CN117634990A - Method for evaluating stability of freshwater ecosystem - Google Patents

Abstract

The invention discloses a method for evaluating the stability of a freshwater ecosystem, which comprises the following steps: forming an aquatic organism data and water environment factor data index original data set of a research area; acquiring a diversity index of each aquatic organism through the aquatic organism data and performing multiple collinearity diagnosis on the water environment factor data to form an index pretreatment data set; carrying out homogenization treatment on the index pretreatment data set; constructing a freshwater ecological system stability evaluation model based on a maximum flow principle; simulating an assessment model of the stability of the freshwater ecological system by using the self-organizing map network; the stability of the freshwater ecological system is evaluated through analysis of entropy values, and driving factors of the stability of the freshwater ecological system are identified through analysis of index weights. The invention has the beneficial effects that: the method can rapidly, accurately and sensitively reflect the stability condition of the freshwater ecosystem, identify the dominant factors driving the freshwater ecosystem to evolve, and is beneficial to the protection of the freshwater ecosystem.

Description

A method for assessing the stability of freshwater ecosystems

Technical field

The present invention relates to the technical field of freshwater ecological health assessment, and in particular to a method for assessing the stability of freshwater ecosystems.

Background technique

Freshwater ecosystem as a complex, open, dynamic, non-equilibrium and non-linear system. For studies on the stability of freshwater ecosystems, some only focus on the biological part of the ecosystem (such as taxonomy, biodiversity, etc.), while some consider both biological and abiotic parts, but only through statistical means (such as Correlation analysis and CCA analysis, etc.) simply connect the two. In complex and changeable environments, living things and non-living things are coupled together through nonlinear interactions, and due to the openness of ecosystems, they themselves often have complex structures. Therefore, previous methods cannot objectively and accurately assess the stability of freshwater ecosystems.

In order to truly describe the dynamic evolution process of freshwater ecosystems in time and space, and scientifically evaluate and predict the stability of river ecosystems, it is necessary to construct a freshwater ecosystem stability assessment model based on the maximum flow principle. In addition to considering aquatic organisms In addition to the importance of the water environment factor indicators themselves, the interaction between each indicator will also be incorporated into the model, which greatly reduces the subjectivity of previous assessment methods and overcomes the incomparability of freshwater ecosystem stability in time and space. characteristics, making the assessment results of freshwater ecosystem stability more accurate.

Contents of the invention

The purpose of the present invention is to provide a method for evaluating the stability of freshwater ecosystems, taking the interaction between aquatic organisms and water environment factor indicators in freshwater ecosystems as the entry point, and based on the principle of maximum flow to improve the stability of freshwater ecosystems. Accuracy and objectivity of stability assessment results.

In order to achieve the above objectives, the technical solution adopted by the present invention is: a method for assessing the stability of freshwater ecosystems, which includes the following steps:

Step S1, for the selected research area, collect aquatic organism data and water environment factor data through field monitoring and indoor experiments, and finally form an original data set of aquatic organism data and water environment factor data indicators in the study area;

Step S2: Calculate the Shannon-Wienner diversity index of each aquatic organism through the aquatic organism data of the indicator original data set in step S1, and perform multicollinearity diagnosis on the water environment factor data of the indicator original data set; the aquatic organism after processing is completed Biological data and water environment factor data form an indicator preprocessing data set;

Step S3: homogenize the indicator preprocessing data set in step S2;

Step S4: Construct a freshwater ecosystem stability assessment model based on the maximum flow principle;

Step S5, use the self-organizing mapping network to simulate the freshwater ecosystem stability evaluation model in step S4 using the indicator preprocessing data set completed by the homogenization process in step S3;

Step S6: Input the entropy value and index weight obtained through the simulation in Step S5 into the Excel table, evaluate the stability of the freshwater ecosystem through the analysis of the entropy value, and identify the driving factors of the stability of the freshwater ecosystem through the analysis of the index weight.

