Evaluation of dispersion models DIPCOT and RIMPUFF used in Decision Support 
Systems for nuclear and radiological emergency response
S. Andronopoulos
1
, L.L. Dyer
2
, E. Davakis
1
and J.G. Bartzis
3
1
National Centre for Scientific Research Demokritos, Institute of Nuclear Technology and Radiation Protection, Environmental 
Research Laboratory, Aghia Paraskevi, Greece
2
Australian Nuclear Science and Technology Organisation (ANSTO): Quality, Safety, Environment and Radiation Protection (QSERP), 
Sydney, Australia
3
University of Western Macedonia, Department of Mechanical Engineering, Kozani, Greece
Introduction
The atmospheric dispersion models DIPCOT and RIMPUFF are 
incorporated for operational use in the Decision Support System (DSS)
RODOS for nuclear emergencies. RIMPUFF is also used in the DSS 
ARGOS.
In this paper an evaluation exercise of DIPCOT and RIMPUFF is 
presented through comparisons of model-predicted with measured gamma 
radiation dose rates in air. 
Models Description
DIPCOT and RIMPUFF are Lagrangian puff models. Concentration of 
nuclides in air and gamma dose rates are calculated at a particular 
location and time by summing the contribution of all neighbouring puffs.
The differences between the two models are located in the use and pre-
processing of meteorological fields, especially the wind velocity field, in the 
movement of the puffs and in the parameterization of turbulence.
Experimental Data Base
? Dispersion of 
41
Ar released operationally from the HIFAR Research 
Reactor (ANSTO, Sydney, Australia)
? 16 different cases covering winter and summer periods of the years 
2002 and 2003 and all the atmospheric stability conditions
? Experimental data: 
41
Ar stack emission rate, meteorological data from 2 
stations and gamma dose rates from 4 monitoring stations located in a 
radius of 5 km around the reactor (15-min time intervals)
The computational domain with terrain 
elevation contours, the 
41
Ar release location, 
the meteorological stations (Met00, Met01) 
and the gamma dose rate detectors (Det9, 
Det16, Det17, Det18)
The terrain is moderately complex with hills 
and valleys and varying land cover: urban 
(south-east part), suburban (central part), 
woods (along the river) and low vegetation 
(north and south-west part) areas
310 311 312 313 314 315 316 317 318
Easting UTM (km)
-3773
-3772
-3771
-3770
-3769
-3768
-3767
-3766
-3765
No
r
thi
n
g UTM
 (k
m - RI
M
P
UFF)
MG
BT
WS
BR
HIFAR
1E-011
1E-010
3E-010
5E-010
1E-009
2E-009
5E-009
1E-008
Contour plot of calculated gamma dose rate overlaid on terrain contours for the case 
of 11/6/2003 at 01:15 EST; left DIPCOT (nGy/hr), right RIMPUFF (Gy/hr) - winter 
case with stable conditions during night time that later turned to neutral and finally 
unstable the next morning
310 311 312 313 314 315 316 317 318 319
Easting UTM (km)
-3773
-3772
-3771
-3770
-3769
-3768
-3767
-3766
-3765
-3764
No
r
t
h
ing UTM
 (k
m - RIM
P
UFF)
MG
BT
WS
BR
HIFAR
1E-011
1E-010
3E-010
5E-010
1E-009
2E-009
5E-009
1E-008
Contour plot of calculated dose  overlaid on terrain contours for the time interval 
10/6/2003, 22:30 to 11/6/2003 10:15 EST; left DIPCOT (nGy), right RIMPUFF (Gy)
310 311 312 313 314 315 316 317 318 319
Easting UTM (km)
-3773
-3772
-3771
-3770
-3769
-3768
-3767
-3766
-3765
-3764
No
r
t
h
i
n
g
 UT
M
 (
k
m
 -
 RI
M
P
UF
F
)
MG
BT
WS
BR
HIFAR
1E-011
1E-010
3E-010
5E-010
1E-009
2E-009
5E-009
1E-008
Contour plot of calculated gamma dose rate overlaid on terrain contours for the case 
of 26/6/2003 at 23:00 EST; left DIPCOT (nGy/hr), right RIMPUFF (Gy/hr) - winter 
case with neutral conditions 
31031 312313314315316317318319
Easting UTM (km)
-3773
-3772
-3771
-3770
-3769
-3768
-3767
-3766
-3765
-3764
No
r
t
h
i
n
g
 UT
M
 (
k
m
 -
 RI
M
P
UF
F
)
MG
BT
WS
BR
HIFAR
1E-011
1E-010
3E-010
5E-010
1E-009
2E-009
5E-009
1E-008
Contour plot of calculated dose overlaid on terrain contours for the time interval 
26/6/2003, 22:30 to 27/6/2003 00:00 EST; left DIPCOT (nGy), right RIMPUFF (Gy)
0
0.5
1
1.5
2
2.5
3
3.5
1:
15
1:
30
1:
45
2:
00
2:
15
2:
30
2:
45
3:
00
3:
15
3:
30
3:
45
4:
00
4:
15
4:
30
4:
45
5:
00
5:
15
5:
30
5:
45
6:
00
6:
15
6:
30
6:
45
7:
00
Time (EST)
G
a
m
m
a
 do
se
 rate
 (nG
y
/h
r)
Meas.
