-------
UJ
s
o
w
o: <
ui a.
*} UJ Q
2 aj i
K uj z
622
O "' '"
— o:
u in
U, LU
u, D
M —I
D'<
13 2 O
< o
UJ I— ^*
> < ~
< a:
2 S
• LU «(
O »-(
LU U X
UI
o
o
a
o
a
u
o
u
in
LU
13
X
U.
o
CO
r
a
a.
a
a.
ill ) 1 »-( C\J p-t
1 t 1 t 1
iH rH jj-l iH CJ C\J
I! 1
a.
2:
Q O ro
LU LU U *
U > - *
< ct LU rz
o: LU D x,
u en —i o
> in < 3
t o > —
: o u
: LU
UJ
in
t-
<
o
KlOOfOKlLflsTf^r^cvJ
in
z
o -*
OTOOOOOOOOOO
tflh-t— h-h-t— »— I— »— J— f-
25
-------
Q;
Z X
rH
a in
UJ UJ
a — u
uj < E
in > >-i
M 1-
o z
0 0
31— M
E ^ O
HI Q: <
xi- a
< Z uj
£ UJ >
U <
u. Z
0 0
u
z
o a
in u
a u f-t
< M CO
n. a o-
§£ "
U U. «4
r: o
• 3 co
< z o-
pn w rt
X —
UJ <
-j r Q
CO W
r- U
2
UJ -*
U Q. f>
UJ Z U X
O UJ X
< Q: 1C
a: uj x
UJ u. O O
> u- u 3
a "
in
UJ r- Z —
U, 3 < O Kl
O -J M *
< 0 r- X
uj > uj < an
tO r- K OX
< s u in o
o: n K D
u £ a x ~
> W UJ U
< X CE <
< a. uj
E
in
UJ r- Z
U. 3 •< O —
O — 1 M Kl
< Q r- X
UJ > LU < t
o > i- or
< E o: in u X
Cct D UJ o
uj r w 03 U —
•< X O <
< UJ
u. in
O uj
uj -j am
y < ox
uj i z
as x
UJ D o U
u. U U 3
U. H-l w
a <
in
a o
UJ M
1 II-
U -J <
o o *• *
uj o: J M
o: 10 x *
a. > a < j<
o Z s s
r- < Q.X
in uj o o
u 3 in 3
X -I [; —
1— 1 > LU
x >
UJ
in
a o
UJ — J M
tx •< <
uj u ^
in a: o m
SUJ -J XX
> < X
r-°§ 55
V) UJ < O O
LU ^ o
— -J (/} «
"J ^ t
X UJ
UJ
-J
UJ
o
o
in >o ^ -o
r-t W f*» CO
AJ ro eo CVJ «0 O
in 03 xT in
i-t
-O -O -O ^>
# § o? ^
in in m in
^ $Q CO W
O* pH -0 <*J
fO f^» fl rH
cj ro , o ro
i i in
i
o o o o
CJ x* (?• 0
cj ui »H r*
vr in t\i eo
w M cj PJ
CJ CJ CJ CJ
Q CO CO CO
rH i— t i-H i-(
FH r-t ^-t rH
a r co
uj o a. t
h- t'J tf) 31
a a. a uj
E E Q. I-
25
-------
5 =
Kl
0 «
UJ UI
> 3
Q; — j uj
UJ > M
O r-
°1 s
§ C M
s ^ to
M o: <
x K a
-< 2: UJ
XI ui >
U «t
u. r:
o o
^, u
S" o
(U
5U PH
M u.
o
UJ —
o 3 ** 2
<« a r-
IU > UJ <
10 p- r-
CE 3 PH
uj £ a x
> PH UJ U
<< X e <
< U, UJ
£
in
UJ p- 2
u. D < O
O -J PH
< a r-
Ul > UJ h-
< sr & tn
o; ID uj
UJ H 01 X
•< X 0 <
< UI
r
u. tn
O UJ
UI —1
u <
UJ
ct r
UJ U
U. 2Z
U_ M
tH X
°r
w
a o
UJ >~t
h— -J H*
U -J <
»-» < u
Q O
uj a: -j
Cc: uj
u. > a
0 Z
K <
(ft UI
UJ 3 VJ
X -J l-
o < r:
HH > UJ
UI
(ft
UJ — J t-t
> -J >-
a < <<
UI U
(ft Q! O
CO UI ^
O >
o a
P- —
crt iu Z
X UJ
UJ
aro
U *
1 £
N
O O
U 3
10
*
*
a£
0 \
^3
,„,
^»
X
O £
U \
13
3
a ro
O X
1 £
N
O n
O 3
-^
X *
<< *
H! 2Z
O 0
3
Kl
X *
<. M
o \
U tl
.3
^
UI
o
o
z:
P-- -O O iH
• • • •
o o in Kt
FH "O ^ f^
pH 1 *$•
I
^ PN »-* O
O* O If) f*.
