Methodology For Environmental Monitoring With Use of Methods of Mathematical Modeling Nadiia Bielikova 1, Daniil Shmatkov 2 1 Research Centre for Industrial Problems of Development of the National Academy of Sciences of Ukraine, Іnzhenerny lane, 1, A, Kharkiv, 61166, Ukraine, 2 Scientific and Research Institute of Providing Legal Framework for the Innovative Development of the National Academy of Legal Sciences of Ukraine, Chernyshevskaya st., 80, Kharkiv, 61002, Ukraine, Abstract The article is aimed at developing methodological support for environmental monitoring using modeling methods, which will optimize the number of studied indicators and facilitate the processing, interpretation and visualization of the results. Developed methodology envisages the following stages: formation of the initial set of partial indicators; factor analysis of partial indicators and reduction of those with a factor load of less than 60%; structuring a set of partial indicators (selection of components and calculation of integrated indicators); application of matrix analysis for grouping of monitoring objects. Keywords 1 Environmental monitoring, sustainable development, modeling, factor analysis, integral indicator, partial indicators. 1. Introduction certain areas of sustainable development, of large amounts of information (variety baselines), of the complexity of their processing and interpretation Sustainable development involves the of results, and of the formulation of conclusions harmonization of development and functioning of require knowledge-intensive approaches. environmental, economic, and social areas. The This article is aimed at developing interaction between these areas is embodied in the methodological support for environmental formation of the outline of direct and indirect monitoring using modeling methods, which relation between the supervision and dynamics of optimize the number of studied indicators and their development and the results achieved. The facilitate the processing, interpretation and problem of sustainable development is complex, visualization of the results. as part of its solution it is advisable to monitor the economy, social area, and the state of the environment using methods that would take into 2. Methodology account the complexity and difficult predictability of this process. Ensuring environmental monitoring within the ICT play a significant role in the developed methodology envisages the following transformation of sustainable development stages: formation of the initial set of partial approaches. Issues of organization and indicators; factor analysis of partial indicators and digitalization of monitoring the functioning of reduction of those with a factor load of less than III International Scientific And Practical Conference “Information Security And Information Technologies”, September 13–19, 2021, Odesa, Ukraine EMAIL: nadezdabelikova@gmail.com (A. 1); d.shmatkov@gmail.com (A. 2). ORCID: 0000-0002-5082-2905 (A. 1); 0000-0003-2952-4070 (A. 2). ©️ 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 (CEUR-WS.org) 60%; structuring a set of partial indicators entropy of the environment features and integral (selection of components and calculation of indicators for estimation of its components. integrated indicators); application of matrix This approach allows us to take into account analysis for grouping of monitoring objects. that the greater the entropy of any partial indicator As the initial set of monitoring indicators is that characterizes a certain feature of any large, in conditions of insufficient time it is component of the environment, the more advisable to use methods of data reduction [1]. disordered the ecological system would be as a One of such methods is the factor analysis which whole. If the entropy of the trait, expressed as a is widely applied in ecological science in various partial exponent, is insignificant, then its weight directions [2–5]. in the total set of traits is also insignificant [6]: 𝑛 Within the framework of the proposed approach, the first stage determines the number of 𝑅(𝑆𝑖 ) = ∑ 𝐻𝑗 𝑏𝑖𝑗 , 𝑖 = 1, 𝑚, (1) factors that should be identified within the 𝑗=1 reduction of monitoring indicators. The initial set where 𝑅(𝑆𝑖 ) – integral value of the object; of partial indicators takes part in the analysis. 