Numerical Modelling of the Pollution Distribution from the

Underwater Discharge of the Wastewater Treatment Plant near the

Heraclean Peninsula of the Crimea under Different Wind Directions

Evgeniya Korshenko

, Nikolay Diansky

1,2,3 b

, Vladimir Fomin

and Irina Panasenkova

Department of Numerical Modelling of Hydrophysical Processes, Zubov State Oceanographic Institute, Moscow, Russia

Faculty of Physics, M.V. Lomonosov Moscow State University, Moscow, Russia

Institute of Numerical Mathematics of the Russian Academy of Sciences, Moscow, Russia

Keywords: Black Sea, Numerical Modelling, Pollution Distribution, Upwelling.

Abstract: The present paper is focused on the numerical modelling of the pollution distribution in the coastal waters of

the Crimean Peninsula from the underwater discharge of the biggest wastewater treatment plant in this area.

Calculations are performed using three-dimensional σ-model of marine circulation INMOM (Institute of

Numerical Mathematics Ocean Model) with high spatial resolution near the Crimean Peninsula for a one-year

period, using WRF (Weather Research and Forecasting model) atmospheric forcing. Simulations are carried

out according to the two experimental scenarios that differ in the presence of the leakage place in the pipeline

and its absence. Simulation results are used to study temporal and spatial structure of the pollution distribution,

taking into account hydrometeorological conditions and their seasonal and synoptic variation.

1 INTRODUCTION

In the case of the Crimean Peninsula, one of the

largest sources of marine coastal waters’ pollution is

connected to the discharge from numerous

wastewater treatment plants (WWTPs). The majority

of these plants was constructed in the 1970’s. Due to

physical and technological wear, they now only

produce mechanical treatment or generally untreated

sewage. The current condition of these plants can

significantly reduce water quality. Numerical

modelling of pollution distribution from the discharge

of the WWTPs makes it possible to estimate the

dynamics of this distribution and its impact on coastal

waters.

This paper presents simulation results of the

pollution distribution from the underwater discharge

of the largest WWTP on the Crimean Peninsula,

which are obtained with INMOM (Institute of

Numerical Mathematics Ocean Model) three-

dimensional σ-model of marine circulation.

https://orcid.org/0000-0003-2310-9730

https://orcid.org/0000-0002-6785-1956

https://orcid.org/0000-0001-8857-1518

https://orcid.org/0000-0002-8338-4825

2 MODEL AND DATA

Three-dimensional σ-model of marine circulation

INMOM has been developed at the INM RAS

(Institute of Numerical Mathematics of the Russian

Academy of Sciences) (Diansky, 2013) and based on

the primitive equations of ocean

hydrothermodynamics with the Boussinesq and

hydrostatic approximations. A regional version of

INMOM for the Black Sea (BS) with nonuniform grid

near the Crimean Peninsula was developed for this

study. To achieve higher horizontal spatial resolution

over a specific region of interest, a spherical

coordinate system with the location of one of the

poles at a point with geographic coordinates 33.96⁰E

and 44.74⁰N was used. As a result, horizontal spatial

resolution varies from 200 m near the new pole to 6-

8 km on the periphery of the BS (Figure 1). The grid

domain in the horizontal plane contained 889×488

points along the model longitude and latitude,

respectively. The vertical coordinate was represented

by twenty nonuniformly distributed σ-levels to provi-

Korshenko, E., Diansky, N., Fomin, V. and Panasenkova, I.

Numerical Modelling of the Pollution Distribution from the Underwater Discharge of the Wastewater Treatment Plant near the Heraclean Peninsula of the Crimea under Different Wind

Directions.

DOI: 10.5220/0007835803730377

In Proceedings of the 5th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2019), pages 373-377

ISBN: 978-989-758-371-1

373

de higher resolution near the surface.

