Bioscience Biotechnology Research Communications

An Open Access International Journal

P-ISSN: 0974-6455 E-ISSN: 2321-4007

Bioscience Biotechnology Research Communications

An Open Access International Journal

Svetlana E. Germanova1*, Vadim G. Pliushchikov2, Polina A. Petrovskaya3, Nikolay V. Petukhov4 and  Tatiana A. Ryzhova5

1Senior Lecturer of the Department of Technosphere Security of the Agrarian and Technological Institute, Рeoples’ Friendship University of Russia (RUDN University) 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation

2Doctor of Agricultural Science, Professor, Director of the Department of Technosphere Security of the Agrarian and Technological Institute, Рeoples’ Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation

3Senior Lecturer of the Department of Landscape architecture and sustainable ecosystem of the Agrarian and Technological Institute, Рeoples’ Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation

4Candidate of Agriculture Science, Associate Professor of the Department of Technosphere Security of the Agrarian and Technological Institute, Рeoples’ Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation

5Candidate of Physical and Mathematical sciences, Senior Lecturer of the Institute of Physical Research and Technology of the Faculty of Physics and Mathematics and Natural Sciences, Рeoples’ Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation

Corresponding author email: yurina_iriha@mail.ru

DOI:

Article Publishing History

Received:

Accepted After Revision:

ABSTRACT:

Conducting effective targeted monitoring of land pollution by oil production and refining is a task that is relevant for all countries, especially for Russia. Much in the efficient solution of this problem depends on the types of pollutant and soil, protection technology. It is important to know the assessments of pollution risks, to localize them. The purpose of the work is system analysis and modeling of land pollution by oil under limited and uncertain data. The hypothesis of analysis and modeling under consideration: “an ecosystem is open, continuously developing and interacting with the environment”. Used methods of analysis-synthesis, decision-making, optimization and simulation of systems, an evaluation technique and procedure, as well as expert and heuristic approaches. Multifactorial and uncertainty of the solved problem, which often complicates research and prediction, are proposed. The main result of our work is the methodology of modeling (forecasting) of the state of the land taking into account bifurcations, making managerial (for example, recreational) decisions. The study was conducted with a focus on predicting self-healing of the environment. The results can be used in the development of intelligent systems for assessing land pollution.

KEYWORDS:

Assessment, Data, Modelling, Oil, Pollution.

Download this article as:

Copy the following to cite this article:

Germanova S. E, Pliushchikov V. G, Petrovskaya P. A, Petukhov N. V, Ryzhova T. A. Modelling and Assessment of Oil Pollution Under Data Constraints. Biosc.Biotech.Res.Comm. 2021;14(2).


Copy the following to cite this URL:

Germanova S. E, Pliushchikov V. G, Petrovskaya P. A, Petukhov N. V, Ryzhova T. A. Modelling and Assessment of Oil Pollution Under Data Constraints. Biosc.Biotech.Res.Comm. 2021;14(2). Available from: <a href=”https://bit.ly/3wIHobt“>https://bit.ly/3wIHobt</a>


INTRODUCTION

Soil is not a renewable resource; its degradation is almost irreversible or associated with duration comparable to human life. Pollution of the earth goes as a chain reaction; it reduces the diversity, stability and organics of the soil, its ability to self-repair. Pollutants from the soil enter on our table – through groundwater, plants, animals on pastures, birds and cause a “bouquet” of diseases, especially oncological ones. The yield of crops and their quality and the ability of the soil to decompose organic pollutants are reduced. Antibiotics, bacteria are getting into the soil, bacterial resistance of the fruiting part of the soil is falling. The polluting capacity of petroleum products and risk situations related to oil production or refining is greatly affected.

About 12.5% of all world oil production is produced in Russia (Pikovsky et al  2015). At the same time, the territories bordering oil production are at risk, sensitive to spills and other methods of pollution. Especially for effects that reduce the self-healing of the environment (ecological niche), which depends on the type and amount and the joint effects of pollutants that increase pollution (“summation effect”). The impact of pollution on land, human health is a multi-criterial and multifactorial problem, often with a lack of data, their uncertainty.