ro *H *o r-*
CW o h* m
O* I*"* r^ *^
r* c\j x± ^3
^ •-* ^ «
(-« m
O4 ff1 ^ O*
O< (7- O O*
O O O O
,-, r-( r-( rH
a r oo
ui o a. i
h- v) uy jz
a. CL CL ui
E r o. h-
27
-------
u *''
< CL
fy II I
UJ U.
> U.
< M
a
u. 3 < O
O -J M
< O I-
UJ > UJ <
U *- H
< — u in
CS 3 M
uj ^ o z
> 1-1 UJ U
< x tx <
< a. LU
UJ I- Z
u. 3 < O
O -I 11
< 0 h-
UJ > LU <
13 > h-
< n a w B5 U
< X O <
< UJ
0 U
U 3
*
ar
(J N
0 E
U N
13
aii
u *
*
i ^
N
O O
U 3
n X
O <
n < U
a o
UJ CK -J
in uj
uj D in
3 < z
M > UJ
X >
UJ -J H
o: < <
UJ U
in a: o
to uj -J
o >
H-°§
in uj <
X *
s c
X *
O X
uo
g
2:
* o
cj i-i
in
-r
r«. r>. r« r^
in
o
O- *H
(0 to
CJ
o-
CO
»•
CO
o: r
iu o Q.
t- in in
a a. a
i: E a.
28
-------
SECTION'IV
CONCLUSIONS
The S02 data base collected at the Kincaid Power Plant during
1980/1981 was used to evaluate the performance or 4 rural models which
had been previously evaluated from data collected in 1975/1976 at
Clifty Creek and Muskingum River. Overall results we^e similar in many
ways to results reported earlier. PPSP clearly overpredicted concentra-
tions in almost every data category while MPSDM, MPTER and TEM exhibited
relatively small but detectable biases depending on averaging period and
data category.
Among the latter three models, MPSDM tended to overpredict the 25
highest observed concentrations while TEM and MPTER tended to underpredict
the 25 hignest observed values. For MPSDM, overpredictions tended to
decrease noticeably with increasing averaging period with 1-hour values
being overpredicted by the greatest extent. For MPTER and TEM, there
was no obvious trend toward greater underpradiction with averaging
period increases. The most noticeable contrast with results from previous
studies occured for MPTER; overprsdictions of 1-hour values ^ere obvious
for Clifty Creek and Muskingum River which were not apparent here, espe-
cially for the more unstable conditions.
Results presented previously by EPRI and those presented in this
addendum are quite close and only differ quantitatively due to slight dif-
ferences in the way the data were processed. The reader is referred to
Appendix E which compares the performance of CRSTER/MPTER reported by EPRI
with results obtained from this study for selected data categories.
29
-------
EPA plans to continue the evaluation of rural models using the tracer
data base collected at Kincaid and other high quality data bases as they
become available. EPA is also in the process of evaluating the results
from various rural studies and plans in the future to issue additional
reports summarizing the performance of rural models across data bases.
Hopefully a format can be developed that is amenable for quantitatively
distinquishing differences in performance levels among the models and among
data bases.
From a regulatory point of view, the ability of the models to
estimate the highest 3-hour and 24-hour average concentrations is
of particular interest. MPTER, MPSDM and TEM-8A were all successful
in estimating the network-wide highest observed 3-hour and 24-hour
observed values within approximately a factor of two. A more robust com-
parison using the second highest 3-hour and second highest 24-hour value
from each of the 30 stations indicated that the better performing models
were able to estimate the second highest values at individual stations
within approximately 50 percent.
