Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій
We studied the possibility of the combined application of screening models to assess the characteristics of sources in accidents at storage facilities for hazardous substances with complex models of atmospheric transport as part of modern decision support systems to calculate air pollution in a wide...
Saved in:
| Date: | 2022 |
|---|---|
| Main Authors: | , , , |
| Format: | Article |
| Language: | English |
| Published: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2022
|
| Subjects: | |
| Online Access: | https://journal.iasa.kpi.ua/article/view/258100 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | System research and information technologies |
| Download file: |  |
Institution
System research and information technologies| _version_ | 1867334424214896640 |
|---|---|
| author | Kovalets, Ivan Bespalov, Viacheslav Maistrenko, Svitlana Udovenko, Oleg |
| author_facet | Kovalets, Ivan Bespalov, Viacheslav Maistrenko, Svitlana Udovenko, Oleg |
| author_institution_txt_mv | [
{
"author": "Ivan Kovalets",
"institution": "Інститут проблем математичних машин і систем Національної академії наук України, Київ"
},
{
"author": "Viacheslav Bespalov",
"institution": "Інститут проблем математичних машин і систем Національної академії наук України, Київ"
},
{
"author": "Svitlana Maistrenko",
"institution": "Інститут проблем математичних машин і систем Національної академії наук України, Київ"
},
{
"author": "Oleg Udovenko",
"institution": "Український центр екологічних та водних проектів, Київ"
}
] |
| author_sort | Kovalets, Ivan |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2022-12-21T22:15:21Z |
| description | We studied the possibility of the combined application of screening models to assess the characteristics of sources in accidents at storage facilities for hazardous substances with complex models of atmospheric transport as part of modern decision support systems to calculate air pollution in a wide range of spatial and temporal scales. The evaporation time following an emergency spill, estimated by screening models, is used to set the emission intensity and calculate the atmospheric transport by the RODOS nuclear emergency response system. For the accident in Chernihiv on March 23, 2022, it was estimated that the maximum permissible concentration of ammonia 0.2 mg/m3 was exceeded at distances up to 75 km from the source. The dependence of the calculated maximum concentrations on time is close to asymptote cmax ~ t-4.5 up to 15 h after emission, which is consistent with the asymptote σ ~ t3/2 for the time dependence of the sizes of puffs following turbulent dispersion of instantaneous releases. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2022.3.04 |
| first_indexed | 2025-07-17T10:27:49Z |
| format | Article |
| fulltext |
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko, 2022
57 Системні дослідження та інформаційні технології, 2022, № 3
UDC 504.3.054
DOI: 10.20535/SRIT.2308-8893.2022.3.04
THE USE OF ENVIRONMENTAL DECISION SUPPORT
SYSTEMS FOR MODELING OF ATMOSPHERIC POLLUTION
FOLLOWING THE CHEMICAL ACCIDENTS
I.V. KOVALETS, V.P. BESPALOV, S.Ya. MAISTRENKO, O.I. UDOVENKO
Abstract. We studied the possibility of the combined application of screening mod-
els to assess the characteristics of sources in accidents at storage facilities for haz-
ardous substances with complex models of atmospheric transport as part of modern
decision support systems to calculate air pollution in a wide range of spatial and
temporal scales. The evaporation time following an emergency spill, estimated by
screening models, is used to set the emission intensity and calculate the atmospheric
transport by the RODOS nuclear emergency response system. For the accident in
Chernihiv on March 23, 2022, it was estimated that the maximum permissible con-
centration of ammonia 0.2 mg/m3 was exceeded at distances up to 75 km from the
source. The dependence of the calculated maximum concentrations on time is close
to asymptote 5.4
max
tc up to 15 h after emission, which is consistent with the as-
ymptote 3/2t for the time dependence of the sizes of puffs following turbulent
dispersion of instantaneous releases.
Keywords: atmospheric dispersion, RODOS system, hazardous substances, ammo-
nia, LASAT, DIPCOT, “Povіtrya” system.
INTRODUCTION
The military aggression of Russia has led, in particular, to numerous cases of
man-made accidents in Ukraine with the emission of potentially hazardous sub-
stances (HS) into the atmosphere. In many cases, hazardous chemicals such as
ammonia, chlorine, and others are stored in liquefied form. As a result of damage
to a tank or pipeline, a rapid release of the gaseous phase into the atmosphere
(primary cloud) and a spill of the liquid phase of the corresponding chemicals
onto the underlying surface with subsequent evaporation (secondary cloud) can
occur simultaneously. Operational forecasting of damage zones as a result of the
spread of clouds formed after the spill of HS in the world and in Ukraine is car-
ried out according to simplified methods, which are called “screening models”
[1]. Such models describe both the formation of an emission source through the
spread and evaporation of a spill and the spread of a cloud of HS. Examples of
such screening models are works [1–4]. In many cases, the initial results of
screening models are only the integral characteristics of the cloud of HS, such as
the radius of the affected area. When propagating at a distance of more than 10–
20 km, it is necessary to take into account the influence of spatially and tempo-
rally variable meteorological conditions on the formation of pollution concentra-
tion fields, which is possible by using state-of-art atmospheric transport models
(ATMs) that simulate cloud propagation based on data from numerical weather
forecast models and are able to calculate atmospheric transport at distances from
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 58
~0.1 to ~1000 km. Examples of modern ATMs are the LASAT [5] and DIPCOT
[6] models as part of the EU nuclear emergency response system RODOS [7], the
WRF-CHEM and CMAQ Euler models [8], the FLEXPART Lagrangian model,
and the CALPUFF Lagrangian-Eulerian model [9] and others. All of the above
models are widely used in the world both for forecasting pollution in real-time
and for analyzing environmental risks associated with possible emissions of pol-
lutants into the atmosphere. In a number of cases, intermediate results of screen-
ing model calculations can be used to set the source characteristics in atmospheric
transport models for further modeling the distribution of a cloud of HS. The pur-
pose of this work is to study the possibility of combined use of screening models
for assessing the characteristics of emission sources during accidents at hazardous
chemicals storage and transportation facilities with the state-of-art atmospheric
transport models for calculating atmospheric pollution over a wide range of spa-
tial and temporal scales.