𝐻𝑗 – the entropy of j-th feature; The essence of the analysis is that during the 𝑏𝑖𝑗 – quantitative assessment of j-th feature of i sequential selection of factors, they include less object; and less variability of monitoring indicators. m – number of objects; Therefore, the decision on when to stop the n – number of features. procedure for the selection of factors depends Matrix analysis is used in this work to largely on the analysis purposes, but one of the visualize the results of monitoring which recommendations to streamline the process of facilitates their interpretation and the possibility selecting the number of factors is to consider the of obtaining homogeneous groups of objects of scree plot. the study by positioning them in different Next, it is proposed to structure the initial set quadrants of matrices. Having a group of objects of indicators for monitoring, which remained after in the study we can isolate their common the factor analysis. Structuring is done by characteristics. To do this, we offer three matrices identifying the components that will be followed or positioning planes: by a generalized assessment of the environment. • Atmosphere – Water resources (IA – IW). All selected components correspond to the main • Soil – Wastes (IS – IWs). components of the living environment for • Forest resources – Nature reserves and monitoring of which the initial set was formed and hunting grounds (IF – INR). which includes the following indicators: Accordingly, the axes of these matrices are IA – integral indicator of the atmosphere integrated indicators for assessing the six selected assessment; components of the environment. IW – integral indicator of the water resources To determine the boundaries of the quadrants assessment; of the matrices, the range of values of the integrated indicators for environment components IS – integral indicator of the soil assessment; estimation can be divided into three parts by the IWs – integral indicator of the wastes golden ratio. The golden ratio is such a assessment; proportional division of a segment into unequal parts in which the whole segment belongs to the IF – integral indicator of the forest resources larger part as much as the largest part belongs to assessment; the smaller one; or in other words, the smaller INR – integral indicator of the nature reserves segment refers to the larger as the larger segment and hunting grounds assessment. refers to all [7]: To calculate the integrated components of the 𝑌𝐵 𝐴𝐵 = = 𝛼, above indicators, it is proposed to use the entropy 𝐴𝐵 𝑌𝐴 (2) method [6], the stages of which (adapted to the where YB, AB, YA – parts of a segment or objectives of the environmental monitoring) are numerical series. the formation of a set of partial indicators and Depending on the defined conditions and their assessment, standardization of partial features of research objects development, as well indicators taking into account their impact on the as properties and role which they carry out in a environment, and calculation of the value of system, nine functions of distribution of the investigated sample are allocated: chaos, Symbol Partial indicators development of elements, development of n.6 Emissions of sulfur dioxide into the properties, development of relations, balance of air from stationary sources of functions of development and preservation, pollution, thousand tons preservation of relations, preservation of n.7 Emissions of nitrogen dioxide into properties, preservation of elements, and collapse the atmosphere from stationary [7,9]. Environmental monitoring objects can be sources of considered as systems with connections and elements. Since the development of this system is pollution, thousand tons unbalanced, in certain periods of time it even n.8 Emissions of carbon monoxide into contains signs of chaos, the most suitable function the air from stationary sources of to describe these processes can be defined as pollution, thousand tons “development of elements” with the appropriate n.9 Emissions of non-methane volatile percentage distribution of parts of the range of organic compounds into the values: [0,0; 0,328) – low, (0,329; 0,735) – atmosphere from stationary sources medium level, (0,736; 1,0) – high level]. of pollution, thousand tons n.10 Emissions of ammonia into the 3. Results atmosphere from stationary sources of pollution, thousand tons At the first stage we formed an initial set of n.11 Emissions of methane into the air partial indicators for assessing the ecological from stationary sources of pollution, sphere of sustainable development of the country thousand tons – its environment. n.12 Drawing of water from natural water The environment has the following main objects, million m3 components that affect health and quality of life: n.13 Drawing of water from natural water air, water, soil, wastes, forests, nature reserves and objects per person, m3 hunting, etc. These components are reflected both n.14 Water loss during transportation, in international indices that assess various aspects million m3 of habitat quality and in statistics to assess the n.15 Use of fresh water, including fresh development of regional environments. Based on and sea