Further, in step S3, the indicator preprocessing data set in step S2 is normalized; specifically:

The stability of freshwater ecosystems increases with the increase of aquatic organisms and water environment factor index values, and pre-normalization processing is performed as shown in formula (1);

(1);

Among them, o represents the oth indicator, b _o represents the normalized value of the oth indicator, x _o represents the measured value of the oth indicator, x _omin represents the minimum value of the oth indicator, and x _omax represents the oth indicator. the maximum value;

The stability of freshwater ecosystems decreases as the index values of aquatic organisms and water environment factors increase. Pre-normalization processing is performed as shown in formula (2);

(2).

Further, the aquatic organism data includes: phytoplankton, zooplankton, benthic animals and fish; the water environment factor data includes total nitrogen, total phosphorus, chemical oxygen demand, turbidity and pH.

Further, the Shannon-Wienner diversity index of aquatic organisms in step S2 is as shown in formula (3):

(3);

in, represents the Shannon-Wienner diversity index, a represents the species, S represents the total number of individuals of all species in the community, is the percentage of species a in the total number of individuals.

Further, multicollinearity diagnosis in step S2 is carried out using the variance inflation factor method, and the variables with the largest variance inflation factor method are deleted in sequence, that is, the collinear water environment factors are deleted until all water environment factors of the variance inflation factor method are less than 5.

Further, in step S4, a freshwater ecosystem stability assessment model is constructed based on the maximum flow principle, where the maximum flow principle means that an open complex system far from equilibrium always looks for an optimization process so that the open complex system can operate under given constraints or The generalized flow obtained at the cost reaches the maximum value; the specific process is as follows:

Step S41, aquatic organisms and water environment factor indicators are the driving force for the evolution of freshwater ecosystems. The driving force for the evolution of freshwater ecosystems indicates the ability of aquatic organisms and water environment factor indicators to accept generalized information flow. The vector form is x=(x _{1 . _ _ _ _} For the calculation of quantity function, see formula (4);

(4);

Among them, J(x) represents the generalized entropy obtained by the volume unit at time t, eta and is the coupling coefficient, i, j, k, and l are the four indicators of aquatic organisms and water environment factors respectively,/> and/> is an exponential term, defined as a potential function, by and/> Reflect the characteristics of the evolution of the comprehensive performance of freshwater ecosystems; x _i , x _j , x _k and x _l are the normalized values of aquatic life and water environment factor indicators i, j, k and l respectively;

The average generalized entropy of all microstates at time t is shown in formula (5);

(5);

in, Represents the general characteristics of information entropy of freshwater ecosystems. ρ(x,t) is the time-varying probability density function of a random sequence set of aquatic organisms and environmental factor indicators in freshwater ecosystems at time t, which satisfies ;

Step S42, the generalized entropy J(x) obtained by the volume unit at time t of the freshwater ecosystem has no boundary and requires constraints to define the freshwater ecosystem; the constraint conditions are converted into a first-order momentum equation to a fourth-order momentum equation, see formula (6):

(6);

Among them, f ₁ is the first-order momentum equation, f ₂ is the second-order momentum equation, f ₃ is the third-order momentum equation, f ₄ is the fourth-order momentum equation, <> represents the statistical average; among them, four must exist in the freshwater ecosystem The interaction between the above aquatic organisms and water environment factor indicators;

Step S43: Optimize the freshwater ecosystem to obtain the maximum generalized entropy under given constraints. Under the given constraints described in formula (6), use Lagrange multipliers to maximize formula (5). Specifically, Expressed as formula (7):

(7);

Among them, ρ is the probability density, exp() represents the exponential function with the natural constant e as the base, Obtained through Lagrangian optimization;

The potential function ensures the stability of the freshwater ecosystem model. The potential function is expressed as:

(8);

Among them, the potential function Represents the basic characteristics of freshwater ecosystem evolution, x represents a random sequence of aquatic organisms and environmental factor indicators in freshwater ecosystems,/> and/> Obtained through Lagrangian optimization,/> is a parameter; use the generalized potential function/> Conduct a theoretical analysis of the formation and evolution of river ecosystems, whose properties are determined by parameters/> To decide, parameters/> Directly determined by the parameters of microscopic dynamic rules that regulate the interaction of information bodies;