In Russia, estimates and models of risk situations with integral effects of pollutants on human health have only begun to be actively studied (Zaitsev et al., 2013).Non-compliance with environmental standards is associated with the risk of costs of oil producing and refining enterprises to compensate for losses in the quality and quantity of land used in the national economy. Now, remote soil sensing, satellite or UAV applications are also successfully used, followed by intelligent data analysis (Abramov et al, 2018,  Van Opheusden, et al 2020 ).

The environment is directly subjected to controlled and controlling influences in oil development (refining) systems and is considered by us as a self-organizing system for which various situational scenarios and evolution strategies are possible. Heuristic and expert methods are an effective method of systemic, complex analysis of environmental impacts. In the work, they are applied to the study of the problems of analyzing environmental pollution.

MATERIAL AND METHODS

Geo-environmental analysis relies, as usual, on monitoring, its results and their applicability to all objects of the environment, and not only, for example, to soil or air. In relation to soil, a complete set of factors is taken into account – erosion, salinization, swelling, waterlogging, etc., as well as self-organizational processes in the environment – migration of substances, restoration, accumulation, etc.The most dangerous factors for oil-contaminated soil are hydrochloric salts, hydrogen sulfide, sulfur, organic chlorides, small fractions, resinous and paraffin compounds, etc.Factors can be classified – according to the properties of the environment, relationships with the environment, the structure (composition) of pollutants and soil, water and air. Data sources, completeness and certainty also affect. For example, using the work data (Pikovsky, et al, 2015), the following table can be compiled.

Table 1. Range of factors influencing oil pollution in Russia

N Factor Max (%) Min (%)
1 Light pollutants 18.2 11.9
2 Paraffin 1.9 0.5
3 Pitches 27.0 19.5
4 Sulfur 2.05 1.55
5 Density (kg/m3) 907 896

The complexity of monitoring activities and data structures, the inability to adapt to their processing of procedures verified by formal means, leads to the need to attract non-traditional approaches. They help out heuristic and expert procedures, modeling methods that we actively use.As noted, their application is complicated by uncertainty and lack of data, without which the relevance of heuristic and expert procedures is impossible. But a tool for analytical processing of monitoring data has been developed to help researchers. Big Data in Computational Social Sciences and Humanities (2018) is processed by Data Mining (Zayar, Mykhaylov, Ye Thu, 2020), mathematical (Kaziev, Shevlokov, 2008) and situational modelling.

A certain pattern of soil pollution can be obtained by the method of bioindicators (bioremediation) or determinants of the degree of pollution of the environment with the help of living organisms – bacteria, algae, conifers, invertebrates, etc. They respond to many environmental factors and impacts, up to oil spills. It occurs according to the scheme “impact – response – information” (Buzmakov, Egorova, Gatina, 2017).But soil bioindication is a complex, thin instrumentation (Israel, Phillipova, Insarov et al, 1982). It uses the natural metabolic activation of microorganisms, plants and the activity of photosynthesis, catalytic reactions that lead to changes in the soil. The method is efficient and simple, without large attachments. But incorrectly selected or transferred to another area, it gives incorrect results. For example, “responds” in various ways depending on the type of soil, humidity, altitude, etc.

The impact of petroleum products on the soil is intensified by suboptimal (unsustainable) requirements, logistics, investments in risk forecasting and damage prevention, the development of infrastructure support for oil production and processing.The effect of the pollutant on the soil can also be non-specific, manifested in the reduction of its resistance, resistance to desertification and diseases due to petroleum products. The amplitude and the exposure intensity period, as well as the pollutant dose.