water, million m3 this, it is proposed to monitor the environment n.16 Use of fresh water, including fresh using the following partial indicators (Table 1). and sea water, used for the needs of the national economy and Table 1 population, million m3 Partial indicators for environmental monitoring n.17 Water saving drawing through the Symbol Partial indicators circulating and recycling water n.1 Emissions of carbon dioxide into the supply, million m3 air from stationary sources of n.18 General drainage, million m3 pollution, thousand tons n.19 Discharge of return waters into n.2 Emissions of pollutants into the air surface water objects, million m3 from stationary sources of pollution, n.20 Discharge of contaminated return thousand tons water into surface water objects, n.3 Emissions of pollutants into the air million m3 from stationary sources of pollution n.21 Discharge of contaminated return per square kilometer, tons water without purification into the n.4 Emissions of pollutants into the air surface water objects, million m3 from stationary sources of pollution n.22 Discharge of insufficiently treated per person kilometer, kg contaminated return water into n.5 Emissions of suspended solids into surface water objects, million m3 the atmosphere from stationary n.23 Discharge of normatively clean sources of pollution, thousand tons without treatment return water into surface water objects, million m3 Symbol Partial indicators Symbol Partial indicators n.24 Wastewater treatment facilities, n.50 Area of transfer of forest areas into million m3 land covered with forest vegetation, n.25 Application of mineral fertilizers per hectares hectare of acreage, kg n.51 Number of illegal felling, units n.26 Application of organic fertilizers per n.52 Damage caused to forestry, millions hectare of acreage, tons of Ukrainian hryvnia n.27 The area of crops fertilized with n.53 The area of hunting lands provided mineral fertilizers, thousand for use, thousand hectares hectares n.54 Land area of nature reserves, n.28 The area of crops fertilized with biosphere reserves and national organic fertilizers, thousand nature parks, hectares hectares n.55 Number of wild animals (ungulates) n.29 Areas where pesticides were used, by objects on the territory of which thousand hectares the lands are located, thousand n.30 Waste generation, thousand tons heads n.31 Waste generation of I–III classes of n.56 Number of wild animals (fur animals) danger, thousand tons by objects on the territory of which n.32 Waste generation per square lands are located, thousand heads kilometer, tons n.57 Number of wild animals (game birds) n.33 Waste generation per capita, kg by objects on the territory of which n.34 Waste disposal, thousand tons the lands are located, thousand n.35 Utilization of wastes of I–III classes of heads danger, thousand tons n.36 Waste incineration, thousand tons The composition of environmental monitoring n.37 Waste disposal in dedicated places objects may vary and depends on the objectives. and facilities, thousand tons In particular, it can be conducted at the global and n.38 Removal of waste of I-III classes of national levels: for countries, regions, cities or danger in specially designated places other territories and settlements. and facilities, thousand tons In this article, the objects of monitoring are n.39 Waste disposal in fly-tipping, defined as regions (administrative-territorial thousand tons units) of Ukraine which have different n.40 Total amount of waste accumulated characteristics of the environment due to different during operation in waste disposal levels of industrial development, climate, sites, thousand tons geographical location, state of natural resources, n.41 Total amount of waste accumulated and other factors. during operation in waste disposal Data collection of partial indicators for sites per square kilometer, thousand environmental monitoring in statistical sources tons allowed us to establish that the objects of n.42 Total amount of waste accumulated monitoring have significant differences in the during operation in waste disposal values of partial indicators n.1 – n.57. For sites per person, thousand tons example, the discrepancy between the maximum n.43 Area of forest destruction, hectares (233,7 thousand tons in the Donetsk region) and n.44 Number of forest fires, units the minimum (0,2 thousand tons in the n.45 The area of forest lands covered by Transcarpathian region) values of sulfur dioxide fires, hectare emissions into the air was 1168,5 times (Table 2). n.46 Area of burned and damaged forest, m3 Table 2 n.47 Area of reforestation, hectares Values of partial indicators for environmental n.48 Area of afforestation, hectares monitoring, 2017 (fragment) [8] n.49 Area of transfer of