285.37
1.76
0.25
0.09
0.62
2.38
Table 3.2.1. Mean observed, mean modeled, mean absolute error (MAE), mean bias (MB),
normalized mean bias (NMB), normalized mean error (NME), and root mean square error
(RMSE) for temperature (K) for the 12US domain.
-------
3.3 Model Performance for Mixing Ratio
Water mixing ratio estimates are compared to the ds472 observation network described earlier
and are presented below for the 36NOAM (Figure 3.3.1) and 12US (Figure 3.3.2) domains.
Regional analysis of statistical metrics for water vapor mixing ratio performance by quarter is
shown in Table 3.3.1 for the 12US domain only.
Mixing ratio is generally overpredicted across most hours of the day with a greater spread in
the bias in the early morning and evening hours. Increased spread in the bias also occurs during
the late Spring to early Fall when increased moisture levels across the country are noted. In
general, the model error is less than a g/kg across the year and all hours of the day.
The monthly spatial distributions of the mixing ratio bias across all hours are shown for the
36NOAM (Figures 3.3.3-3.3.6) and 12US (Figures 3.3.7-3.3.10) domains. Monthly spatial
distributions of the mixing ratio bias across daytime hours only are shown for the 36NOAM
(Figures 3.3.11-3.3.14) and 12US (Figures 3.3.15-3.3.18) domains. Hourly distributions of the
observed and predicted mixing ratios are shown for each Climate Region and quarter in Figure
3.3.19. As noted in the earlier figures, a general overprediction of moisture is observed across
much of the year, particularly in the Winter and Spring in the eastern US for both the 36NOAM
and 12US domains. An underprediction during the months of September and October is noted
in both domains. In the 12US simulation, there is a noticeable overprediction in the southeast
during July. Some slight variations appear across regions, with a noticeable underprediction of
moisture that persists across the Southeast for much of the year. During daytime hours only,
there is an underprediction during the late Spring months through the Summer period in both
the 36NOAM and 12US simulations. The slight overprediction during the Winter and late Fall
periods persists during the daytime hours, as well.
57
-------
Mixing Ratio Bias
! ! ! ! ; 1 .TTTTTTTTTT, j ; ! ! !
8 0 s s s a $ $ $ $ $ $ 6 $ e s s a b s s B B B
! | ! | 1 i i i i 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 T T 1 1 1
0 1 2 3 4 5 0 7 8 9 ID 11 12 13 14 15 16 17 18 10 20 21 22 23
Hour of day (GMT)
Mixing Ratio Bias
- - 4= ^ X
rn
cJzi
^ ^ ^ 1.
UJ
L-l
i
i i ri -j 1 i '
! 1 L
Mixing Ratio Error
-r-
-r-
-p
i
i
a
i
i
H
a
i
i
B
T
-r-
1
J
1
F
i
M
1
A
1
M
i
j
mon
I
j
i
A
1
S
1
o
1
N
1
D
Mixing Ratio Fractional Bias
T
T
^ T
-
_T_
-J-
-1-
n 1 i 1 1 1 1 1 1 1 1 r
J FMAMJ JASOND
Mixing Ratio Fractional Error
mon
Figure 3.3,1. Distribution of hourly bias by hour and hourly bias, error, fractional bias, and
fractional error for water vapor mixing ratio by month for the 36NOAM domain.
58
-------
Mixing Ratio Bias
_g_$-
H -0 0- 0
n 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r~
0 1 2 3 4 5 8 7 8 9 ID 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour of day (GMT)
Mixing Ratio Bias
-r-
ir* 1111 r~' 111 ' '
" L+J 'i1 1,-J C=l
i
^ "t3 T ^ ^
-J6 o
~r
M
Mixing Ratio Error
8
8
Mixing Ratio Fractional Bias
-T"
M
Mixing Ratio Fractional Error
Figure 3.3.2. Distribution of hourly bias by hour and hourly bias, error, fractional bias, and
fractional error for water vapor mixing ratio by month for the 12US domain.
59
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220101 AND 20220131
Figure 3.3.3. Spatial distribution of water vapor mixing ratio bias (g/kg) across ail hours for the
months of January, February, and March (top to bottom) for the 36NOAM domain.