RESULTS AND DISCUSSION

With fixed pollution, for example, releases of the pollutant into the soil with subsequent physicochemical compounds, pollution fields and concentrations of the pollutant in the medium are formed. The usual problem is the identification of parameters, pollution data or their lack, uncertainty. Here we propose the following approach.Let a pollutant with a concentration of u(x,y,h,t) enter the soil to the depth h ∈ [O,H] and by the area determined by the coordinates  x,y by the moment t.

To smooth out the lack of information and “compress” it, its possible inaccuracies, as well as reduce the number of variables, average as follows:

It’s assumed that at the interval [t;oT ], the function


is little dependent on the initial moment. The pollution concentration field is a complex vector object. It can be associated with a scalar value – the average territorial (according to the ecological niche  with area determined by the measure (Ω)).

The value of  in the territory of Ω coincides with  We call this value, by analogy with (Pikovsky, Ismailov, Dorokhova, 2015), the background value of the concentration of the pollutant in the territory of Ω. Evasion of Ū can occur under the influence of cleaning measures or natural, anthropogenic influences, forming “peaks”. The very value of  can greatly depend on the territory, . The most complete picture nevertheless does not give the average value, but the mathematical expectation and distribution function u over x, y, h and t.When modeling the state of soil contaminated with petroleum products, system models and principles of forecasting the response of land, soil to pollution are necessary.

Balance models of semi-empirical type, expert and heuristic procedures (Bestuzhev-Lada, 1982) can be used to obtain information on the degree of response of soil, land to changes in pollutant concentration with the above measure. Let us give an example of this approach. If we could receive optimum, for example, in size of residual dispersion, a form of dependences of pollution on each considered factor of possible to define multiple-factor best dependence of a look.


where  – weight coefficients of significance of the criterion of the type “pollutant concentration – pollution value (pollution changes)” for the i-th pollutant. This is an indicator of participation in the total pollution of the i-th pollutant. In the deterministic version, the traditional least squares method is used, and in the random nature of impacts, the maximum likelihood method is used, for example (Van Opheusden, Acerbi, Ma, 2020). This approach will take into account self-organization, adaptive management regimes based on evolutionary potential for self-cleaning of soil with the help of bioactive substances. Here, the methods of Social Mining (Kaziev, Kazieva, Khizbullin, Takhumova, 2019) should be applied, flexible models that take into account connections throughout the contaminated territory. Its state will be taken into account in the following evolutionary way.

We define the state vector x, activity s(x), the functionality of the activity (or tension, fatigue) of processes in the system. They will provide additional or generalized information on the contaminated area.For example, if a pasture is restored or knocked out of circulation at a rate defined in following case

then the potential of this process and the entire system can be set by the functionality of the form:

where a is the coefficient of natural land regeneration, x is spatial, t is temporal variables. Above the pace – above the potential, or vice versa. The soil cover specified at the initial moment will be depleted under E < 1.The threshold concentrations above which changes in soil occur, as well as the adaptive response to changes in oil concentrations, should first be determined. If there is contamination of land in the oil production zone, then we consider the vector of concentrations of pollutants (factor vector) –


and the vector of concentrations (MPC) –


and the effectiveness of the effect of the pollutant (participation in pollution) soil:


Taking into account the “total” effect – the enhancement of the polluting capacity, its toxicity and the ability to act. To assess impacts, we propose a system of ordinary and fissile equations of the form:

where mij is the weight corresponding to the relative influence of pollutant i, enhanced (attenuated) by the presence of pollutant j in the soil, y10 – MPC of pollutant j in the absence of other pollutants.Weights can be set according to various methods. The simplest and most understandable:


then the solution of the given system in logarithmic form can be represented in the following form:

follows an increase in pollution at and  – its weakening. From the  values according to calculations (forecast) we select the largest. Similarly, we find minimum values. They are used in situational modeling (simulation calculations) to make relevant design decisions, for example, agricultural ones.In the above approach to determining a multifactor dependency, it’s important to know the forms of the basis dependencies To determine them in conditions of lack of data, we propose to use simulation simulations with a database of such functions, formed by experts.

