=Paper=
{{Paper
|id=Vol-3006/66_regular_paper
|storemode=property
|title=Modeling of the spatial distribution of the components for the ecosystem of the Novosibirsk reservoir
|pdfUrl=https://ceur-ws.org/Vol-3006/66_regular_paper.pdf
|volume=Vol-3006
|authors=Aleksandr A. Tskhai,Vladislav Yu. Ageikov,Aleksandr N. Semchukov
}}
==Modeling of the spatial distribution of the components for the ecosystem of the Novosibirsk reservoir==
https://ceur-ws.org/Vol-3006/66_regular_paper.pdf
Modeling of the spatial distribution of the
components for the ecosystem of the Novosibirsk
reservoir
Aleksandr A. Tskhai1,2 , Vladislav Yu. Ageikov1 and Aleksandr N. Semchukov1
1
Institute for Water and Environmental Problems SB RAS, Barnaul, Russia
2
Polzunov Altai State Technical University, Barnaul, Russia
Abstract
The object of the study is the ecosystem of the largest in Western Siberia — the Novosibirsk reservoir.
The aim of the study is forecast the response of hydrobiocenosis on the implementation of different
methods for the aquatic ecosystem restoration. Novelty: structural-dynamics modeling of ecological
processes based on the reproduction of biogeochemical cycles of limiting elements in the conditions of
spatial heterogeneity for the reservoir is performed. A preliminary conclusion is formulating about the
main role of autochthonous processes in the eutrophication of the Novosibirsk reservoir. A comparative
assessment of the influence for three variants of washing the reservoir with a flood wave on the annual
variability of the phytoplankton content and nitrate concentration in three characteristic parts of the
Novosibirsk reservoir was carried out.
Keywords
Novosibirsk reservoir, ecosystem, eutrophication, restoration, spatially inhomogeneous, prediction,
model.
1. Introduction
The Novosibirsk reservoir (NR) is the largest in Western Siberia. Trophic status of the NR
(in the end of 20th century) has attributed to the oligo-mesotrophic type [1]. The study of the
ecosystem state for the NR in the last thirty years indicates the development of eutrophication
in this water body: “. . . The largest recorded chlorophyll “a” concentrations (up to 45.5 mg/m3 )
corresponded to highly eutrophic water bodies. The occasional concentrations of chlorophyll
“a” up to 298–316 mg/m3 recorded during the so-called “water blooming” because of the local
and short-term mass development of cyanoprokaryotes, correspond to the hypereutrophic level
and serve as environmental risk factor for the fishery and recreational use of the reservoir, as
well as for drinking water supply” [2]. Eutrophication is the excessive development of primary
productivity processes, accompanied by the degradation of water communities, which, as a
result, can lead to the impossibility of use a water body by inhabitants.
Therefore, today the question of measures to restore the ecosystem of the NR is relevant.
World practice shows that the relevant measures are very expensive projects. For example, the
cost of restoring Great Lakes ecosystems that have been eutrophied is estimating by the Federal
SDM-2021: All-Russian conference, August 24–27, 2021, Novosibirsk, Russia
" tskhai@iwep.ru (A. A. Tskhai)
© 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org)
557
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
Environmental Protection Agency (US) to be more than a billion dollars [3]. Therefore, the use
of mathematical modeling to assess the environmental consequences of the rehabilitation of
the NR has become an important issue when considering various ways to improve it.
A model “Biogen” of the biogeochemical transformation of the limiting elements nitrogen
and phosphorus (for example, [4]) has used to assess the state of the reservoir ecosystem under
study. Recently, a spatially inhomogeneous modification of this model has developed [5]. Using
structural-dynamic modeling, the ecohydrological mechanism of unusual spatial distributions
of phytoplankton in the NR has reproduced and evaluated [6].
The results of the application of the developed model approach for predicting the conse-
quences of applying to the object under study known in the world practice [7, 8] measures for
the improvement of aquatic ecosystems are presenting here.