60
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220401 AND 20220430
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220501 AND 20220531
-JrlP:
a
wv* * %
ฎx.-/y
AN V
.'v
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220601 AND 20220630
* V J *,
V-i>
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Figure 3.3.4. Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of April, May, and June (top to bottom) for the 36NOAM domain.
61
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220701 AND 20220731
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220901 AND 20220930
Figure 3.3.5. Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of July, August, and September (top to bottom) for the 36NOAM domain.
62
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221001 AND 20221031
. Q 2 %J\
. ** * ^
' >"T>_ 1 M- V - ^ - &J&
Figure 3.3.6, Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of October, November, and December (top to bottom) for the 36NOAM domain.
63
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220101 AND 20220131
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220301 AND 20220331
Figure 3.3.7. Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of January, February, and March (top to bottom) for the 121IS domain.
64
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220501 AND 20220531
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220601 AND 20220630
Figure 3.3.8. Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of April, May, and June (top to bottom) for the 12US domain.
65
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220701 AND 20220731
> A ti*. . vK'i * * . *ซ?
j. sCnViV. : w-i
, w " ฆ * m~ 0m V/^" -
s. w .$f
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220801 AND 20220831
X* * t
cv-r
>AAi- r^4r
** * ' . ^".
A * I* v t. 1 * %i <
... ....
'i.V-*Wr* A
?>
'J -
* - f\\-
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220901 AND 20220930
"1
j
%>ฆ::*ป. - ฐ *, / ,
t" ? 2*{
Figure 3.3.9, Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of July, August, and September (top to bottom) for the 12US domain.
66
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221001 AND 20221031
Figure 3.3.10. Spatial distribution of water vapor mixing ratio bias (g/kg) across all hours for the
months of October, November, and December (top to bottom) for the 12US domain,
67
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220101 AND 20220131
^ -Kwi
c [
VL
"ฃv.
w
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220201 AND 20220228
Figure 3.3.11. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of January, February, and March (top to bottom) for the 36NOAM domain.
68
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220401 AND 20220430
-K-
_jC
"; i/ , r'i
\\ v 1 j* /i, v ฆ >
c
y ซv
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220501 AND 20220531
Figure 3.3.12. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of April, May, and June (top to bottom) for the 36NOAM domain.
69
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220701 AND 20220731
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220801 AND 20220831
Figure 3.3.13. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of July, August, and September (top to bottom) for the 36NOAM domain.
70
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221001 AND 20221031
a.; f ; * 1 f
A.ir! rilfc
Figure 3.3.14, Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of October, November, and December (top to bottom) for 36NOAM domain.
71
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220101 AND 20220131
Figure 3.3.15. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of January, February, and March (top to bottom) for the 12US domain.
72
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220401 AND 20220430
Figure 3.3.16. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of April, May, and June (top to bottom) for the 12US domain.
73
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220701 AND 20220731
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20220901 AND 20220930
Figure 3.3.17. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of July, August, and September (top to bottom) for the 12US domain.
74
-------
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221001 AND 20221031
; vV: ' v
I I
<3* iซ I ฎ rfc\
- I ซ k " * -,% ฆ t- ' o
I *. : .x*. * 4 ,h iSK
* * ฆ ** s. V"\JI
' '* v.-
-i* aO,
^W" ฆ>
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221101 AND 20221130
<
1
lffc
ฆ Mt ฆ * .ป VnEBW I
"ป]#* i? Av -
WHX ^
V.
i *. 1 V\%'
. x *r > ^ <
4 r V ' . ' *. \\
i V .y:^r' \>
' * ;
W" ป' ' 3
>f -
4PY
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Mean bias of Mixing Ratio (g/kg) Date: BETWEEN 20221201 AND 20221231
%
t
> - - ! - r ฆ
^ . . vfi i*vwOsi? .v
l ___ 1 ^ ' > -r x
>,
l-iVW . _
-*ซS5&i #
T ฆ:;,
-3
-2
-1
-0.5
-0.25
0
0.25
0.5
1
2
3
Figure 3.3.18. Spatial distribution of water vapor mixing ratio bias (g/kg) across daytime hours
for the months of October, November, and December (top to bottom) for the 12US domain.
75
-------
76
-------
\
Figure 3.3.19. Hourly average distribution of observed arid water vapor mixing ratios (g/kg) for
the 12LJS domain in the Northeast, Northwest, Northern Rockies & Plains, Ohio Valley, South,
Southeast, Southwest, Upper Midwest, and West Climate Regions (respectively, top to bottom)
for each quarter.