For example, one can use the following types of functions:

and others.As a criterion for the adequacy of single-factor constituents, a residual dispersion is usually taken. But for the necessary testing of zero hypotheses, “sufficient process depth” and others, the probabilities should be identified, both of the pollution risks themselves, as well as the possibility of their localization and subsequent neutralization.Data are not always normally distributed, as, however, are their errors. Therefore, a preliminary statistical analysis should be performed. As a procedure for such an analysis, a working procedure can be used (Fetisov, Kolesnikov, 2018).We’ll suggest another algorithm.

Let  be a vector of the state of the medium (not necessarily pollution, as above), for example, x1 is the sulfur content,x2  is the acidity of the soil (PH), etc. The following procedure may then be used.

  1. Entering experimental data .
  2. We believe i: = 1.
  3. We find the optimal form of communication for , for example, using a bank of dependencies (logistic, fractional-rational, lognormal, etc.).
  4. If i < n, then ii+1 and go to item 3.
  5. We divide the medium into cellular structures (discretization of space) according to the principles of cell neighborhood.
  6. Leaving from the starting point (for example, the pollutant where the pollutant enters), we find pollution according to the pollutant transfer model.

Design of forecast models – after preliminary statistical analysis and application of Big Data and Data Mining, with the help of which we improve predictive analytics on structural, phase responses to the effects of the pollutant. Health risk monitoring (Deryabin, Unguryan, Buzinov, 2019) is associated with high costs and data redundancy. You can get rid of such costs using intelligent analytical systems, situational criteria for information and modeling (forecasting) with optimal termination of monitoring (“at the level of rationality and profitability”).The effect of petroleum products on the ecosystem is not additive by type of pollutants and their concentrations. If the impact is close to the state under normal, natural conditions, then intra-system changes, alterations are possible. If the effect is significantly different from normal, then a transition to a new state of equilibrium is possible. Such is the process of evolution – the search for a new state of equilibrium and “balancing” (adjustment) near this point does not always go smoothly.

Quantitative indicators of soil, the entire ecosystem, necessary data for prediction, modeling, their certainty, sensitivity, etc. also change.When recultivating land, involving them again in economic circulation, without taking into account soil and bioclimatic restrictions (on land categories) on drilling wells, the use of acid-water and washing solutions cannot be dispensed with. As without forecasting, simulating their effects, without taking into account the effects of pollutants on the anthropogenic risk load of the ecosystem, system analytics and the technique of “squeezing maximum information from minimal data.”

For example, the criterion for recoverability may be the ability to grow plants, their density and stability, pH in soil samples (up to 0.5).

Methods (procedures) for analyzing and evaluating data with restrictions on them, methods for classifying and ranking land should be developed. This will allow the transition to legal environmental assessment standards.In the course of the development of these studies, one can consider sources of random data of Markov observations, which may be the Markov observations, with a matrix of transition probabilities. This situation corresponds to the used statistical data processing procedure, for example, with a given quality function, if observations of the use of the environment are acting ahead of time.If a vector observations or experts polling is specified, then the observation corresponds to a Markov chain with an identifiable matrix of transition probabilities. The structure and complexity of data, monitoring determines the assessment of environmental pollution. Using evolutionary stochastic modeling, it’s possible to reduce the complexity of obtaining and processing data, and to increase the efficiency of ecological decision-making.

CONCLUSION

Extraction and processing of oil resources can lead to irreversible consequences, risks. Monitoring, assessment and prevention of soil contamination should take into account the ranges of permissible concentrations, deviations from the possible residual content of harmful compounds in the soil. In our opinion, the main risks for land, soil cover are the practical alienation of land (desertification or salinization), accumulation of poorly neutralized and toxic impurities. The risk of bifurcations in the environment, during the operation of fields and oil refining sites, is associated with them (so far although theoretical).