2. Materials and methods
A comprehensive study of the NR conducted in 1981/1982 yielded a body of data allowing for
a tentative model description of the processes in the reservoir ecosystem [1]. The scenario
of annual dynamics of component concentrations in the head section of the reservoir (the
Kamen’-na-Obi) was selected based on actual concentrations of mineral forms of N (NH4 , NO2 ,
NO3 ), P, O2 , chlorophyl “a”, zooplankton, phytoplankton, organic matter and detritus with the
use of empirical relationships [4].
The box approximation [5] is used for simulation of spatial heterogeneity of environmental
processes. To do that, ten areas of the reservoir divided into three layers by depth (30 boxes in
total) are used.
Seven areas are located along the main stream, two — in shallow waters near the left bank,
and one — in the Berd Bay (Figure 1). If depth exceeds 2 m, the upper and lower layers are
Figure 1: The Novosibirsk reservoir areas.
558
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
assuming to be of 1 m thick. The bottom layer is absent if depth is less than 1 m. Under ice
conditions, depth H is the distance from the ice low boundary to the bottom.
To simulate environmental processes based on solving three-dimensional hydrothermal and
box’s biogeochemical tasks, the variability of the spatial-temporal structure of water flows,
matters and temperatures for the conditions of 1981 was reproduced [5].
The hydrothermal block of the “Biogen” model has defined using the following initial in-
formation as the average daily parameters of each box. The list of data is as follows: water
temperature, volume, metric area, length, width and height of the box; the metric area of contact
with the bottom; the metric area of the ice-free surface; the thickness of the ice; the distance from
the box to the water surface or the lower surface of the ice; positive and negative components
of the water discharges into this box from others.
Transformation and dynamics of nine 𝐶𝑖 variables (Figure 2) are simulating to reproduce the
processes of biogeochemical transformation of nitrogen and phosphorus compounds as well as
oxygen regime in the surface and intermediate boxes. These variables relate to a water column,
where 𝑍𝑂 — zooplankton biomass, 𝐹 — phytoplankton biomass, 𝑁 𝐻4, 𝑁 𝑂2, 𝑁 𝑂3 — mineral
forms of N, 𝐷 — suspended substances, 𝐶 — dissolved organic substances, 𝐼 — mineral P and
𝑂2 — oxygen.
Extra six variables are adding to do calculations for bottom boxes. 𝐶𝐵 is organic matter (OM)
involved in metabolic processes. Interstitial forms of phosphorus and nitrogen are 𝑃 𝐵 and 𝑁 𝐵,
respectively. Forms of phosphorus and nitrogen sorbed at solid phase of bottom sediments are
designating as 𝑃 𝑆 and 𝑁 𝑆. Passive OM in the bottom sediment are marked as variable 𝐶𝑁 .
Figure 2: A scheme of biochemical components transformation in the aquatic ecosystem according to
the model “Biogen”.
559
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
Equations of the “Biogen” model for description of the 𝑖-th component transformation into
the 𝑗-th box are as follows:
𝑑(𝐶𝑖𝑗 · 𝑊𝑗 )
= 𝑊𝑗 · 𝑅𝑖𝑗 + 𝑄𝑗𝑞 𝐶𝑖𝑗 + 𝐽𝑖𝑗 · Ω𝑗 + 𝐺𝑗𝑖 𝐿𝑗 ,
∑︁ ∑︁
𝑄𝑘𝑗 · 𝐶𝑖𝑘 − (1)
𝑑𝑡 𝑞
𝑘
where 𝑖 = 𝑍𝑂, 𝐹 , 𝑁 𝐻4, 𝑁 𝑂2, 𝑁 𝑂3, 𝐷, 𝐶, 𝐼, 𝑂2; 𝐶𝑖𝑗 — the concentration of the 𝑖-th
component in the 𝑗-th box; 𝑊𝑗 — the volume of the 𝑗-th box; 𝑡 — the time; 𝑅𝑖𝑘 — the rate of
biochemical transformation of 𝐶𝑖 in the 𝑘-th box; 𝑄𝑘𝑗 — the water discharged from the 𝑘-th
box to the 𝑗-th box; 𝐽𝑖𝑗 — the mass flow of the 𝑖-th component through the interfacial surface
into the 𝑗-th box; Ω𝑗 — the metric area of the interfacial surface of the 𝑗-th box; 𝐺𝑗𝑖 — the lateral
load of the 𝑖-th component in the 𝑗-th box; 𝐿𝑗 — the length of the 𝑗-th box.