METHODS
In this work, we use the atmospheric transport model of the RODOS system of
the European Union for nuclear emergency response [7]. It was recently imple-
mented in Ukraine to predict the consequences of radiation accidents and analyze
the risks associated with possible accidents [10, 11]. The RODOS system uses
several atmospheric transport models. In this work, we used the LASAT model of
Lagrangian particles [5], which is the most resource-demanding, but at the same
time the most suitable for application in the case of emissions near the Earth’s
surface. In the LASAT model, particles are considered idealized material points,
which are not assigned a spatial size. Other atmospheric models of the RODOS
system use so-called “puffs” — instantaneous emissions that represent long-term
emission. Each puff is assigned a Gaussian distribution of matter in space, and the
vertical and horizontal size of the puff is determined by the dispersions of this
distribution in the horizontal and vertical planes 22 , vh . The distribution disper-
sions increase with time, but when propagating in flows with a large velocity
shift, for example, near the Earth’s surface, the symmetry of the distribution in real
clouds is violated, which may be not well described by the corresponding models.
The input meteorological data of the RODOS system in Ukraine are the
results of numerical weather forecasting of the WRF–Ukraine system, which
operates at the Ukrainian Hydrometeorological Center and uses the modern WRF
meteorological model for calculations [12]. The weather prediction data of the
WRF–Ukraine system have been repeatedly verified on the basis of
hydrometeorological measurements in Ukraine [13].
The source parameters were calculated using the “Povitrya” system [14],
which is in the public domain [15]. The system “Povіtrya” implements screening
models for predicting damage zones during HS emissions [3, 4], the intermediate
results of which are the mass of the primary cloud (m1) and the evaporation time
(τ) of the secondary cloud. The mass of the primary cloud depends, in particular,
on the boiling point of the given substance, the heat of boiling, the specific heat of
the substance, and the air temperature. Since the total mass of matter in the
container m0 is set by the user, the emission intensity qs during the formation of a
secondary cloud by evaporation can obviously be calculated by the formula:
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 59
.
ττ
102 mmm
qx
(1)
Here 2m is the mass of the secondary cloud. For most substances in the
model [3], the mass of the primary cloud is neglected, and then 02 mm . In [16],
examples of the application of Ukrainian and foreign models for calculating the
evaporation rate sq and the mass of the primary cloud and their comparison are
given. It is shown that Ukrainian [3, 4] and foreign [2] models give comparable
results.
Fig. 1 presents a flowchart for the calculation of air pollution based on the
combined use of screening models to assess the characteristics of sources of
emissions of hazardous chemicals with state-of-art models of atmospheric
transport. For the calculations of the screening model, it is necessary to set a
relatively small number of input parameters: the volume of the spill, the
conditions of the spill (was there a spill containment pallet or not), the storage
conditions of the substance (under pressure or not), the type of terrain (for
example, “flat terrain”), the type of vegetation (for example “grassland”),
meteorological conditions at the time of the spill (wind speed and direction,
surface air temperature, soil temperature if available, atmospheric stability). The
masses of primary and secondary clouds calculated on the basis of the screening
model, as well as the evaporation time, are used to set the emission intensity in
the atmospheric transport model according to the formula (1). The mass of the
primary cloud is accounted for as an instantaneous or short-term emission. The
calculated characteristics of the source are transferred to the ATM, which
calculates the spread of atmospheric pollutants taking into account meteorological
conditions changing in time and space. In this work, the transfer of the estimated
characteristics of the emission source to the ATM was performed manually. The
result of the ATM calculation is the field of near-surface concentration of the
pollutant distributed in time and space.
MODEL VERIFICATION
Let us compare the results of modeling the depth of propagation of a chlorine
cloud after a rapid spill of 67 tons of this substance, performed by the “Povіtrya”
Fig. 1. Flowchart of calculation of air pollution based on the combined application of
screening models to assess the characteristics of sources of emissions of hazardous
chemicals with state-of-art models of atmospheric transport
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 60
system, with calculations of international atmospheric transport models [17], spe-
cially designed to calculate the propagation of heavy gases, such as chlorine. The
meteorological conditions of the modeling scenario [17] were as follows: wind
speed of 3 m/s, stability category ‘F’ (very stable atmosphere), and air tempera-
ture of 25 °С. When modeling such an event by the “Povіtrya” system according
to the method [3], the depth of the cloud propagation zone is 20 km, and the
evaporation time is 90 min.
The range of concentrations calculated by eight models [17] at different dis-
tances from the source is shown in Fig. 2. As can be seen from the presented data,
the range of concentrations calculated by the models in [17] for a distance of 25
km is from 2 to 30 ppm. According to the American Industrial Hygiene Associa-
tion (AIHA), this range includes the maximum concentration that does not cause
irreversible health effects when exposed for one hour ERPG2=3 ppm, and the
maximum concentration of chlorine that does not threaten life ERPG3 20 ppm,
[18]. Therefore, according to the calculations of the models in [17], at a distance
of 25 km, it is necessary to take countermeasures for the event under considera-
tion, which is consistent with the cloud propagation depth of 20 km according to
the results of the “Povіtrya” system.
Fig. 2 also shows the results of calculating the maximum chlorine concentra-
tions by the RODOS system at the same distances as the models in [17]. The
emission source in the RODOS system was set according to the formula (1),
where τ=90 min according to the calculation of the “Povіtrya” system, and the
mass of the primary cloud was neglected. The emission height was assumed to be
10 m since the emission height in the RODOS system cannot be less than 10 m.