Calc.
0
1
2
3
4
5
6
7
8
18:
1
5
18:
3
0
18:
4
5
19:
0
0
19:
1
5
19:
3
0
19:
4
5
20:
0
0
20:
1
5
20:
3
0
20:
4
5
21:
0
0
21:
1
5
21:
3
0
21:
4
5
22:
0
0
22:
1
5
22:
3
0
22:
4
5
23:
0
0
23:
1
5
23:
3
0
23:
4
5
Time (EST)
G
a
m
m
a
 d
o
s
e
 r
a
te
 
(
n
G
y
/h
r
)
Meas .
Calc .
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1
8:
15
1
8:
30
1
8:45
1
9:
00
1
9:
15
1
9
:30
1
9
:45
2
0:
00
2
0:
15
2
0:
30
20
:
4
5
2
1:00
2
1:
15
2
1
:30
2
1
:45
2
2:0
0
2
2:
15
2
2:
30
2
2
:45
23
:
0
0
2
3:
15
2
3:
30
2
3:
45
Time (EST)
Gamma
 Do
s
e
 Ra
t
e
 
(
n
G
y
/
h
r
)
Meas.
Calc.
Comparison of 
measured and 
calculated 
gamma dose 
rates for the case 
of 25/7/2003 (up) 
22/6/2003 (down) 
for DIPCOT (left) 
and RIMPUFF 
(right)
Case DIPCOT RIMPUFF
NMSE CORR FA2 FB NMSE CORR FA2 FB
250703
0.46 0.81 0.93 0.28 0.78 0.61 0.73 -0.15
220603
0.46 0.48 0.73 0.07 5.61 0.34 0.64 -0.75
081103 0.69 0.88 0.82 0.37 0.86 0.09 0.82 -0.11
251102 1.52 0.63 0.42 0.71 0.44 0.78 0.75 -0.24
151202
2.10 0.90 0.57 0.81 0.50 0.71 0.86 0.26
060603
1.20 0.45 0.56 0.32 6.29 -0.21 0.44 1.27
140603 0.68 0.51 0.27 -0.46 6.52 -0.13 0.46 0.06
130603
1.34 0.23 0.33 -0.23 9.26 0.03 0.40 -0.39
260603
0.83 0.46 0.50 -0.12 3.61 -0.21 0.13 0.90
170603
4.62 0.13 0.00 1.36 1.42 0.18 0.40 0.84
090703 5.73 0.31 0.00 1.43 4.98 0.05 0.10 1.35
291102
2.16 0.79 0.80 0.30 1.07 0.81 0.80 -0.08
131202
13.11 1.00 0.80 1.26 13.11 1.00 0.80 1.26
171202
1.55 0.89 0.50 0.68 2.30 0.73 0.60 0.83
070603 0.58 0.42 0.51 0.22 0.38 0.60 0.64 0.10
100603
1.06 0.06 0.50 -0.25 3.92 0.01 0.29 -0.48
All
1.48 0.45 0.50 0.17 4.39 0.22 0.53 0.02
CONCLUSIONS
Contour plots of calculated gamma radiation dose rates and accumulated 
doses, reveal similar behaviors between the two models in the majority of 
cases. They also show the effects of the changing wind direction during the 
simulated periods.
Based on the statistical indices, the models performance varies between 
cases. Overall, the median values of the indices are within the suggested 
ranges according to the BOOT documentation, with the exception of one 
index. This is a satisfactory result considering the rather strict requirement of 
pairing both in space and time that has been imposed for this evaluation 
exercise.
Models Evaluation and Intercomparison
The models evaluation has been performed through comparisons of 
calculated vs. measured gamma radiation dose rates from the 4 monitoring 
stations located in a radius of 5 km around the reactor:
? time-history plots
? statistical analyses through BOOT software
Contour plots of gamma dose rates and accumulated doses have been 
used for the model intercomparison.
Statistical indices for model performance obtained by BOOT software (gamma dose rate
values paired in space and time). Red identifies cases satisfying the model acceptance 
criteria