77
-------
Climate Region
Season
Mean Obs
Mean Mod
MAE
MB
NMB
NME
RMSE
Northeast
Q1
2.98
3.26
0.46
0.28
9.49
15.44
0.65
Q2
7.61
7.82
0.74
0.21
2.8
9.74
1.02
Q3
12
12.1
0.95
0.1
0.82
7.9
1.27
Q4
4.97
5.18
0.56
0.2
4.11
11.33
0.76
N. Rockies & Plains
Q1
2.21
2.53
0.46
0.32
14.6
20.81
0.66
Q2
6.15
6.11
0.83
-0.04
-0.62
13.44
1.21
Q3
9.66
9.56
1.13
-0.1
-1.06
11.68
1.55
Q4
3.08
3.29
0.45
0.21
6.71
14.76
0.66
Northwest
Q1
4.04
4.17
0.52
0.13
3.11
12.9
0.72
Q2
5.79
5.61
0.65
-0.18
-3.17
11.22
0.91
Q3
8.04
8.18
0.97
0.14
1.76
12.03
1.31
Q4
4.5
4.73
0.57
0.23
5.15
12.65
0.77
Ohio Valley
Q1
3.31
3.59
0.51
0.28
8.6
15.39
0.7
Q2
9.23
9.3
0.97
0.07
0.75
10.48
1.32
Q3
13.49
13.35
1.1
-0.14
-1.02
8.17
1.47
Q4
4.78
4.94
0.57
0.16
3.38
11.84
0.78
South
Q1
4.7
4.9
0.63
0.21
4.37
13.35
0.9
Q2
12.13
12.05
1.25
-0.08
-0.64
10.33
1.7
Q3
14.66
14.55
1.24
-0.11
-0.78
8.47
1.64
Q4
7.21
7.35
0.81
0.14
1.93
11.17
1.15
Southeast
Q1
6.33
6.55
0.72
0.22
3.44
11.39
0.97
Q2
11.97
12.16
1.08
0.19
1.57
9.03
1.44
Q3
16.08
16.08
1.24
-0.01
-0.05
7.69
1.63
Q4
8.03
8.02
0.82
-0.01
-0.09
10.16
1.11
Southwest
Q1
2.47
2.87
0.6
0.41
16.44
24.2
0.81
Q2
4.3
4.52
0.89
0.22
5.2
20.79
1.28
Q3
9.36
9.36
1.25
0
0.04
13.34
1.66
Q4
3.79
4.08
0.63
0.29
7.58
16.63
0.84
Upper Midwest
Q1
2.06
2.22
0.35
0.17
8.11
16.9
0.51
Q2
7.02
7.15
0.8
0.13
1.85
11.39
1.18
Q3
11.35
11.31
0.98
-0.04
-0.36
8.66
1.33
Q4
3.77
3.9
0.44
0.13
3.57
11.75
0.64
West
Q1
4.45
4.45
0.7
0.01
0.17
15.73
0.98
Q2
6.25
6.18
0.84
-0.07
-1.09
13.41
1.17
Q3
9.16
9.44
1.12
0.28
3.02
12.17
1.55
Q4
5.6
5.71
0.72
0.11
1.94
12.83
1
Table 3.3.1. Mean observed, mean modeled, mean absolute error (MAE), mean bias (MB),
normalized mean bias (NMB), normalized mean error (NME), and root mean square error
(RMSE) for water vapor mixing ratio (g/kg) for the 12US domain.
78
-------
3.4 Model Performance for Precipitation
Monthly total rainfall is plotted for each grid cell to assess how well the model captures the
spatial variability and magnitude of convective and non-convective rainfall. As described earlier,
the PRISM estimations for rainfall are only within the continental United States. With lightning
assimilation in the 12US simulation mentioned earlier, the model will either trigger (suppress)
convection when lightning is observed (not observed). This assimilation is particularly useful in
constraining the model's convection scheme that at times has been observed to be inaccurately
active. WRF rainfall estimates by month are shown for all grid cells in the domain. Monthly total
estimates are shown in Figures 3.4.1 through 3.4.12 for the 36NOAM domain and Figures 3.4.13
through 3.4.24 for the 12US domain. Domain-wide biases for the 36NOAM and 12US domains
are shown in Table 3.4.1.