Model depths were specified by using the

GEBCO data on the Earth's topography with a high

spatial resolution of 30’’ available to download from

www.gebco.net. However, the initial high-resolution

data were smoothed several times and limited by a

minimum depth of 4 m. This is a necessary procedure

for σ-models because the transformation of the

vertical coordinate is used and the function describing

undisturbed sea depth should be nonzero and

sufficiently smooth (Diansky, 2013).

Figure 1: Grid domain of the σ-model of marine circulation

INMOM with nonuniform grid near the Crimean Peninsula.

The initial conditions of temperature and salinity

were constructed using three-dimensional monthly

mean climatic fields of the Azov and the BS. The data

from the Atlas were taken as climatic fields of

temperature and salinity of the Azov Sea. In the case

of the BS, the data were provided by the Marine

Hydrophysical Institute of the Russian Academy of

Sciences (INM RAS) with a horizontal resolution of

0.1°x0.0625° and with 36 vertical levels, from 0 to

2150 m (Ivanov et al., 2011). All described data were

interpolated to the model grid domain.

The calculation of atmospheric forcing for a

numerical modelling period of 2011 year was

performed using the bulk formulas (Gill, 1986) based

on characteristics that were obtained by non-

hydrostatic atmosphere model WRF (Weather

Research and Forecasting model) with a spatial

resolution of 10 km and with a time resolution of 1

hour.

The discharge of the most significant rivers

flowing into the BS was specified according to the

climatological runoff data (Jaoshvili, 2002).

3 RESULTS AND DISCUSSION

The regional version of INMOM with high spatial re-

solution near the Crimean Peninsula described above

was used to calculate the distribution of pollution

from the underwater discharge of the WWTP KOS-1

“Yuzhnye” near the Heraclean Peninsula according to

two experimental scenarios. In the first scenario (I),

the entire sewage discharge occurred through the pipe

head at a depth of 80 m. The length of the pipeline

itself was considered to be 3.300 m. In the second

scenario (II), we assumed the presence of a leakage in

the pipeline at a distance of 500 m from the entrance

of the pipeline into the sea and at a depth of 33-35 m,

through which half of all sewage was discharged

(Figure 2).

Figure 2: The location of the pipeline is indicated by a solid

line. The location of the pipe head and leakage are marked

with stars. The entrance of the pipeline into the sea and the

location of its bend are indicated by circles. The vertical

cross-section to the coast is marked by a dotted line.

The pollution distribution in the case of both

experiments was calculated for a period of one year –

2011 – and the annual discharge volume of sewage of

38.300 thousand cubic meters was taken into account.

To perform the calculations, the volume

concentration of the conditional pollutant in the

sewage was assumed to be completely polluted and

equal to 1. It was taken into account that sewage

affected the distribution of salinity. Therefore, at each

model time step, the flow of pollutants into the

nearest model cells was calculated according to the

volumetric flows of pipes received for this model

step, multiplied by the corresponding concentration,

and instantaneous dilution of pollutants in the

volumes of the corresponding cells. A similar method

of calculating the pollution distribution was

previously used for the coastal waters of Big Sochi

and described in Diansky et al., 2013.

ONM-CozD 2019 - Special Session on Observations and Numerical Modeling of the Coastal Ocean Zone Dynamics

374

Graphs representing change in the volume

concentration of conditional pollutant on the sea

surface were drawn at points corresponding to the

pipe head location and the location of the leakage

place for both experiments and the entire 2011-year

calculation period.

Figure 3: Volume concentration of the conditional pollutant

on the sea surface for 2011 year at the points corresponding

to the location of the pipe head (a, c) and leakage place (b,

d) for two experiments: in the case of the entire sewage

discharge through the pipe head (I) (a, b) and the sewage

discharge through the pipe head and through the leakage

place in equal proportions (II) (c, d). The concentration

scale of conditional pollutant is presented in dimensionless

units and marked along the y axis.