As can be seen from the data presented in the figure, the concentrations calculated
by RODOS are underestimated compared to other models according to [17] and
are outside the ranges concentrations of these models up to a distance of 5 km.
c,
p
pm
v
x, km
Fig. 2. Maximum concentrations of chlorine in the air calculated by the RODOS system
on the cloud symmetry axis for the conditions of the computational experiment according
to [17] and the duration of the source calculated by the “Povіtrya” system; concentration
ranges according to the calculations of 8 models in [17] are shown by vertical lines
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 61
This difference is most likely due to buoyancy effects that affect meteorological
fields in the vicinity of the cloud when the density of the cloud is significantly
different from the density of the surrounding air [19]. These effects may be im-
portant for modeling the propagation of chlorine at high concentrations since the
latter is a heavy gas whose density at atmospheric pressure is about 2.2 times that
of air. Buoyancy effects are taken into account by models [17], which are spe-
cially designed for calculating heavy gases and are ignored by both the RODOS
system and almost all ATMs used to calculate atmospheric transport at a distance
of more than 30 km since taking into account buoyancy effects requires separate
approaches to modeling.
Let us evaluate the significance of buoyancy effects for this problem. In
[19], a parameter was introduced that characterizes the significance of such ef-
fects depending on the emission conditions and meteorological conditions. Ac-
cording to [19], buoyancy effects can be neglected if the following condition is
satisfied:
15.0
)/g( 3/1
00
U
Dq
. (2)
Here U is the wind speed, D is the source diameter, 0q is the volumetric gas
flow rates in the source (m3/s), aagg /)(g0 is buoyancy, g is the
gravitational acceleration, g is the gas density at atmospheric pressure, a is
the air density. To use formula (2), we estimate the parameters: 12g0 m/s2,
5.40 q m3/s (the estimate of the evaporation time according to the data of the
“Povіtrya” system was used), 3U m/s. To estimate the source diameter, we
use the approach [3], in which it is assumed that during a spill on the real surface
of the Earth, the spill diameter ceases to increase at a spill basin height of less
than 0.05 m due to the impact of minor surface irregularities. Using this estimate,
it is possible to obtain an estimate of the spill diameter 33D m. Substituting
these data into formula (2), we obtain 4.0 , so the effects of buoyancy are im-
portant for this problem. When estimating pollution levels near the source the
buoyancy effects could be properly accounted for by using specialized models of
the heavy gas dispersion, such as those used in [17].
As can be seen from Fig. 2, at distances greater than 5 km, the results calcu-
lated by RODOS fall within the range of model calculations [17]. If at a distance
of 5 km the results of RODOS are at the lower limit of the calculation range of
other models, then at a distance of 25 km they are already inside the range. The
explanation for this is that the effects of buoyancy decrease with distance due to
the mixing of the cloud with the surrounding air [19]. As a result, the distance to a
dangerous concentration of ERPG2=3 ppm according to the calculations of the
RODOS system is 38 km, which is within a significant interval of the correspond-
ing distances according to the models [17]. The lower limit of this interval is
≈20 km (obtained by interpolating the lower limits of the intervals depicted in
Fig. 2 to a value of 3 ppm), and the upper limit is ≈46 km (obtained by extrapolat-
ing the upper limits of the intervals in Fig. 2 to a value of 3 ppm).
As can be seen from Fig. 2, the difference between the calculations of all
models is large — almost 10 times, and the level of discrepancy practically does
not decrease with distance, although the effects of buoyancy cease to affect cloud
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 62
propagation. The corresponding discrepancies between the distances from the
source to the places where the fixed concentration is reached are also very signifi-
cant, as mentioned above. In addition to negative cloud buoyancy effects and dif-
ferences in the results of evaporation submodels, such differences in calculated
concentrations could be explained by the difficulties in accurately calculating the
propagation of pollutants near the surface, where horizontal velocity gradients are
large. This leads to the effect of “shear dispersion” and the corresponding addi-
tional dilution of the substance concentration [20].
RESULTS OF CALCULATIONS OF THE AMMONIA EMISSION IN
CHERNIHIV
By using the proposed methodology, we carried out modeling of the spread of
ammonia vapors after damage to the ammonia pipeline during the shelling of one
of the industrial facilities in Chernihiv on March 23, 2002 [21]. Since the total
volume of ammonia emission is unknown, in this work the estimate m0=50 tons
was accepted. Such a mass corresponds to a spill through a hole with a diameter
of 10 cm at a pressure in the container of 100 psig = 689 kPa for about 4 min [17].
The pressure of 689 kPa is close to the typical pressure in ammonia pipelines of
862 kPa [22]; therefore, the error in the given estimate is determined mainly by
the unknown spill time.
Fig. 3 shows the results of calculations of the “Povіtrya” system for the sce-
nario under consideration. The meteorological conditions in the calculations were
identical to those observed on the day and time of the accident: northwest wind,
wind speed 1 m/s, air temperature 16 °C, stability category C. The calculated
cloud propagation depth was 3 km. The calculated evaporation time according to
Fig. 3. The results of the calculation by the “Povіtrya” system of the consequences of an
accident in Chernihiv with a leak of 50 tons of ammonia
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 63
the model [3] was 84 min, and the corresponding value obtained according to
the model [4] 62 min. According to the calculations, the mass of the primary
cloud was m1=11.7 tons, that is, approximately 23% of the total emission volume.