In the intermediate and bottom boxes, the lateral load 𝐺𝑗𝑖 is assumed to be zero. The runoff
from the reservoir banks makes up less than 6% of the river’s water flow [1]. Special calculations
for N and P compounds arrived from the reservoir bank (0–12% of the river’s biogenic load) do
not reveal a significant difference in annual variability of the simulated components.
Terms and values of internal parameters of the “Biogen” model are given in [4]. Unknown
parameters of the “Biogen” model [4] were found within the intervals of typical values by
minimizing Cr — the Theil’s statistical criterion [9] for components of the aquatic ecosystem:
⎯
⎸ 𝑛
⎸∑︁
⎷ (𝑋𝑖 − 𝑌𝑖 )2
𝑖=1
Cr = ⎯ ⎯ , (2)
⎸ 𝑛 𝑛
⎸∑︁ ⎸∑︁
𝑋𝑖2 + ⎷ 𝑌𝑖2
⎸
⎷
𝑖=1 𝑖=1
where 𝑛 is the number of observations, and 𝑋𝑖 and 𝑌𝑖 are the simulated and observed values of
substance concentrations, respectively. This model is used as a basis for monitoring the aquatic
ecosystems of the Ob river [10].
In the current model approaches (the review is given, for example, in [11]), the dynamics of
ecosystems are considered, as a rule, within the framework of the equilibrium approximation,
which makes it impossible to characterize new for the hydrobiocenosis, significantly non-
equilibrium cases, such as degradation or restoration for the state of the aquatic ecosystem.
Tools which allow to flexibly adjust the internal structure of the ecosystem model are needed.
To take into account intensive impacts on the ecosystem, it is necessary. Such changes lead the
hydrobiocenosis out of the circle of past equilibrium states.
In order to construct adaptive models, one should set a criterion for choosing from a set of
trajectories for ecosystem behavior under non-equilibrium conditions. The structural-dynamic
models (SDM) that appeared at the end of the 20th century in the search for a possible response
of an ecosystem to a violation of its stability use the assumption of non-equilibrium thermody-
namics. It is about the preference for a development trajectory that corresponds to the possible
maximum order of a living system in specific conditions [7].
The first steps have been taken on this path only if time changes in the simulated ecosystem
are taken into account, which, in particular, did not allowed to take into account the role of
560
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
spatial mechanisms that determine the intra-water mass and energy flows of substances that
occur in real ecosystems (for example, [8]).
In this regard, in this study, the development of a structural-dynamic approach aimed at taking
into account spatial factors was carried out. This allowed us to perform a model assessment
of the impact of the possible use of methods for restoring water ecosystems on the spatial
variability of the ecosystem components of the NR.
3. Results and discussion
By modeling the biogeochemical cycles of transformation of biogenic elements, the features
of the hydrochemical and hydrobiological regimes of the NR for the conditions of 1981 are
reproduced. The study uses the results of parameterization of the “Biogen” model in the zero-
dimensional approximation [4]. The simultaneous solution (in 30 boxes) of zero-dimensional
tasks with account of heat, water and mass exchange allowed us to estimate the spatially
inhomogeneous dynamics of the components of the water ecosystem of the NR.
To simulate the ecosystem restoration of the NR, the results of the application of the most
well-known method of managing the aquatic ecosystem by reducing the biogenic load from
the catchment area are presented in a scenario manner. This method consists in carrying out a
complex of water protection measures, which would reduce the pollution of the studied water
body.
Theoretically, when implementing effective measures in the catchment area, it is possible to
reduce the content of nitrogen and phosphorus compounds in the river to 80% or even 60% of
the actual observed concentrations. How would this affect the state of the ecosystem of the NR?
To answer this question, special simulation was carried out with the variation of the biogenic
load on the NR, which allowed us to obtain the spatial and temporal distribution of pollution
in the water areas. The difference between the content of pollutants in the case of a decrease
in the biogenic load is reduced compared to the reality, but only by a few percent. This result
suggests that autochthonous processes play a major role in eutrophication in the NR. As noted
above, the costs of limiting the biogenic load are very high.