The results of the proposed study and the expert heuristic procedure will allow to identify, predict risks and damage from pollution in conditions of insufficient statistical monitoring data (by volume, accuracy, structure).The results of the study will help in the development of predictive intelligent systems, for example, expert ones, as well as moving from traditional MPC to differentiated sanitary and epidemiological standards, which is very relevant during the pandemic COVID-19. This situation is relevant and should be taken into account.

ACKNOWLEDGEMENTS

This paper has been supported by the RUDN University Russia Strategic Academic Leadership Program.

REFERENCES

Abramov, N.S., Makarov, D.A., Talalaev, A.A., Fralenko, V.P. (2018) “Modern methods of intelligent processing of DZZ data”. Software systems: theory and applications. 9: 4 (39), 417-442.

Bestuzhev-Lada, I. (1982) “Working book on forecasting”. Moscow: Thought. 426p.

Big Data in Computational Social Sciences and Humanities (2018). “Big Data in Computational Social Science and Humanities”. 1-25. DOI: 10.1007/978-3-319-95465-3_1.

Buzmakov, S.A., Egorova, D.O., Gatina, E.L. (2017) “Dose-effect of oil pollution of soils on the biotic component of ecosystems” Bulletin RUDN (ser. “Ecology and Safety of Life”), 25: 2. 217-229. DOI 10.22363/2313-2310-2017-25-2-217-229.

Deryabin, A.N., Unguryan, T.N., Buzinov, R.V. (2019) “The risk to health of the population connected with an exposition of chemicals of the soil” Risk analysis to health. 3. 18-25. DOI:10.21668/health.risk/201.

Fetisov, D.V., Kolesnikov, A.N. (2018). “Monitoring of subsoil use processes based on processing of aerospace images” News of the Tula State University (Technical Sciences series). 2, 216-226.

Germanova, S.E., Ryzhova, T.A., Kocheva, M.V., Fedorova, T.A., Petukhov, N.V. (2020) “Situational modelling of oil pollution risks monitored by distributed monitoring”   Amazonia Investiga,  9: 25. 44-48.

Israel, Yu.A., Phillipova, L.M., Insarov, G.E., Semevsky, F.N., Semenov, S.M. (1982) “Impact of background pollution of the natural environment on biota: problems of assessment and forecast” Problems of ecological monitoring and modeling of the ecosystem, Yu. A. Israel (Ed.). Vol. V. – Leningrad: Hydrometeoisdat, pp. 6-18.

Kaziev, V.M., Shevlokov, V.Z. (2008) “Modeling of land alienation in the agro-industrial complex” International Agricultural J. 5, 56-58.

Kaziev, V., Kazieva B., Khizbullin F., Takhumova O. (2019) “Increasing the Information Content of Social Network Groups and Clients Using Social Mining”. International Conference on Man-Power-Law-Governance: Interdisciplinary Approaches (MPLG-IA 2019). Advances in Social Science, Education and Humanities Research. 374. 180-184.

Pikovsky, Yu.I., Ismailov, N.M., Dorokhova, M.F. (2015) “Fundamentals of oil and gas geoecology”: a textbook / A.N. Gennadieva (Ed.). Moscow. 400 p.

Van Opheusden, B., Acerbi, L., Ma, W. (2020) “Unbiased and efficient log-likelihood estimation with inverse binomial sampling”. PLoS Comput. Biol. 16(12): https://doi.org/10.1371/journal.pcbi.1008483.

Zaitsev, N.V., Shur, P.Z., Kiryanov, M.R., Kamaltedinov, M.R., Tsinker, M.Yu. (2013) “Methodological approaches to assessing population health risk based on emotional models”. Public health and habitat. 1. 4-6.

Zayar, Aung, Mikhaylov, Ilya, Ye Thu, Aung (2020). “Data Mining Methods for Solving Classification Problem of Oil Wells”. BDET 2020: Proceedings of the 2020 2nd International Conference on Big Data Engineering and Technology. 40–44. https://doi.org/10.1145/3378904.3378911.