Overall, the model captures the general spatial patterns and magnitude of the precipitation
across the US throughout the year. Precipitation is generally underpredicted across the
southern and eastern portions of the US during the spring and winter months. There is a
general overprediction that is noted across the western US and portions of the Plains states,
particularly in areas of complex terrain (e.g., northern CA, the Rockies, etc.), especially during
the Summer months. The overprediction is particularly noticeable in the 36NOAM simulation,
where lightning assimilation data were not used and thus, convective precipitation was
unconstrained (notable domain-wide biases in summer months are up to 0.6" greater in the
36NOAM simulation versus the 12US simulation).
79
-------
Precipitation, January 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
160
140
120
100
80
60
40
20
0
PRISM
Model
Inches Inches
Difference
Inches
Figure 3.4.1. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for January for the 36NOAM domain.
80
-------
Precipitation, February 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Difference
Inches
3.4.2. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in) and the
difference (bottom) for February for the 36NOAM domain.
81
-------
Precipitation, March 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Inches
Figure 3.4.3. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for March for the 36NOAM domain.
82
-------
Precipitation, April 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Inches
Figure 3.4.4. PRISM analysis (top left) arid WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for April for the 36NOAM domain.
83
-------
Precipitation, May 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Difference
Inches
Figure 3.4.5. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for May for the 36NOAM domain.
84
-------
Precipitation, June 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Inches
Figure 3.4.6. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for June for the 36NOAM domain.
85
-------
Precipitation, July 2022
Difference
Inches
Figure 3.4.7. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for July for the 36NOAM domain.
160
140
120
100
80
60
40
20
0
Model
86
-------
Precipitation, August 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Difference
Inches
Figure 3.4.8. PRISM analysis (top left) arid WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for August for the 36NOAM domain.
87
-------
Precipitation, September 2022
PRISM
Inches Inches
Difference
160
140 -
120 -
100 -
80 -
60 -
40 -
20 -
0-
0 25 50 75 100 125 150 175
Inches
Figure 3.4.9. PRISM analysis (top left) arid WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for September for the 36NOAM domain.
Difference
160
140
120
100
80
60
40
20
0
Model
88
-------
Precipitation, October 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Difference
Inches
Figure 3.4.10. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for October for the 36NOAM domain.
89
-------
Precipitation, November 2022
160
140 -
120 -
100 -
80-
60-
40 -
20-
0--
0
PRISM
160
140
120
100
80
60
40
20
0
Model
Inches Inches
Inches
Figure 3.4.11. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for November for the 36NOAM domain.
90
-------
Precipitation, December 2022
PRISM
Model
160
0
25
50
75
100
125
150 175
6 9 12 15
Inches
160
6 9 12 15 18
Inches
Difference
0
25
50
75
100
125
150 175
-5 -4 -3 -2 -1 1 2 3 4 5
Inches
Figure 3.4,12, PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for December for the 36NOAM domain.
91
-------
Precipitation, January 2022
PRISM
Model
0
100
1 f ^5 I
200
300
400
9 12
Inches
15 18
0
-sn r, ! r
100 200
300
400
9 12
Inches
15 18
Difference
300
250
200
150
100
50
0
IX
0
100
200
300
400
-5 -4 -3 -2 -1 1 2 3 4 5
Inches
Figure 3.4.13. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for January for the 12US domain.
92
-------
Precipitation, February 2022
-1 1
Inches
Figure 3.4.14. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for February for the 12US domain.
Difference
PRISM
Model
93
-------
Precipitation, March 2022
PRISM Model
Inches Inches
300 -
250 -
200 -
150 -
100 -
50 -
0-
0 100 200 300 400
Inches
Figure 3.4.15. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for March for the 12US domain.
Difference
94
-------
Precipitation, April 2022
;ฆ> " 1 ,-=
0 100 200 300
400
^* 1 1 -
9 12
Inches
15
18
0
-sn f ! r
100 200
300
400
9 12 15
Inches
18
Difference
0
100
200
300
400
-5 -4 -3 -2 -1 1 2 3 4 5
Inches
Figure 3.4.16. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for April for the 12US domain.
95
-------
Precipitation, May 2022
Model
PRISM
9 12
Inches
9 12
Inches
Difference
Inches
Figure 3.4.17. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for May for the 12US domain.