Another way to restore water quality: the maximum flushing of a reservoir by a flood wave
is used in the world (for example, [8]). To simulate the response of the ecosystem of the NR
under consideration, in addition to the real one, the first and second scenarios for managing the
operation mode of the hydroelectric power station were formed (Figure 3).
The meaning of the first scenario option was to pass the maximum volume of water through
the dam during the flood, but so that during further operation it would be possible to navigate
river vessels through the NR toward middle part of the Ob river. In the second scenario, and
this restriction is removed, the maximum water level during the filling phase is such that during
the subsequent low-water periods there is enough water for the water supply of settlements
only. The modeling result of the ecosystem responses was shown in Figures 4–9.
The implementation of such scenario options may have a number of other water management
and environmental consequences, which should be taken into account when planning the
actual operation of the reservoir. In this study, only one aspect was considered: how will a
scenario-defined flood wave wash affect the intra-annual dynamics of phytoplankton content
561
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
Figure 3: Inflow and outflows, levels of the Novosibirsk reservoir from November 1980 until January
1982 for three variants: real; intermediate and maximal washout.
Figure 4: Annual variability of phytoplankton content in the 6th area for three regimes of the Novosibirsk
reservoir exploitations in 1981.
Figure 5: Annual variability of nitrate concentrations in the 6th area for three regimes of the Novosibirsk
reservoir exploitations in 1981.
562
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
Figure 6: Annual variability of phytoplankton content in the 9th area for three regimes of the Novosibirsk
reservoir exploitations in 1981.
Figure 7: Annual variability of nitrate concentration in the 9th area for three regimes of the Novosibirsk
reservoir exploitations in 1981.
Figure 8: Annual variability of phytoplankton content in the 10th area for three regimes of the Novosi-
birsk reservoir exploitations in 1981.
563
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
Figure 9: Annual variability of nitrate concentration in the 10th area for three regimes of the Novosibirsk
reservoir exploitations in 1981.
and other controlled water quality indicators. In addition, the entire array of information that is
obtained as a result of simulation, but due to the limited volume is not given in this publication,
should be analyzed.
The results of the corresponding calculations of the content of phytoplankton and nitrates in
the case of the three above-described modes of operation of the hydroelectric power station for
the surface layers of three water areas of the main part of the NR and the Berd Bay are presented.
The first of them was considered the 6th water area, the most susceptible to eutrophication
(Figures 4 and 5). Secondly, the 9th water area, which provides water for the Novosibirsk water
intake, is here (Figures 6 and 7). Finally, the Berd Bay is of particular importance as a source of
water for the Berd water intake (Figures 8 and 9).
An increase in the intensity of reservoir flushing due to a flood wave would lead to a noticeable
decrease in the maximum level of phytoplankton content in the surface layers of the water
areas of the main part of the NR (by one and a half to two times).
The implementation of the second scenario does not differ much from the real version of the
temporal picture of phytoplankton in the Byrd Bay. However, the maximum washing (the third
scenario) would lead to a temporary shift of the peak of the algae content by almost a month
in the 10th water area. This can be understood by taking into account the peculiarities of the
water exchange of the Berd Bay with the near-dam part of the NR in the summer low water.
The maximum permissible concentration of nitrates in the Russian Federation for water bodies
of fisheries significance is 9.1 gN/m3 . The considered regimes would not cause a significant
change in the content of nitrate pollution in the water areas of the main part in the HR (Figures 5
and 7). At the same time, in the 10th water area from the end of May, the nitrate content in the
case of the third scenario would significantly (by 2–3 times) decrease compared to the first and
second, and, starting from mid of August — until the end of the year, on the contrary, it would
slightly increase. The explanation is in the corresponding alternation of periods for activity of
phytoplankton.