Perform translation transformation on formula (8) and diagonalize the transformed constant term matrix to obtain formula (9):

(9);

in, is the stability entropy value of freshwater ecosystem,/> It is formed by the interactive connection of various indicators and is a key parameter to evaluate the stable state of freshwater ecosystems. a _i represents the connection weight corresponding to the indicator x _i , and x _i is the normalized value of the aquatic life and water environment factor indicator i;

According to the relationship between the potential function equation and the dynamic evolution equation, the stochastic evolution equation of the orderly pattern of the freshwater ecosystem is derived, expressed as formula (10):

(10);

in, express The first derivative with respect to time, λ is the eigenvalue of the matrix composed of a _i , is a nonlinear interaction term that reflects the interaction between indicators, and F(t) is a random term.

Furthermore, the self-organizing mapping network in step S5 is a type of unsupervised learning neural network model, which has the ability to handle complex nonlinear problems and simulate or learn the sample space or the unknown external environment.

Further, in step S5, the indicator preprocessing data set completed by the homogenization process in step S3 is simulated using a self-organizing mapping network, specifically: the data completed by the normalization processing is put into the self-organizing mapping network of MATLAB 2018 for simulation and training. The step size is set to 300, and the indicator weight and the entropy value indicating the stable level of the freshwater ecosystem are obtained through simulation.

The beneficial effects of the present invention are: (1) Relying on aquatic organisms and water environment factor data to conduct freshwater ecosystem stability assessments, taking into account the importance of the aquatic organisms and water environment factor indicators themselves, and also involving the interaction between the indicators. relationship; a freshwater ecosystem stability assessment model is constructed based on the maximum flow principle, and a self-organizing mapping network is used for simulation. It can not only intuitively compare the evolution trends of freshwater ecosystem stability in different times and spaces, but also judge the driving force of freshwater through indicator weights. key factors in ecosystem evolution.

(2) This method combines aquatic organisms and water environment factors, analyzes the freshwater ecosystem as a dynamic whole, weakens a single cause-and-effect relationship, and the results obtained may be closer to the essence. Based on the principle of maximum flow, the stability of the freshwater ecosystem is evaluated from the perspective of information. The method provided by the present invention has strong operability and can quickly, sensitively, accurately, comprehensively and objectively reflect the stability status of the freshwater ecosystem, and explore Identify key driving factors and provide important theoretical basis for protection and restoration measures of freshwater ecosystems.

Description of drawings

Figure 1 is a flow chart for freshwater ecosystem stability assessment;

Figure 2 shows the configuration of the aquatic ecosystem in different sections of the main stream of the Yellow River;

Figure 3 is a weight diagram of each indicator in the aquatic ecosystem of different reaches of the main stream of the Yellow River;

Figure 4 shows the results of aquatic ecosystem stability assessment in different reaches of the Yellow River mainstream.

Detailed ways

In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention will be further elaborated below in conjunction with specific implementation modes. It should be understood that the specific embodiments described here are only used to explain the present invention and are not intended to limit the present invention.

As shown in Figure 1, a method for assessing the stability of freshwater ecosystems includes the following steps:

Step S1, for the selected research area, collect aquatic organism data and water environment factor data through field monitoring and indoor experiments, and finally form an original data set of aquatic organism data and water environment factor data indicators in the study area;

Step S2: Calculate the Shannon-Wienner diversity index of each aquatic organism through the aquatic organism data of the indicator original data set in step S1, and perform multicollinearity diagnosis on the water environment factor data of the indicator original data set; the aquatic organism after processing is completed Biological data and water environment factor data form an indicator preprocessing data set;

Step S3: homogenize the indicator preprocessing data set in step S2;

Step S4: Construct a freshwater ecosystem stability assessment model based on the maximum flow principle;

Step S5, use the self-organizing mapping network to simulate the freshwater ecosystem stability evaluation model in step S4 using the indicator preprocessing data set completed by the homogenization process in step S3;

Step S6: Input the entropy value and index weight obtained through the simulation in Step S5 into the Excel table, evaluate the stability of the freshwater ecosystem through the analysis of the entropy value, and identify the driving factors of the stability of the freshwater ecosystem through the analysis of the index weight.