96
-------
Precipitation, June 2022
PRISM
0
100
1 f ^5 I
200
300
400
9 12
Inches
15 18
0
-sn f ! r
100 200
300
400
9 12
Inches
15 18
Difference
WtJ
0
100
; t -a L| i
200
300
400
-5 -4 -3 -2 -1 1 2 3 4 5
Inches
Figure 3.4.18. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for June for the 12US domain.
97
-------
Precipitation, July 2022
PRISM Model
Inches Inches
300 -
250 -
200 -
150 -
100 -
50 -
0-
0 100 200 300 400
Inches
Figure 3.4.19. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for July for the 12US domain.
Difference
98
-------
Precipitation, August 2022
PRISM
Figure 3.4.20. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for August for the 12US domain.
Model
9 12
Inches
Difference
9 12
Inches
-1 1
Inches
99
-------
Precipitation, September 2022
9 12
Inches
9 12
Inches
-1 1
Inches
Figure 3.4.21. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for September for the 12US domain.
Difference
PRISM
Model
100
-------
Precipitation, October 2022
9 12
Inches
9 12
Inches
-1 1
Inches
Figure 3.4.22. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for October for the 12US domain.
Difference
PRISM
Model
101
-------
Precipitation, November 2022
PRISM
Model
0
100
1 f ^5 I
200
300
400
9 12
Inches
15 18
0
-sn f ! r
100 200
300
400
9 12 15 18
Inches
Difference
0
100
200
300
400
-5 -4 -3 -2 -1 1 2 3 4 5
Inches
Figure 3.4.23. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for November for the 12US domain.
102
-------
Precipitation, December 2022
PRISM
-1 1
Inches
Figure 3.4.24. PRISM analysis (top left) and WRF (top right) estimated monthly total rainfall (in)
and the difference (bottom) for December for the 12US domain.
Model
9 12
Inches
9 12
Inches
Difference
103
-------
Month
36NOAM
12US
January
0.12
0.19
February
0.09
0.13
March
0.15
0.13
April
0.31
0.16
May
0.24
0.02
June
0.25
0.01
July
0.55
0.02
August
0.56
-0.08
September
0.05
-0.07
October
-0.20
-0.15
November
-0.17
-0.03
December
-0.11
0.06
Table 3.4.1. Domain-wide biases (in inches) of total precipitation in the 36NOAM and 12US
domains by month.
3.5 Model Performance for Solar Radiation
Photosynthetically activated radiation (PAR) is a fraction of shortwave downward radiation and
is an important input for the biogenic emissions model for estimating isoprene (Carlton and
Baker, 2011). Isoprene emissions are important for regional ozone chemistry and play a role in
secondary organic aerosol formation. Radiation performance evaluation also gives an indirect
assessment of how well the model captures cloud formation during daylight hours.
Shortwave downward radiation estimates are compared to surface-based measurements and
shown below for the 36NOAM (Figure 3.5.1) and 12US (Figure 3.5.2) domains6.
In general, WRF slightly overpredicts shortwave radiation across all months of the year,
showing a greater spread in the overprediction during the late Spring to early Fall months.
Overall, the median bias in WRF for all months of the year is roughly 10-20 W/m2.
More variability is noted on an hourly basis as WRF overpredicts shortwave radiation across all
daytime hours. The median bias during the hours of most downward shortwave radiation is less
than 10-20 W/m2. A greater spread in the overprediction is noted during the afternoon to early
evening hours when the sun is highest in the sky. The model's inability to accurately simulate
sub grid clouds at both the 36km and 12km resolution is likely the cause of these errors.
6 As noted above, the 12US WRF simulation used lightning data assimilation whereas the 36NOAM WRF simulation
did not.
104
-------
Shortwave Radiation Bias: 36NOAM
Month
Shortwave Radiation Bias: 36NOAM
T
*r |
T 1 ,
T T ! ฆ !
0 q o , *oq000G-
T T
* !
i
i
i
ซ i
g n
T
* T
A P
' 1 i i ป i : ฆ i
i * ^ i ! i ;
" - i i
_L
< >
4 '
*
1 i J
1
1 '
J-
r
iiiiiiiiiiiiiiiii 1iiiir
00 02 04 06 08 10 12 14 16 18 20 22
Hour of day (GMT)
Figure 3.5.1. Distribution of hourly bias for shortwave radiation (W/m2) by month (top) and by
hour of the day (bottom) for the 36NOAM domain.