forest areas of Values natural regeneration into land The monitoring object n.5 n.6 n.7 covered with forest vegetation, hectares Vinnytsia region 17,0 71,9 10,6 Values Table 3 The monitoring object Eigenvalues of factors obtained by the principal n.5 n.6 n.7 components’ method Volyn region 1,4 0,4 0,5 Cumulative % Total – Cumulative Dnipropetrovsk region 86,5 66,8 31,2 Factor Eigenvalue – variance –% Donetsk region 76,2 233,7 44,8 Eigenvalue Zhytomyr region 2,7 1,0 1,6 1 21,2 37,28 21,2 37,3 Transcarpathian region 0,4 0,2 0,7 2 7,2 12,6 28,4 49,8 Zaporizhya region 13,1 79 31,9 3 5,9 10,3 34,3 60,1 Ivano-Frankivsk region 37,3 129,6 14,5 4 4,6 8,1 38,9 68,3 Kyiv region 12,4 14,3 4,8 5 3,0 5,3 41,9 73,5 Kirovograd region 4,0 0,9 1,4 Luhansk region 10,4 33,3 8,1 6 2,5 4,4 44,4 77,9 Lviv region 8,4 39,8 6,8 Mykolayiv region 3,6 0,7 2,6 As can be seen from table 3, the first and Odessa region 3,6 1,9 2,4 second factors explain 49,83%, i.e. half of the variance of the initial indicators of environmental Poltava region 6,3 7,4 10 monitoring, and all six factors – 77,92% of the Rivne region 2,6 0,6 2,8 total variance. Therefore, in the process of factor Sumy region 3,5 3,1 3,2 analysis, it is possible to identify either two main Ternopil region 1,5 0,3 1 factors (factor 1 and factor 2) and to reduce those Kharkiv region 6,5 11,3 7,8 indicators of environmental monitoring that are Kherson region 1,2 0,7 0,3 not included in their composition or to identify six factors and to reduce those indicators of Khmelnytsky region 2,8 2,5 5,3 environmental monitoring that are not included in Cherkasy region 8,8 5,0 10 their composition. Chernivtsi region 0,9 0,4 0,3 Leaving for analysis factors 1 and 2 and Chernihiv region 3,9 6,4 3,6 reducing the indicators of environmental Consideration of the scree plot (Fig. 2) allows monitoring which have a factor load of more than a researcher to determine the place where the 60% and explain 49,83% of the total variance, the decline in the eigenvalues of the factors from left following results were obtained: to right is slowed down as much as possible. In • Composition of factor 1 “Dangerous”: this graph, this place corresponds to the number n.1, n.2, n.3, n.4, n.5, n.6, n.7, n.8, n.11, n.18, of factors equal to six. But the maximum n.19, n.20, n.21, n.22, n.24, n.33, n.34, n.35, variability of the initial indicators is explained by n.37, n.40, n.41, n.42. the first and second factors (Table 3). • Composition of factor 1 “Permissible”: n.12, n.13, n.15, n.16, n.25, n.44, n.45, n.46, n.47, n.49, n.51, n.55, n.56. Thus, factor 1 includes monitoring indicators that characterize the negative phenomena of the environment: emissions of hazardous substances into the atmosphere, different types of waste generation, etc. Given the composition of the indicators that fall into factor 1, it can be called “Dangerous” because it has a negative impact on the environment. Factor 2 includes monitoring indicators that characterize less dangerous phenomena: water intake and its use for various purposes, application of mineral fertilizers to soil, etc. Given the Figure 1: Graph of eigenvalues of environmental composition of the indicators of factor 2, it can be monitoring factors (scree plot) conditionally called "Permissible" because it has a permissible and, in some cases, positive impact Figure 2: Positioning of monitoring objects in the on the environment. matrix “Soil–Wastes” According to the criterion of factor load less than 60%, the following indicators were reduced: Characteristics of matrix quadrants are the n.9., N.10, n.14, n.17, n.23, n.26, n.27, n.28, n.29, following: n.31, n.32, n.36, n.38, n.39, n.43, n.48, n.50, n.52, HH – high assessment of soil – high level of n.53, n.54, n.57. wastes management; The six selected factors include indicators that HM – high assessment of soil – medium level characterize areas of environmental monitoring of wastes management; such as air and water pollution by various types of HL – high assessment of soil – low level of hazardous substances (factor 1 and factor 3); wastes management; wastes management (factor 3); use of natural MH – medium assessment of soil – high level resources for different purposes (factor 2); of wastes management; restoration of forest resources (factor 4); soil MM – medium assessment of soil – medium management (factor 5); loss of forest stands level of wastes management; (factor 6). ML – medium assessment of soil – low level Factors 1–3 were the largest in terms of the of wastes management; number of included indicators, and only one LH – low assessment of soils – high level of indicator was included in factor 6, which wastes management; corresponds to the general rule of factor analysis LM – low assessment of soil – medium