Example:

This invention uses aquatic organisms and water environment factors as basic data, builds a stability evaluation model based on the maximum flow principle to evaluate the stability of freshwater ecosystems, and uses water ecosystems in different sections of the main stream of the Yellow River as research objects to conduct evaluations.

A total of 44 sampling sections were arranged along the source area of the main stream of the Yellow River to the estuary, including 26 natural river sections and 6 typical reservoirs (Longyangxia, Liujiaxia, Qingtongxia, Wanjiazhai, Sanmenxia and Xiaolangdi Reservoirs). Each reservoir includes Three sampling sections are respectively arranged at the beginning, middle and end of the reservoir. At these 44 sampling sections, water environmental factors in the main stream of the Yellow River were measured and aquatic biological samples were collected.

There are 17 types of water environment factors, including water temperature (WT), conductivity (Cond), dissolved oxygen (DO), pH, oxidation-reduction potential (ORP), total dissolved solids (TDS), flow rate (V), and turbidity (Tur), chemical oxygen demand (COD), total phosphorus (TP), total dissolved phosphorus (TDP), orthophosphate phosphorus, total nitrogen (TN), ammonia nitrogen, nitrate nitrogen, nitrite nitrogen and total solubility Nitrogen (TDN).

There are 3 types of aquatic life samples, including phytoplankton (FYZW), zooplankton (FYDW) and benthos (DQDW).

Multicollinearity diagnosis is used for water environment factors, and the largest variables are deleted in order until all variables are less than 5. There are a total of 12 remaining water environment factors, including water temperature (WT), conductivity (Cond), and dissolved oxygen (DO). , pH, redox potential (ORP), flow rate (V), turbidity (Tur), chemical oxygen demand (COD), total phosphorus (TP), total dissolved phosphorus (TDP), nitrite nitrogen and total dissolved Nitrogen (TDN).

Calculate the Shannon-Wienner diversity index of three aquatic organisms.

The data of three kinds of aquatic organisms Shannon-Wienner diversity index and 12 kinds of water environment factors were normalized. The results are shown in Table 1:

Table 1 Normalization results of aquatic organisms and water environment factor indicators in the main stream of the Yellow River

The normalized data of each indicator are made into radar charts according to different river sections (see Figure 2). In each radar chart, the line box represents the characteristics of the river ecosystem network, which is caused by aquatic organisms and water environment factors. indicators are generated. They are flow patterns of generalized information flux, which more intuitively and vividly express the spatial expansion process of freshwater ecosystem structure.

The normalized data of each indicator is then imported into the self-organizing mapping network of MATLAB 2018 for simulation according to different river sections. Finally, the indicator weight a _ij (see Figure 3) and the entropy value indicating the stability level of the freshwater ecosystem are obtained through simulation. (See Figure 4).

Among them, the indicator weight reflects the competition and cooperation between indicators in the freshwater ecosystem. They reflect the impact of indicators on entropy values. The greater the weight of the indicator, the greater the impact of the indicator on the entropy value.

It can be seen from Figure 3 that the indicators with weights greater than 0.6 in the aquatic ecosystem of the four reaches of the Yellow River include Cond, V, Tur and TDP. The average values of these four indicator weights are 0.787, 0.706, 0.775 and 0.732 respectively.

It can be seen that Cond has the greatest impact on the stability of the Yellow River water ecosystem, followed by Tur, TDP and V.