To simulate the propagation of the cloud by the RODOS system, the evapo-
ration time of the cloud was taken to be 60 min. The simulation was carried
out on the basis of meteorological data calculated by the WRF–Ukraine weather
forecasting system. The RODOS system performed calculations on two computa-
tional grids:
1) calculation with the minimum resolution of the spatial grid, which is equal
to 50 m, and the total size of the calculation area is 40×40 km (hereinafter re-
ferred to as “grid 1”);
2) calculation with a coarser minimum resolution of the spatial grid, which is
equal to 500 m, and the total size of the computational area is 400×400 km (here-
inafter referred to as “grid 2”).
Fig. 4 shows the calculated concentration field one hour after the start of the
emission when the concentration values in the near zone around the emission are
maximum (this is the point at which the emission stops). According to [18], the
maximum concentration of ammonia that does not threaten the occurrence of irre-
versible or other serious health effects is ERPG2=150 ppm=105 mg/m3. The
isoline of the corresponding concentration is shown in Fig. 3. The maximum dis-
tance from the emission point to this isoline, according to RODOS calculations,
was: 8.1L km. The corresponding value of the cloud propagation depth calcu-
lated by the “Povіtrya” system according to the model [3] is 3 km. This discrep-
ancy can be considered small given the typical discrepancies between models dis-
cussed in the previous section.
Fig. 4 shows a cell with the maximum concentration value, which is equal to
910 mg/m3 and is reached at a distance of 200 m from the source. According to [18],
Fig. 4. Ground air pollutant concentrations calculated by the RODOS system (mg/m3) as
a result of the emission of 50 tons of ammonia one hour after the start of the emission;
the calculation was performed on grid 1 for meteorological conditions on March 23, 22;
dashed isoline corresponds to the concentration of 150 ppm=105 mg/m3 (normative value
of ERPG2 according to [18])
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 64
ammonia concentrations of 1050 mg/m3 and higher are deadly. In the considered
scenario, such concentrations are not achieved, which is consistent with the above
media report that, according to preliminary data, there were no casualties caused
by this event. But it should be noted that, according to the information [3], a con-
servative estimate when calculating an accident at an ammonia pipeline or
a product pipeline can be considered an outflow of a substance in the amount of
500 tons. When calculating such a conservative scenario by the RODOS system
for meteorological conditions on March 23, 2022, the size of the affected area in
the direction of the wind, which corresponds to the above value of the lethal con-
centration, is 1.8 km.
Fig. 5 shows the isolines of the concentration of 0.2 mg/m3 calculated by the
RODOS system on grid 2 for different time moments after the accident. Accord-
ing to the Order of the Ministry of Health of Ukraine No. 52 dated December 14,
2020, the value of 0.2 mg/m3 for ammonia is the maximum permissible concen-
tration (MPC). As can be seen from the location of the isolines shown in Fig. 5, at
first the cloud moves to the southeast, but over time the wind changes direction,
and the cloud continues to move in a northeasterly direction. MPC exceedances
are observed during 9.5 hours of cloud movement at distances up to 75 km, after
which concentrations become lower than MPC. Owing to the dilution of the cloud
and the corresponding increase in its horizontal and vertical dimensions, the
maximum concentrations in the cloud rapidly decrease. The maximum value
achieved within each isoline depicted in Fig. 5 is also shown in the figure. Calcu-
Fig. 5. Isolines of surface concentrations of ammonia corresponding to the standard
MPC=0.2 mg/m3, calculated by the RODOS system on grid 2 for different time moments
after the accident in Chernihiv (1, 5, 7, 9.5 h) for meteorological conditions on March 23,
2022. For each time, the figure shows the maximum concentrations in the cloud
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 65
lation on a coarse grid leads to a strong underestimation of the concentration near
the source: one hour after the start of the emission, the maximum value on grid 2
is 212 mg/m3, while on grid 1 the corresponding value is 910 mg/m3. However,
over time, this difference disappears. For example, 5 hours after the start of the
emission, the maximum concentrations calculated on both grids differ by less
than 10%.
The time dependence of the maximum concentrations calculated by the
RODOS system on grids 1 and 2 with an interval of 1 hour is shown in Fig. 6.
Since the results are presented only for the times when the cloud is in the compu-
tational domain, the time series corresponding to the plot for grid 2 is longer than
for grid 1. To test the sensitivity of the results to the model used, in addition to the
LASAT model that was predominantly used in this work. Fig. 6 shows the results
of the DIPCOT model, also available in the RODOS system. Although DIPCOT
also uses a Lagrangian approach to modeling stochastic particle motions in a tur-
bulent medium, in contrast to LASAT, DIPCOT particles are characterized by
finite sizes, that is, they are “puffs”, as discussed above.
From the graphs presented, it can be seen that in all calculations three time
intervals can be distinguished: 1) 21 t hours, the maximum concentration
decreases rapidly; 2) ht 42 , the concentration decreases slowly; 3) ht 4 ,
the concentration drops rapidly again. The interval 10 t h is not considered,
since during this period an emission occurs and the maximum concentration is
approximately constant. As can be seen from the above results, during intervals 1)
and 3) the maximum concentration decreases with time approximately as:
5.4
max ~ tc , (3)
and the corresponding asymptotics is also shown on the graph. This dependence
of the maximum concentration on time is in accordance with the classical results
of the theory of turbulent motion of particles, according to which the vertical and
Fig. 6. Dependence of the highest concentrations in the computational area on time after
the start of the emission according to the results of RODOS calculations on grids 1 and 2;
for grid 2, the results of two RODOS models are presented: LASAT and DIPCOT; the
results are presented only for the times when the cloud is inside the computational
domain
t, h
c m
ax
~t-4.5
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 66
horizontal dimensions of a medium particle after an instantaneous emission in-
crease following the dependence [23]:
2/3~ t . (4)
Indeed, in the case of an instantaneous emission of a substance, if the height
of the emission (puff) above the Earth’s surface is much less than its characteristic
vertical size σv, that is typical for the long-term existence of the puff [24], the
maximum concentration on the Earth’s surface will be determined by the depend-
ence 12
max ~ vhc . Substituting expression (4) into the last relation, we obtain
the asymptotic dependence (3), which agrees with the graphs in Fig. 5 at time in-
tervals 1) and 3). Some deviations from this dependence on time intervals 1)
and 3) are explained by the fact that for turbulent dispersion in real meteorologi-
cal fields that vary in time and space, no analytical relationships for the sizes of
puffs with time can be fulfilled exactly.