105
-------
Shortwave Radiation Bias: 12US 2022 w/ Lightning
c\j
E
03
5
Month
Shortwave Radiation Bias: 12US 2022 w/ Lightning
T
1 T
1 T 1
! J i 7 !
0 fl A * , I iphO
ฆ
a
4
i
i r
~ E
r
T
1
i
1
4
4
a
i
i
I T -|
ป 1
0BF
! i * 1 : : ;
i - - i ;
_L
LJ L
i
i
a
a
a
.
i
a
i
i
i
i
L "*ฆ
T i
i
~i 1 i 1 1 1 1 1i 1 1 1 1 i 1 1 1i 1 r
00 01 02 03 04 05 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour of day (GMT)
Figure 3.5.2. Distribution of hourly bias for shortwave radiation (W/rrs2) by month (top) and by
hour of the day (bottom) for the 12US domain.
4 CLIMATE REPRESENTATIVENESS OF 2022
Figures 4,1 and 4.2 show the divisional rankings for observed temperatures across the US for
2022. A climatic representation of the precipitation for 2022 is shown in Figures 4.3 and 4.4.
These types of plots are used to determine whether the meteorological conditions in a specific
year are near-normal or anomalous. Additionally, we can make determinations of their
suitability for use in photochemical modeling in terms of a specific year's conduciveness for
photochemical production of secondary pollutants.
In 2022, temperatures were average to above average for most months of the year. Much
below average temperatures were noted in the Northwest and Northern Plains during April.
Much above average temperatures were noted in the South in April through June. Record or
106
-------
near-record warmth was observed in the Northeast in August, as well as various portions of the
West, Northwest, and Southern Plains during July through October.
With regards to precipitation, 2022 was highly variable in terms of record to near-record
drought or rainfall amounts. Much of the year, most of the country varied between below
average to above average rainfall amounts, with no discernible pattern that persisted through
any season or region. Record dry conditions were noted in California during January, May, and
July. Record wet conditions were noted in the Northern Plains during December.
107
-------
Divisional Average Temperature Ranks
January 2022
Period: 1895-2022
Record Much Betow Near Above Much Record
Coldest Betow Average Average Average Above Warmest
Average Average
Divisional Average Temperature Ranks
March 2022
Period: 1895-2022
Much Betow Near Above Much
Below Average Average Average Above
Average Average
Divisional Average Temperature Ranks
May 2022
Period: 1895-2022
Divisional Average Temperature Ranks
February 2022
Period: 1895-2022
Much Betow Near Above Much
Betow Average Average Average Above
Average Average
Divisional Average Temperature Ranks
April 2022
Much Beiow Near Above Much
Below Average Average Average Above
Average Average
Divisional Average Temperature Ranks
June 2022
Period: 1895-2022
Record Much Betow Near Above Much Record Record Much Betow Near Above Much Record
Coldest Below Average Average Average Above Warmest Coldest Beiow Average Average Average Above Warmest
Average Average Average Average
Figure 4.1 Climatic temperature rankings by climate division: January to June 2022.
http://www.ncdc.noaa.gov/temp-and-precip/maps.php
108
-------
Divisional Average Temperature Ranks
July 2022
Period-1895-2022
Divisional Average Temperature Ranks
August 2022
Period: 1895-2022
Divisional Average Temperature Ranks
September 2022
Period: 1895-2022
Record Much Betow Near Above Much Recwd
Coldest Below Average Average Average Above Warmest
Average Average
Divisional Average Temperature Ranks
November 2022
Period: 1895-2022
Average
Average
Divisional Average Temperature Ranks
October 2022
Period: 1895-2022
Record Mud) Below Near Above Much Record
Coldest Below Average Average Average Above Warmest
Average Average
Divisional Average Temperature Ranks
December 2022
Period: 1895-2022
Record Much Betow Near Above Much Recwd
Coldest Below Average Average Average Above Warmest
Average Average
Record Much Below Near Above Mud) Recotd
Coldest Below Average Average Average Above Warmest
Average Average
Figure 4.2 Climatic temperature rankings by climate division: July to December 2022.