It can be seen from Figure 4 that the water ecosystem of the main stream of the Yellow River self-organizes from 0 to 130 steps, resulting in an oscillation process under the influence of the environment. When the number of steps exceeds 130, as the number of steps increases, the dominant patterns of the water ecosystem in different reaches of the Yellow River mainstream remain unchanged, indicating that the simulation results are accurate and credible.

Entropy value of aquatic ecosystem stability from the source area to the lower reaches of the Yellow River They are 6.864, 5.855, 5.282 and 3.939 respectively.

It can be seen that the water ecosystem stability in the source area of the Yellow River is the best, followed by the upstream and middle reaches, and the downstream is the worst.

In summary, the freshwater ecosystem stability assessment method based on aquatic organisms and water environment factor data and based on the maximum flow principle can be well applied to the assessment of freshwater ecosystem stability.

The above are only examples of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the content of the description of the present invention, or directly or indirectly applied in other related technical fields, shall be regarded as Likewise, it is included in the patent protection scope of the present invention.

Claims (7)

1. A method for assessing the stability of a freshwater ecosystem, characterized by: the method comprises the following steps:

step S1, aiming at a selected research area, aquatic organism data and water environment factor data are collected through field monitoring and indoor experiments, and finally an aquatic organism data and water environment factor data index original data set of the research area is formed;

step S2, calculating Shannon-wiener diversity index of each aquatic organism through the aquatic organism data of the index original data set in the step S1, and performing multiple collinearity diagnosis on the water environment factor data of the index original data set; the processed aquatic organism data and the processed water environment factor data form an index pretreatment data set;

step S3, carrying out homogenization treatment on the index pretreatment data set in the step S2;

s4, constructing a freshwater ecological system stability evaluation model based on a maximum flow principle;

step S5, the index pretreatment data set after the homogenization treatment in the step S3 is simulated by using a self-organizing map network to obtain an evaluation model of the stability of the freshwater ecosystem in the step S4;

s6, inputting the entropy value and the index weight obtained by simulation in the step S5 into an Excel table, evaluating the stability of the freshwater ecological system through analysis of the entropy value, and identifying driving factors of the stability of the freshwater ecological system through analysis of the index weight;

in the step S3, carrying out homogenization treatment on the index pretreatment data set in the step S2; the method comprises the following steps:

the stability of the freshwater ecological system is improved along with the increase of index values of aquatic organisms and water environment factors, and the pre-normalization treatment is carried out as shown in a formula (1);

（1）；

wherein o represents the o-th index, b _o A unification value, x, representing the o index _o Represents the measured value, x, of the o index _omin Represents the minimum value of the o index, x _omax Represents the maximum value of the o index;

the stability of the freshwater ecological system is reduced along with the increase of index values of aquatic organisms and water environment factors, and the pre-normalization treatment is carried out as shown in a formula (2);

（2）。

2. a method for assessing the stability of a freshwater ecosystem according to claim 1, wherein: the aquatic life data includes: phytoplankton, zooplankton, benthonic animals and fish; the water environment factor data includes total nitrogen, total phosphorus, chemical oxygen demand, turbidity and pH.

3. A method for assessing the stability of a freshwater ecosystem according to claim 2, wherein: step S2, shannon-Winner diversity index of aquatic organisms is shown as a formula (3):

（3）；

wherein,represents Shannon-Winner diversity index, aRepresents species, S represents the total number of individuals of all species in the community,/->Is the percentage of the a-th species to the total number of individuals.

4. A method for assessing the stability of a freshwater ecosystem according to claim 3, wherein: and S2, performing multiple collinearity diagnosis by adopting a variance expansion factor method, and deleting the variable with the maximum variance expansion factor method in sequence, namely deleting the collinearity water environment factors until all the water environment factors of the variance expansion factor method are smaller than 5.