level – reduction of the number of indicators included of wastes management; in each subsequent selected factor due to reduced LL – low assessment of soil – low level of variability of indicators. wastes management. The analysis allowed us to conclude that the Coordinates of positioning points in the matrix minimum number of factors that can be identified are the following: Ukraine (0,371; 0,735); is two factors, and the maximum is six factors. Vinnytsia region (0,134; 0,773); Volyn region And in addition, the logic of this study allowed us (0,623; 0,775); Dnipropetrovsk region (0,144; to recommend the second option of factor analysis 0,234); Donetsk region (0,384; 0,710); Zhytomyr according to which there are six factors region (0,462; 0,777); Transcarpathian region influencing the environment which explain the (0,707; 0,776); Zaporizhzhya region (0,109; maximum indicators variability. 0,761); Ivano-Frankivsk region (0,671; 0,769); Thus, as a result of the reduction, the initial set Kyiv region (0,574; 0,772); Kirovograd region of environmental monitoring indicators was (0,072; 0,639); Luhansk region (0,392; 0,769); reduced from 57 to 45. Lviv region (0,571; 0,764); Mykolayiv region After calculating the entropy of integrated (0,241; 0,762); Odessa region (0,072; 0,775); indicators that characterize the state of the Poltava region (0,321; 0,686); Rivne region selected components of the environment (Fig. 1), (0,710; 0,775); Sumy region (0,215; 0,738); we carried out the positioning of monitoring Ternopil (0,469; 0,776); Kharkiv (0,070; 0,740); objects in three matrices, for example, the matrix Kherson region (0,352; 0,761); Khmelnytsky “Soil–Wastes” is presented in Fig. 2. region (0,354; 0,774); Cherkasy region (0,362; 0,776); Chernivtsi region (0,658; 0,777); Wastes (Iws) 1,0 Chernihiv region (0,230; 0,775). LH MH HH The quadrants that are on the line of [names of five [names of [name of one development of monitoring objects from the worst regions] twelve regions] region] to the best condition are Dnipropetrovsk, 0,735 LM MM HM Donetsk, and Rivne regions. [names of four [name of one regions] region] 4. Conclusions 0,328 LL ML HL The given methodology for environmental [name of one monitoring with use of methods of mathematical region] modeling allows a researcher to draw conclusions 0,328 0,735 1,0 Soil (Is) with regard to a habitat condition as a whole in the country, to define the most dangerous state of ecology according to its administrative units, and of multidimensional objects of arbitrary to analyze results of an environment condition nature, Moscow, YKAR, 2004. assessment according to its components. [8] N. N. Moiseev, E. P. Ivanilov, E. M. The proposed methodology support provides Stolyarova, Optimization methods, Moscow, the implementation of the complex approach to The science, 1978. the establishment of monitoring of an ecological [9] O. M. Prokopenko (Ed.) Statistical data component of sustainable development and to Environment of Ukraine for 2017, Kyiv, strengthen its scientific substantiation. Its State Statistics Service of Ukraine, 2018. advantage is the ease and high implementation opportunities through ICT tools to reduce the time of monitoring and systematic analysis for clear conclusions and recommendations for more effective implementation of the concept of sustainable development. 5. References [1] D. Shmatkov, N. Bielikova, N. Antonenko, O. Shelkovyj, Developing an environmental monitoring program based on the principles of didactic reduction, European Journal of Geography, volume 10, number 1, 2019, 99– 116. [2] A. Deraemaeker, K. Worden, A comparison of linear approaches to filter out environmental effects in structural health monitoring, Mechanical systems and signal processing, volume 105, 2018, 1–15. https://doi.org/10.1016/j.ymssp.2017.11.045 [3] M. S. Guerreiro, I. M. Abreu, Á. Monteiro, T. Jesus, A. Fonseca, Considerations on the monitoring of water quality in urban streams: a case study in Portugal, Environmental monitoring and assessment, volume 192, 2020, 1–11. https://doi.org/10.1007/s10661- 020-8245-y [4] A. H. Hirzel, J. Hausser, D. Chessel, N. Perrin, Ecological‐niche factor analysis: how to compute habitat‐suitability maps without absence data?, Ecology, volume 83, number 7, 2002, 2027–2036. [5] O. Ovaskainen, G. Tikhonov, A. Norberg, F. Guillaume Blanchet, L. Duan, D. Dunson, T. Roslin, N. Abrego, How to make more out of community data? A conceptual framework and its implementation as models and software, Ecology letters, volume 20, issue 5, 2017, 561–576. https://doi.org/10.1111/ele.12757 [6] Infocommunications. Science and Technology. IEEE PIC S&T 2015. pp. 269 – 271. [7] V. Y. Vasylev, V. V. Krasylnykov, S. Y. Plaksyi, T. N. Tiahunova, Statistical analysis