It should be noted that, according to theoretical studies in [23], when the du-
ration of the movement of puffs is much longer than the Lagrangian integral time
scale of turbulence in the atmosphere (i.e., several hours after the emission), the
law of growth of puff’s sizes (4) should be replaced by more slow dependence
3/4~ t . Then the time dependence of the maximum concentration would be de-
scribed by the relation 25.2
max ~ tc , which obviously does not occur according to
the calculations of both RODOS models, the data of which are presented in Fig. 6.
Such a discrepancy between theory and model calculations may be the subject of
further research.
A sharp slowdown in the rate of fall of the maximum concentration occurs in
time interval 2) from the second to the fourth hour after the emission, exactly
when the direction of the cloud movement changes from southeast to northeast.
This is associated with the advancement of the pressure trough from the north and
the corresponding change in wind direction. At this time interval, the wind speed
drops below 0.5 m/s for a short time. Hence calm conditions arise when the existing
cloud moves almost without mixing. Within two hours, a stable transport is estab-
lished in the northeast direction, the wind speed increases to 3.5 m/s, and mixing
is restored.
CONCLUSIONS
In the present work, we studied the possibility of the combined application of
screening models for assessing the characteristics of emission sources during ac-
cidents at hazardous chemicals storage and transportation facilities with complex
models of atmospheric transport, functioning as part of state-of-art environmental
decision support systems, for calculating air pollution in a wide range of spatial
and time scales – from ~100 m to ~100 km. The evaporation time and the mass of
the primary cloud in case of an accidental spill of a substance are calculated by
using the web system “Povitrya” based on the calculation of screening models for
assessing affected areas in emergency situations at industrial facilities. These pa-
rameters are then used to set the emission intensity and calculate the atmospheric
transport by the RODOS system of the European Union for nuclear emergency
response. The advantage of using simplified screening models in comparison with
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 67
the more complex models of releases results from the fact that relatively few pa-
rameters need to be specified for the application of simplified models (volume of
the spill, storage and meteorological conditions, type of terrain). The values of the
corresponding parameters could be estimated in real-time.
The chlorine concentrations calculated in this way for the scenario of an ac-
cidental release are compared with the results of international models specially
designed to calculate the spread of heavy gases. It was found that the concentra-
tions calculated by the RODOS system for this scenario are consistent with the
results of other models at distances of more than 5 km and are underestimated at
shorter distances. This is due to the effect of negative buoyancy on the propaga-
tion of the chlorine cloud. With distance, the effects of buoyancy cease to affect
cloud propagation, and as a result, the distance to a dangerous concentration, ac-
cording to RODOS calculations, is within the interval of the results of other models.
The proposed approach was used to analyze the spread of ammonia vapor as
a result of damage to the pipeline in Chernihiv on March 23, 2022, caused by hos-
tilities. The amount of ammonia emission for this event is unknown. A conserva-
tive estimate of the emission of 500 tons for this event leads to an estimate of the
length of the affected area according to the RODOS system up to 1.8 km from the
source in the direction of the wind. Based on information from the media about
the absence of victims, an assumption was made about the emission of 50 tons of
ammonia. Based on the calculations of the RODOS system using the data of the
WRF–Ukraine meteorological forecasting system, it was estimated that, due to
the spread of the cloud, the maximum permissible concentration of ammonia,
MPC 2.0 mg/m3, was exceeded at distances up to 75 km from the point of emis-
sion. Exceeding the MPC in a certain area means that the cloud could have a dan-
gerous effect on the organisms located in the corresponding area. An assessment
of the degree of such exposure can be made based on the results of the models
used in this work, but requires the involvement of experts in toxicology and is not
the subject of this article.
The temporal behavior of the highest concentrations calculated by the
RODOS system was analyzed. The dependence of maximum concentrations on
time is asymptotic, close to 5.4
max ~ tc , which is consistent with the classical
asymptotic relation 2/3~ t for the time dependence of the size of puffs (instant
emissions) during their turbulent mixing in the atmosphere on time intervals
smaller than the Lagrangian integral time scale of turbulence Lt . However, since
the propagation was calculated for time intervals up to 15 hours, which is more
than Lt , in future theoretical studies on atmospheric dispersion, it would be
necessary to clarify the possibility of fulfilling the dependence 2/3~ t on large
time intervals.
The method of combining screening models for evaluating the characteristics
of sources in the case of severe man-made accidents at industrial facilities with
state-of-art models of atmospheric transport studied in this work can be used to
forecast the consequences of such events for the environment and people, for the
development of new and improved decision support systems for environmental
safety management, and to assess risks due to operation of industrial objects.
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 68
REFERENCES
1. J.E. Cudney-Black et al., Hazardous Chemical Inventory Guidelines, Purpose and
Process. Sandia National Lab., Albuquerque, New Mexico, USA, Rep. SAND2021-
8639, 2021. doi: 10.2172/1821974.
2. C.J.H. van den Bosch and R.A.P.M. Weterings, Eds., Methods for the calculation of
physical effects due to releases of hazardous materials (liquids and gases) – ‘Yellow
Book’ (Publication series on dangerous substances); 3d ed. The Hague, The Nether-
lands: Director-General of Labor : Committee for the Prevention of Disasters, 2005.