The results of the corresponding calculations of the content of phytoplankton and nitrates in
the case of the three above-described hydroelectric power station operating modes for the surface
layers of two water areas. First of them, the 6th area most susceptible to eutrophication was
considered (Figure 4). Secondly, the 9th area is which water outflow enters for the Novosibirsk
564
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
water intake from (Figure 5). It can be seen that an increase in the flushing of the reservoir due
to the flood wave would lead to a noticeable decrease in the maximum level of phytoplankton
content in the surface layers of the water areas of the NR to one and a half to two times. The
maximum permissible concentration of nitrates in the Russian Federation for water objects of
fishery significance is 9.1 g/m3 . Considered modes would not have caused a essentially increase
in the content of nitrate contamination (Figures 6 and 7).
4. Conclusion
Model estimates of the impact of the use of different methods of restoring the aquatic ecosystem
on the spatial distribution of ecosystem components are obtained, including for the case of three
variants of the HB operation mode during a flood. The result of calculations for the surface
layers of the main part of the Novosibirsk reservoir (6th , 9th ) and the Berd Bay (10th ) water
areas of phytoplankton content and types of pollution is given on the example of nitrates for
the case of three modes of annual operation of the NR. The results of preliminary scenario
calculations show that the use of reservoir flushing during a flood would significantly affect the
level of phytoplankton content in the surface layers of the selected water areas for comparison.
A preliminary conclusion is formulated about the main role of autochthonous processes in the
eutrophication of the Novosibirsk reservoir.
Acknowledgments
The study carried out within the Research Programs of the Institute for Water and Environmental
Problems SB RAS and the Polzunov Altai State Technical University; was funded by the Russian
Foundation for Basic Research and the government of Altai Krai, the Russian Federation (grant
No. 18-41-220002).
References
[1] Podlipskii Yu.I., Chaikovskaya T.S. Comprehensive studies of the Novosibirsk reservoir.
Moscow: Gidrometeoizdat, 1985.
[2] Vasiliev O.F. Long-term dynamics of water-ecological regime of the Novosibirsk reservoir.
Novosibirsk: SB RAS Publ. House, 2014.
[3] The great lakes restoration initiative: Background and issues. Available at: https://www.
everycrsreport.com/reports/R43249.html#_Toc370914293.
[4] Tskhai A.A., Ageikov V.Yu. Disturbance of sustainability of the reservoir ecosystem: A
model approach for assessing and forecasting the long-term process of eutrophication //
Journal of Sustainable Development of Energy, Water and Environment Systems. 2021.
Vol. 9. Is. 1. 1080327. DOI:10.13044/j.sdewes.d8.0327.
[5] Tskhai A.A., Ageikov V.Yu., Semchukov A.N. Modelling of transformation features for
nitrogen and phosphorous compounds in the conditions of the Novosibirsk reservoir //
Water Sector of Russia. 2020. Vol. 6. P. 87–102. DOI:10.35567/1999-4508-2020-6-5.
565
Aleksandr A. Tskhai et al. CEUR Workshop Proceedings 557–566
[6] Tskhai A., Ageikov V., Semchukov A. Hydrological paradoxes of phytoplankton distribution
in the Novosibirsk reservoir // EGU General Assembly 2021. Online, 19–30 Apr 2021,
EGU21-287. DOI:10.5194/egusphere-egu21-287.
[7] Jørgensen S.E. Lake management. Oxford: Pergamon Press, 1980.
[8] Kong X.-Z., Jørgensen S.E., He W., Qin N., Xu F.-L. Predicting the restoration effects by a
structural dynamic approach in Lake Chaohu, China // Ecological Modelling. 2013. Vol. 266.
P. 73–85.
[9] Theil H. Applied economic forecasting. North-Holland: Amsterdam, 1971.
[10] Tskhai A., Puzanov A. Kovalevskaya N., Kirillov V. New approach to aquatic ecosystems
monitoring for the Ob-river // Proceedings of the International Association of Hydrological
Sciences. 2020. Vol. 383. P. 375–379. DOI:10.5194/piahs-383-375-2020.
[11] Vinçon-Leite B., Casenave C. Modelling eutrophication in lake ecosystems: A
review // Science of Total Environment. 2019. Vol. 651. Pt 2. P. 2985–3001.
DOI:10.1016/j.scitotenv.2018.09.320.
566