Based on the results presented in Figure 3 (a-d),

the most adverse period of the pollution spread from

deeper layers to the surface occurred during the cold

autumn-winter and subsequent spring seasons, when

the BS vertical water exchange increased and when

vertical stratification was weakened due to

convective cooling and turbulent mixing (the last one

also occurred due to the increased intensity of winds

during these seasons). Due to the upper layers

warming, as well as the wind speed decrease after the

beginning of the summer season, a density jump layer

was formed, which usually prevented the release of

pollution from deeper layers to the sea surface.

However, in the case of the upwelling, which is

found to be a rather common phenomenon in the

coastal area of the Crimean Peninsula in the summer

season, the polluted water from deeper layers can

reach the surface. For the entire period of performed

calculations, several similar situations were

identified. We conducted an analysis of one of them,

dated 18.08.2011, and present it below (Figure 4).

Figure 4 (a, b) shows the pollution concentration

on the sea surface on 18.08.2011 under two scenarios

(I, II). These maps were reconstructed in a geographic

coordinate system by interpolating numerical

modelling results on the uniform 1 km grid area. The

scale of volume concentrations is presented in

dimensionless units, as in Figure 3. The results of

numerical modelling confirmed that, depending on

the experimental conditions, the most significant

differences in the pollution concentration were

observed at the sea surface. According to the graphs

(Figure 3), the highest concentration of the pollution

was noted in the experiment with the sewage

discharge through the pipe head and through the

leakage place in equal proportions. This was quite

predictable because of the closer location of the

leakage to the surface (~33-35 m) than the pipe head

itself (~80 m).

Figure 4 (c, d) shows pollution concentration

maps on the vertical cross-section to the shore (shown

in Figure 2) on 18.08.2011, also for both scenarios (I,

II). The map data was based on the calculation results,

transformed into a geographic coordinate system by

interpolating model data on the uniform 1-km grid

area, followed by interpolation on 36 vertical levels

from 0 to 2150 m and on the vertical cross-section

itself. Therefore, vertical coordinates are measured in

meters. The scale of volume concentrations, as in the

case of the pollution concentration on the surface, is

presented in dimensionless units.

Numerical Modelling of the Pollution Distribution from the Underwater Discharge of the Wastewater Treatment Plant near the Heraclean

Peninsula of the Crimea under Different Wind Directions

375

Figure 4: Volume concentration of the conditional pollutant

on the sea surface on 18.08.2011 in the case of the entire

sewage discharge through the pipe head (I, a) and the

sewage discharge through the pipe head and through the

leakage place in equal proportions (II, b). The concentration

scale of conditional pollutant is presented in dimensionless

units and shown on the right. The vectors of wind velocity

near the sea surface with the scale of arrows in cm/s are also

indicated on the right of the figures and are additionally

marked on the maps. The volume concentration on the

vertical section, the position of which is shown in Figure 2

on 18.08.2011 in the case of two different scenarios: (I, c);

(II, d). The concentration scales in dimensionless units are

shown below. (II, e) - temperature distribution on the

vertical cross-section to the shore on 18.08.2011. The scale

in °C is shown below. Vertical coordinates are measured in

meters.

Using the presented pictures and considering the

temperature distribution on the vertical cross-section

in Figure 4 (e), it can be concluded that on 18.08.2011

there was an upwelling caused by the north-east wind

when the coast was to the left of the wind direction.

According to existing studies, this situation is found

to be quite common in the coastal area of the BS

(Silvestrova et al., 2017). At the same time, the

integral Ekman transfer in the upper quasi-

homogeneous layer was directed from the coast,

which lead to a compensatory rise of the underlying,

colder waters, which sometimes come to the sea

surface (Silvestrova et al., 2017).

Figure 5 shows wind velocity and its direction in

August 2011 at the closest point to the pipe head and

the leakage place location, according to the WRF data

on the wind speed at 10 m height. The time scale

along the x axis is represented by the number of the

day in a year. Respectively, the date of 18.08.2011 is

equivalent to 230 days. According to the analyzed

graph, several days before the date of interest, the

wind strengthened and the north-west wind along the

coastline prevailed.