3. “Recommendations on forecasting the consequences of spills (emissions) of
hazardous chemicals at industrial facilities and transport,” (in Ukrainian), Ministry of
Ukraine of Emergencies and Affairs of Population Protection from the Conse-
quences of Chernobyl catastrophe. Kyiv, 2001. Accessed on: Jun 6, 2022. [online].
Available: https://zakon.rada.gov.ua/laws/show/z0326-01.
4. “Method of assessment of chemical situation following accidents at facilities con-
taining toxic substances,” (in Russian), Ministry of Defense of USSR. M., 1989.
5. U. Janicke and L. Janicke, “Lagrangian particle modeling for regulatory purposes;
A survey of recent developments in Germany,” in Proc. of the 11th International
Conference on Harmonisation within Atmospheric Dispersion Modelling for Regula-
tory Purposes, HARMO 2007, Cambridge, UK, 2007, pp. 109–113.
6. S. Andronopoulos, E. Davakis, J.G. Bartzis, and I. Kovalets, “RODOS meteorologi-
cal pre-processor and atmospheric dispersion model DIPCOT: a model suite for ra-
dionuclides dispersion in complex terrain,” Radioprotection, vol. 45, no. 5, pp. 77–84,
2010. doi: 10.1051/radiopro/2010017.
7. C. Landman, J. Pasler-Sauer, and W. Raskob, “The decision support system
RODOS,” in The Risks of Nuclear Energy Technology. Science Policy Reports. Hei-
delberg, Germany: Springer-Berlin, 2014, pp. 337–348. doi: 10.1007/978-3-642-
55116-1_21.
8. C. Gao et al., “Two-way coupled meteorology and air quality models in Asia: a sys-
tematic review and meta-analysis of impacts of aerosol feedbacks on meteorology
and air quality,” Atmospheric Chemistry and Physics, vol. 22, pp. 5265–5329, 2002.
doi: 10.5194/acp-22-5265-2022.
9. S. Khan and H. Quamrul, “Review of developments in air quality modelling and air
quality dispersion models,” J. of Environmental Engineering and Science, vol. 16,
no. 1, pp. 1–10, 2021. doi: 10.1680/jenes.20.00004.
10. I. Kovalets and O. Romanenko, “Use of Nuclear Emergency Response System for
Assessment of Transboundary Transfer and Radiological Risks of the Potential Acci-
dental Releases at Khmelnitsky NPP,” Advances in Intelligent Systems and Comput-
ing, vol. 1265, pp. 3–12, 2021. doi: 0.1007/978-3-030-58124-4_1.
11. Y. Kyrylenko et al., “Source Term Modelling for Event with Liquid Radioactive Ma-
terials Spill,” Studies in Systems, Decision and Control, vol. 298, pp. 261–279, 2020.
doi: 10.1007/978-3-030-48583-2_17.
12. J.G. Powers et al., “The Weather Research and Forecasting Model: Overview, Sys-
tem Efforts, and Future Directions,” Bulletin of the American Meteorological Soci-
ety, vol. 98, no. 8, pp. 1717–1737, 2017. doi: 10.1175/BAMS-D-15-00308.1.
13. A. Khalchenkov and I. Kovalets, “Evaluation of Spectral/Grid Nudging Methods for
Weather Analysis and Forecasting in Kyiv Region with the Use of WRF Mesoscale
Meteorological Model,” Advances in Intelligent Systems and Computing, vol. 1265,
pp. 13–23, 2021. doi: 10.1007/978-3-030-58124-4_2.
14. I.V. Kovalets et al., “Povitrya web-based software system for operational forecasting
of atmospheric pollution after manmade accidents in Ukraine,” Science and Innova-
tion, vol. 13, no. 6, pp. 11–22, doi: 10.15407/scin13.06.013.
The use of environmental decision support systems for modeling of atmospheric pollution …
Системні дослідження та інформаційні технології, 2022, № 3 69
15. System “Povitrya”. [Online]. Available: https://cloud-2.bitp.kiev.ua/airsystem_ eng-
lish/ airsystem.html.
16. M. Oksanych, V.P. Bespalov, and I.V. Kovalets, “Comparison of methods for calcu-
lating the parameters of evaporation of hazardous liquids from the surface of the spill
basin on a hard surface,” (in Ukrainian), Mathematical Machines and Systems, no. 2,
p. 49–59, 2017.
17. S.R. Hanna, O.R. Hansen, M. Ichard, and D. Strimaitis, “CFD model simulation of
dispersion from chlorine railcar releases in industrial and urban areas,” Atmospheric
Environment, vol. 43, no. 2, pp. 262–270, 2009. doi: 10.1016/j.atmosenv. 2008.09.081.
18. OSHA Occupational Chemical Database, US Department of Labor, Occupational
Safety and Health Administration (OSHA), May 2022. [Online]. Available:
https://www.osha.gov/chemicaldata.
19. R.E. Britter, “Atmospheric dispersion of dense gases,” Annual Reviews of Fluid Me-
chanics, vol. 21, pp. 317–344, 1989. doi: 10.1146/annurev.fl.21.010189.001533.
20. A.H. Kumar, S.J. Thomson, T.R. Powers, and D.M. Harris, “Taylor dispersion of
elongated rods,” Physics Review Fluids, vol. 6, no. 9, Sept. 2021, Art. no. 094501.
doi: 10.1103/PhysRevFluids.6.094501.
21. “Aggressors’ shell damaged an ammonia pipeline in Chernihiv,” Ukrayinska
Pravda, 23.03.2022. [Online]. Available: https://www.pravda.com.ua/eng/news/2022
/03/23/7334017/.