http://www.ncdc.noaa.gov/temp-and-precip/maps.php
109
-------
Divisional Precipitation Ranks
January 2022
Period: 1895-2022
Divisional Precipitation Ranks
Divisional Precipitation Ranks
March 2022
Period: 1895-2022
Divisional Precipitation Ranks
April 2022
Period: 1895-2022
Record Much Betow Near Above Much Record
Driest Below Average Average Average Above Wettest
Average Average
Divisional Precipitation Ranks
May 2022
Period: 1895-2022
Record Much Beiow Near Above Much
Dries) Betow Average Average Average Above
Average Average
Divisional Precipitation Ranks
P
%
X
ฉ
Record Much Betow Near Above Much Record
Driest Below Average Average Average Above Wettest
Average Average
Record Much Beiow Near Above Much
Driest Below Average Average Average Above
Average Average
Figure 4.3 Climatic rainfall rankings by climate division: January to June 2022.
http://www.ncdc.noaa.gov/temp-and-precip/maps.php
110
-------
Divisional Precipitation Ranks
July 2022
Period-1895-2022
Divisional Precipitation Ranks
August 2022
Period: 1895-2022
Divisional Precipitation Ranks
September 2022
Period: 1895-2022
Divisional Precipitation Ranks
October 2022
Period: 1895-2022
Record Much Betow Near Above Much Record
Driest Below Average Average Average Above Wettest
Average Average
Divisional Precipitation Ranks
Period: 1895-2022
Record Much Beiow Near Above Much
Dries) Betow Average Average Average Above
Average Average
Divisional Precipitation Ranks
December 2022
Period: 1895-2022
Record Much Betow Near Above Much Record
Driest Below Average Average Average Above Wettest
Average Average
Record Much Beiow Near Above Much
Driest Below Average Average Average Above
Average Average
Figure 4.4 Climatic rainfall rankings by climate division: July to December 2022.
https://www.ncdc.noaa.gov/sotc/
111
-------
5 REFERENCES
Boylan, J.W., Russell, A.G., 2006. PM and light extinction model performance metrics, goals, and
criteria for three-dimensional air quality models. Atmospheric Environment 40, 4946-4959.
Carlton, A.G., Baker, K.R., 2011. Photochemical Modeling of the Ozark Isoprene Volcano:
MEGAN, BEIS, and Their Impacts on Air Quality Predictions. Environmental Science &
Technology 45, 4438-4445.
Cooper, O.R., Stohl, A., Hubler, G., Hsie, E.Y., Parrish, D.D., Tuck, A.F., Kiladis, G.N., Oltmans, S.J.,
Johnson, B.J., Shapiro, M., Moody, J.L., Lefohn, A.S., 2005. Direct Transport of Midlatitude
Stratospheric Ozone into the Lower Troposphere and Marine Boundary Layer of the Pacific
Ocean. Journal of Geophysical Research - Atmospheres 110, D23310,
doi:10.1029/2005JD005783.
ENVIRON, 2008. User's Guide Comprehensive Air Quality Model with Extensions. ENVIRON
International Corporation, Novato.
Gilliam, R.C., Pleim, J.E., 2010. Performance Assessment of New Land Surface and Planetary
Boundary Layer Physics in the WRF-ARW. Journal of Applied Meteorology and Climatology 49,
760-774.
Heath, Nicholas K., Pleim, J.E., Gilliam, R., Kang, D., 2016. A simple lightning assimilation
technique for improving retrospective WRF simulations. Journal of Advances in Modeling Earth
Systems. 8. 10.1002/2016MS000735.
Langford, A.O., Reid, S.J., 1998. Dissipation and Mixing of a Small-Scale Stratospheric Intrusion
in the UpperTroposphere. Journal of Geophysical Research 103, 31265-31276.
Otte, T.L., Pleim, J.E., 2010. The Meteorology-Chemistry Interface Processor (MCIP) for the
CMAQ modeling system: updates through MCIPv3.4.1. Geoscientific Model Development 3,
243-256.
Skamarock, W.C., Klemp, J.B., Dudhia, J., Gill, D.O., Barker, D.M., Duda, M.G., Huang, X., Wang,
W., Powers, J.G., 2008. A Description of the Advanced Research WRF Version 3.
Stammer, D., F.J. Wentz, and C.L. Gentemann, 2003, Validation of Microwave Sea Surface
Temperature Measurements for Climate Purposes, J. Climate, 16, 73-87.
112
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/R-24-001
Environmental Protection Air Quality Assessment Division March 2024
Agency Research Triangle Park, NC
113
-------