30
-------
REFERENCES
1. Londergan, R. J., et al., "Evaluation of Rural Air Quality
Simulation Models."" E"PA~-450/4-83-003, Environmental Protection
Agency, Research Triangle Park, N. C., July 1982.
2. Fox, D. G. "Judging Air Quality Mode: Performance," (A Summary
of the AMS Workshop on Dispersion Model Performance, Woods, Hole,
MA, Sept. 1980). Bulletin of the Amer. Met. Soc., Volume 62,
No. 5, May 1981, pp. 599-609.
3. Cox, W. M. and G. K. Moss, "Evaluation of Rural Air Quality
Simulation Models, Addendum A: Muskingum River Data Base."
EPA-450/4-83-003a, Environmental Protection Agency, Research
Triangle Park, N. C. June 1985.
4. Reynolds, D. S. et al ., "Operational Validation of Gaussian Plume
Models at a Plains Site." EPRI EPA-3076, Project 1616-9, Electric
Power Research Institute, Palo Alto, California, 1984.
5. Bowne, N. E., et al . "Overview, Results and Conclusions for the EPRI
Plume Model Val idation and Development Project: Plains Site." EPRI
EA-3074, Project 1616-1, Electric Power Research Institute, Palo
Alto, California, 1983.
6. Cher, M., et al ., "Plume Model Validation and Development Field
Measurements - Plains Site," EPRI EA-3064, Project 1616-8. Electric
Power Reserach Institute, Palo Alto, California 1984.
7. Baldridge, K. W., "Model Evaluation Support System Users Manual."
Prepared by Computer Science Corporation for EPA, Computer Science
Corporation, Research Triangle Park, N. C. 1983.
8. Statistical Evaluation Subsystem Users Manual." Environmental
Protection Agency, Research Triangle Park, N. C., 1985.
31
-------
APPENDIX A
Statistics For 25 Highest Values
-------
a
UJ
PREOIC
OCATIOI
g —i
< a
o
a
UJ UJ
a: M
i HICHEST OBSE
UNPAIRED IH 1
N
in
UJ
U_ 13
O -J
O __
M O
o: n
< t-
i: o:
u Z
UJ
y
< o
u
UJ
_ J OJ
m o
i-
o:
O
X
IH
M
1-
o
/ERAGIM
<
MPTER
^
UJ
a
o
c
03
(1930 4 1
KIHCAID
Z _- _
V M O LU ^
u *- to re o
H D )~4 0, I-H
Z) w << I U
UJ 1— ZZ CQ O£
CS V) O O U.
a -
0 0
—
UJ I/)
u z — —
z < a. fi
UJ »-H U *
a a *
u. i: \
u. O (0
t-H U. U ID
UJ f.n
U UJ — — '
^ o a. KI
UJ \
a. < O to
I-H u D
au_~~
Q Q- —
UJ U Kl
UJ t~ «— ./
< HI iu S
Q: a Z) N.
LU UJ — J CD
> CU < 13
< a. > -
Q O -»
Ul uj U K»
O > — *
< a *
QC LiJ IU ^
UJ CO D N
> C" -J CD
< o < a
UJ
V)
o
in
CO
Kl
0
5
o
1
o
iH
*
(VI
o
0
o
l-t
o
CO
»H
1
Kt
CM
O
MICNS/ALL
( A-4A ) 76
t- tn
2
-J U.J
*t UJ
l-l
sssss
o o o o o
W *Q O iH *y
o o o a o
i i i
S § 2 5 5
O O O O O
^r^cvj
CJ CM 1
( i-H r-t 1 i~*
1
CM CM 1 1
sT Kl 1
r^ in *H sj CT» tn OJ Ut Lf» f"* TO CJ 1 <*- sX ''1 Nj" t -1 ^ '
1 t-H 1 II 1 1 1 1 1 CJ t- 1
1
1 c-t •— t IO CJ
CM CM fO iH CM CM CM rO CM i-H (\J CM CM •-* CJ rH CJ CJ f 1 CJ ri CJ J- PO fO
i-^ rH i- 1 r- 1 •- 1 i-* (— 4 j-H i-t <-) CJ CJ CJ CJ p>J OJ C-J l 1 CJ C 1 fM
dobooboooaaoooooooaoouuLJu
wiwwto 01 (rta)(nwww^tf)««(rttntnvioTW (,)«.» tn to
A-l