22. S. Papavinasam, “Chapter 2 - Oil and Gas Industry Network,” in Corrosion Control
in the Oil and Gas Industry, S. Papavinasam, Ed., Amsterdam, The Netherlands,
Gulf Professional Publishing, 2014, ch. 2, pp. 41–131, doi: 10.1016/B978-0-12-
397022-0.00002-9.
23. T. Mikkelsen, S. Thykier-Nielsen, and S. Hoe, “Chapter 2.16 Medium-range puff
growth,” in Air Pollution Modeling and Its Application XVIII (Developments in En-
vironmental Science, vol. 6), C. Borrego and E. Renner, Eds., Amsterdam, The
Netherlands, Elsevier, 2007, pp. 243–252, doi: 10.1016/S1474-8177(07)06216-X.
24. T. Mikkelsen, “Atmospheric Dispersion: Basic,” in Encyclopedia of Environmetrics
Vol.1, A.H. El Shaarawi, W.W. Piegorsch, Eds., Chichester, UK, John Wiley &
Sons, 2002, pp. 85–106, doi: 10.1002/9780470057339.vaa025m.
Received 06.06.2022
INFORMATION ON THE ARTICLE
Ivan V. Kovalets, ORCID: 0000-0002-5553-3299, The Institute of Mathematical Ma-
chines and Systems Problems of the Ukraine National Academy of Science, Ukraine, e-
mail: ivkov084@gmail.com
Viacheslav P. Bespalov, ORCID: 0000-0001-6841-1533, The Institute of Mathematical
Machines and Systems Problems of the Ukraine National Academy of Science, Ukraine,
e-mail: bespalovviacheslav@gmail.com
Svitlana Ya. Maistrenko, ORCID: 0000-0002-1316-9732, The Institute of Mathematical
Machines and Systems Problems of the Ukraine National Academy of Science, Ukraine,
e-mail: maistrsv@ukr.net
Oleg I. Udovenko, ORCID: 0000-0001-5679-8291, Ukrainian Centre for Water and
Ecology Issues, Ukraine, e-mail: oleg.udovenko@gmail.com
ВИКОРИСТАННЯ СИСТЕМ ПІДТРИМАННЯ РІШЕНЬ З ЕКОЛОГІЧНОЇ
БЕЗПЕКИ ДЛЯ МОДЕЛЮВАННЯ ЗАБРУДНЕННЯ АТМОСФЕРИ
ВНАСЛІДОК ХІМІЧНИХ АВАРІЙ / І.В. Ковалець, В.П. Беспалов,
С.Я. Майстренко, О.І. Удовенко
I.V. Kovalets, V.P. Bespalov, S.Ya. Maistrenko, O.I. Udovenko
ISSN 1681–6048 System Research & Information Technologies, 2022, № 3 70
Aнотація. Досліджено можливість комбінованого застосування скринінгових
моделей для оцінювання характеристик джерел у разі аварій на об’єктах збері-
гання небезпечних речовин зі складними моделями атмосферного перенесення
у складі сучасних систем підтримання рішень для розрахунку атмосферного
забруднення у широкому діапазоні просторових і часових масштабів. Час ви-
паровування у разі аварійного розливу, оцінений скринінговими моделями,
використовується для задання інтенсивності емісії і розрахунку атмосферного
поширення системою ядерного аварійного реагування RODOS. Для аварії в
Чернігові 23.03.2022 оцінено, що перевищення граничної допустимої концент-
рації аміаку 0.2 мг/м3 відбувається на відстанях до 75 км від джерела. Залеж-
ність розрахованих максимальних концентрацій від часу має асимптотичний
характер близький до 5.4
max
tc до 15 год після викиду, що узгоджується з
асимптотичним співвідношенням 2/3t для часової залежності розмірів
хмар у разі турбулентної дисперсії миттєвих викидів.
Ключові слова: атмосферна дисперсія, система RODOS, небезпечні речовини,
аміак, LASAT, DIPCOT, система «Повітря».
|
| id | journaliasakpiua-article-258100 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:27:49Z |
| publishDate | 2022 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/6c/e38c1c26df8837d5dd77b777f981986c.pdf |
| spelling | journaliasakpiua-article-2581002022-12-21T22:15:21Z The use of environmental decision support systems for modeling of atmospheric pollution following the chemical accidents ИСПОЛЬЗОВАНИЕ СИСТЕМ ПОДДЕРЖКИ РЕШЕНИЙ ОБЕСПЕЧЕНИЯ ЭКОЛОГИЧЕСКОЙ БЕЗОПАСНОСТИ ДЛЯ МОДЕЛИРОВАНИЯ ЗАГРЯЗНЕНИЯ АТМОСФЕРЫ ВСЛЕДСТВИЕ ХИМИЧЕСКИХ АВАРИЙ Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій Kovalets, Ivan Bespalov, Viacheslav Maistrenko, Svitlana Udovenko, Oleg atmospheric dispersion RODOS system hazardous substances ammonia LASAT DIPCOT “Povіtrya” system атмосферна дисперсія система RODOS небезпечні речовини аміак LASAT DIPCOT система “Повітря” We studied the possibility of the combined application of screening models to assess the characteristics of sources in accidents at storage facilities for hazardous substances with complex models of atmospheric transport as part of modern decision support systems to calculate air pollution in a wide range of spatial and temporal scales. The evaporation time following an emergency spill, estimated by screening models, is used to set the emission intensity and calculate the atmospheric transport by the RODOS nuclear emergency response system. For the accident in Chernihiv on March 23, 2022, it was estimated that the maximum permissible concentration of ammonia 0.2 mg/m3 was exceeded at distances up to 75 km from the source. The dependence of the calculated maximum concentrations on time is close to asymptote cmax ~ t-4.5 up to 15 h after emission, which is consistent with the asymptote σ ~ t3/2 for the time dependence of the sizes of puffs following turbulent dispersion of instantaneous releases. Исследована возможность комбинированного применения скрининговых моделей для оценки характеристик источников при авариях на объектах хранения опасных веществ со сложными моделями атмосферного переноса в составе современных систем поддержки решений для расчета атмосферного загрязнения в широком диапазоне пространственных и временных масштабов. Время испарения при аварийном