ONM-CozD 2019 - Special Session on Observations and Numerical Modeling of the Coastal Ocean Zone Dynamics

376

Figure 5: The wind velocity (m/s) and direction for August

2011 at the point closest to the pipe head and to the leakage

place according to the WRF data. The velocity of the wind

is represented by a shaded gray graph; wind direction

corresponds to red vectors, measured from the north

azimuth. The scale of the vectors in m/s is placed at the

bottom of the picture. The timeline is represented by the

numbers of days in a year.

It additionally confirmed the presence of the

upwelling phenomenon and it can be concluded that

because of the integral Ekman transfer arising due to

the intensification of northwestern and western

winds, polluted water raised from the deeper layers

and reached the sea surface at the sites of sewage

discharges, especially in the case of the leakage,

which can seriously affect the state of the marine

environment of the considered coastal area.

4 CONCLUSIONS

As part of the presented study, we developed a

regional version of the INMOM for the BS with the

nonuniform grid, horizontal spatial resolution of

which is increasing from 200 m near the Crimean

Peninsula to 6-8 km on the periphery of the BS. It

allowed us to calculate the pollution distribution from

the underwater discharge of the WWTP with the high

spatial resolution and gave us an opportunity to study

the temporal and spatial structure of the pollution

distribution, taking into account hydrometeorological

conditions and their seasonal and synoptic variability.

It was shown that the most adverse period of the

pollution transport from deeper layers to the sea

surface was occurring during the cold autumn-winter

and subsequent spring seasons, because of the

increasing vertical water exchange and weakening of

vertical stratification. However, even after the

beginning of the summer season, when a density jump

layer is usually formed in the BS, due to the

upwelling, the polluted water from deeper layers can

reach the surface.

The analysis of the numerical modelling results

from the experiments with two different scenarios

showed that in the presence of a leakage place in the

pipeline the release of pollution to the sea surface

occurred more intensively than in the case of its

absence. Therefore, the state of the pipeline also

requires further research to take all necessary

measures to reduce the anthropogenic load on the

ecological state of the coastal area. It is also noted that

the pollution distribution beyond the sea surface

horizon should be researched further.

ACKNOWLEDGEMENTS

The presented study was funded by Russian

Foundation for Basic Research according to the

research projects № 18-35-00512, № 17-05-41101.

REFERENCES

Diansky, N.A., 2013. Modelling of the ocean circulation

and study of its response to short-term and long-term

atmospheric forcing. 1

ed. Physmatlit, Moscow. (in

Russian)

Diansky, N.A., Fomin, V.V, Zhokhova, N.V., Korshenko,

A.N., 2013. Simulations of Currents and Pollution

Transport in the Coastal Waters of Big Sochi.

Atmospheric and Oceanic Physics, 49(6): 611–621.

Gill, A., 1986. Atmosphere and ocean dynamics. Vol. 1.

Moscow, Mir, 396 p. (in Russian).

Ivanov, V.A., Belokopytov, V.N, 2011. Oceanography of

the Black Sea. NAS of Ukraine, Marine Hydrophysical

Institute, Sevastopol. (In Russian)

Jaoshvili, S, 2002. The rivers of the Black Sea. European

Environmental Agency, eds. I. Chomeriki, G.

Gigineishvili, and A. Kordzadze, Technical Report No.

71.

Silvestrova, K.P., Zatsepin, A.G., Myslenkov, S.A., 2017.

Coastal upwelling in the Gelendzhik area of the Black

Sea: Effect of wind and dynamics. Oceanology, 57(4):

469-477.

Numerical Modelling of the Pollution Distribution from the Underwater Discharge of the Wastewater Treatment Plant near the Heraclean

Peninsula of the Crimea under Different Wind Directions

377