разливе, оцененное скрининговыми моделями, используется для задания интенсивности эмиссии и расчета атмосферного распространения системой ядерного аварийного реагирования RODOS. Для аварии в Чернигове 23.03.2022 оценено, что превышение максимальной допустимой концентрации аммиака 0.2 мг/м3 происходит на расстояниях до 75 км от источника. Зависимость рассчитанных максимальных концентраций от времени имеет асимптотический характер, близкий к cmax~t-4.5 до 15 ч после выброса, что согласуется с асимптотическим соотношением σ~t3/2 для зависимости размеров облаков от времени при турбулентной дисперсии мгновенных выбросов. Досліджено можливість комбінованого застосування скринінгових моделей для оцінювання характеристик джерел у разі аварій на об’єктах зберігання небезпечних речовин зі складними моделями атмосферного перенесення у складі сучасних систем підтримання рішень для розрахунку атмосферного забруднення у широкому діапазоні просторових і часових масштабів. Час випаровування у разі аварійного розливу, оцінений скринінговими моделями, використовується для задання інтенсивності емісії і розрахунку атмосферного поширення системою ядерного аварійного реагування RODOS. Для аварії в Чернігові 23.03.2022 оцінено, що перевищення граничної допустимої концентрації аміаку 0.2 мг/м3 відбувається на відстанях до 75 км від джерела. Залежність розрахованих максимальних концентрацій від часу має асимптотичний характер близький до cmax ~ t-4.5 до 15 год після викиду, що узгоджується з асимптотичним співвідношенням σ ~ t3/2 для часової залежності розмірів хмар у разі турбулентної дисперсії миттєвих викидів. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2022-10-30 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/258100 10.20535/SRIT.2308-8893.2022.3.04 System research and information technologies; No. 3 (2022); 57-70 Системные исследования и информационные технологии; № 3 (2022); 57-70 Системні дослідження та інформаційні технології; № 3 (2022); 57-70 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/258100/264893 |
| spellingShingle | атмосферна дисперсія система RODOS небезпечні речовини аміак LASAT DIPCOT система “Повітря” Kovalets, Ivan Bespalov, Viacheslav Maistrenko, Svitlana Udovenko, Oleg Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title | Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title_alt | The use of environmental decision support systems for modeling of atmospheric pollution following the chemical accidents ИСПОЛЬЗОВАНИЕ СИСТЕМ ПОДДЕРЖКИ РЕШЕНИЙ ОБЕСПЕЧЕНИЯ ЭКОЛОГИЧЕСКОЙ БЕЗОПАСНОСТИ ДЛЯ МОДЕЛИРОВАНИЯ ЗАГРЯЗНЕНИЯ АТМОСФЕРЫ ВСЛЕДСТВИЕ ХИМИЧЕСКИХ АВАРИЙ |
| title_full | Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title_fullStr | Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title_full_unstemmed | Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title_short | Використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| title_sort | використання систем підтримання рішень з екологічної безпеки для моделювання забруднення атмосфери внаслідок хімічних аварій |
| topic | атмосферна дисперсія система RODOS небезпечні речовини аміак LASAT DIPCOT система “Повітря” |
| topic_facet | atmospheric dispersion RODOS system hazardous substances ammonia LASAT DIPCOT “Povіtrya” system атмосферна дисперсія система RODOS небезпечні речовини аміак LASAT DIPCOT система “Повітря” |
| url | https://journal.iasa.kpi.ua/article/view/258100 |
| work_keys_str_mv | AT kovaletsivan theuseofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT bespalovviacheslav theuseofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT maistrenkosvitlana theuseofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT udovenkooleg theuseofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT kovaletsivan ispolʹzovaniesistempodderžkirešenijobespečeniâékologičeskojbezopasnostidlâmodelirovaniâzagrâzneniâatmosferyvsledstviehimičeskihavarij AT bespalovviacheslav ispolʹzovaniesistempodderžkirešenijobespečeniâékologičeskojbezopasnostidlâmodelirovaniâzagrâzneniâatmosferyvsledstviehimičeskihavarij AT maistrenkosvitlana ispolʹzovaniesistempodderžkirešenijobespečeniâékologičeskojbezopasnostidlâmodelirovaniâzagrâzneniâatmosferyvsledstviehimičeskihavarij AT udovenkooleg ispolʹzovaniesistempodderžkirešenijobespečeniâékologičeskojbezopasnostidlâmodelirovaniâzagrâzneniâatmosferyvsledstviehimičeskihavarij AT kovaletsivan vikoristannâsistempídtrimannâríšenʹzekologíčnoíbezpekidlâmodelûvannâzabrudnennâatmosferivnaslídokhímíčnihavaríj AT bespalovviacheslav vikoristannâsistempídtrimannâríšenʹzekologíčnoíbezpekidlâmodelûvannâzabrudnennâatmosferivnaslídokhímíčnihavaríj AT maistrenkosvitlana vikoristannâsistempídtrimannâríšenʹzekologíčnoíbezpekidlâmodelûvannâzabrudnennâatmosferivnaslídokhímíčnihavaríj AT udovenkooleg vikoristannâsistempídtrimannâríšenʹzekologíčnoíbezpekidlâmodelûvannâzabrudnennâatmosferivnaslídokhímíčnihavaríj AT kovaletsivan useofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT bespalovviacheslav useofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT maistrenkosvitlana useofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents AT udovenkooleg useofenvironmentaldecisionsupportsystemsformodelingofatmosphericpollutionfollowingthechemicalaccidents |