5. The method for assessing the stability of a freshwater ecosystem of claim 4, wherein:

in the step S4, a freshwater ecological system stability evaluation model is built based on a maximum flow principle, wherein the maximum flow principle refers to that an open complex system far away from equilibrium always searches for an optimization process, so that generalized flow obtained by the open complex system under given constraint conditions or cost reaches the maximum value; the specific process is as follows:

step S41, the aquatic organism and the water environment factor index are driving forces of the freshwater ecosystem evolution, and the written vector is in the form of x= (x) ₁ , x ₂ , ……, x _n ) The volume unit of a gamma space formed by n freshwater ecosystem subsystems is dx= (dx) ₁ , dx ₂ , ……, dx _n ) The method comprises the steps of carrying out a first treatment on the surface of the The generalized flux function calculation of the freshwater ecological system is shown in a formula (4);

（4）；

wherein J (x) represents generalized entropy, eta and eta obtained by the volume unit at the moment tFor coupling coefficient, i, j, k, l is water respectivelyFour indexes of biological and water environment factors, < ->And->Is an exponential term, defined as a potential function, by +.>Andreflecting the characteristic of the comprehensive performance evolution of the freshwater ecological system; x is x _i , x _j , x _k And x _l Normalized values of the aquatic organism and the water environment factor indexes i, j, k and l are respectively obtained;

the average generalized entropy of all microscopic states at the moment t is shown in a formula (5);

（5）；

wherein,representing general characteristics of information entropy of the freshwater ecosystem, wherein rho (x, t) is a time-varying probability density function of a random sequence set of aquatic organism and environmental factor indexes in the freshwater ecosystem at the moment t, and the method satisfies>；

Step S42, the generalized entropy J (x) obtained by the volume unit at the moment t of the freshwater ecosystem has no boundary, and constraint is needed to define the freshwater ecosystem; the constraint condition is converted into a first-order momentum equation to a fourth-order momentum equation, and the formula (6) is as follows:

（6）；

wherein f ₁ For the first order momentum equation, f ₂ Is a second order momentum equation, f ₃ Is a third order momentum equation, f ₄ In the form of a fourth order momentum equation,< >representing a statistical average;

step S43, optimizing the freshwater ecological system, obtaining the maximum generalized entropy under the given constraint condition, and maximizing the formula (5) by utilizing Lagrangian multipliers under the given constraint condition described by the formula (6), wherein the formula is specifically expressed as the formula (7):

（7）；

where ρ is the probability density, exp () represents an exponential function based on the natural constant e,obtained through Lagrangian optimization;

the potential function ensures the stability of the freshwater ecosystem mode, and is expressed as:

（8）；

wherein the potential functionRepresents the basic characteristics of the evolution of the freshwater ecosystem, x represents the random sequence of the aquatic organism and environmental factor indexes in the freshwater ecosystem, and +.>And->Obtained by Lagrangian optimization, +.>Is a parameter;

performing translation transformation on the formula (8) and diagonalizing transformation on the transformed constant term matrix to obtain a formula (9):

（9）；

wherein,is the stability entropy value of the fresh water ecological system, a _i Indicating index x _i Corresponding connection weight, x _i The normalized value of the index i of the aquatic organism and the water environment factor is obtained;

according to the relation between the potential function equation and the dynamic evolution equation, a random evolution equation of the ordered mode of the freshwater ecological system is derived and expressed as a formula (10):

（10）；

wherein,representation->The first derivative with respect to time, lambda is defined by a _i Eigenvalues of the constituent matrix->Is a nonlinear interaction term reflecting the interaction between indices, and F (t) is a random term.

6. A method for assessing the stability of a freshwater ecosystem according to claim 5, wherein: in step S5, the self-organizing map network is an unsupervised learning neural network model, which has the capability of processing complex nonlinear problems, and simulates or learns a sample space or an external unknown environment.

7. The method for assessing the stability of a freshwater ecosystem of claim 6, wherein: in step S5, the index preprocessing data set after the homogenization treatment in step S3 is simulated by using a self-organizing map network, which specifically includes: the data after normalization processing is put into a self-organizing map network of MATLAB 2018 for simulation, the training step length is set to 300, and index weight and entropy value representing the steady level of the freshwater ecosystem are obtained through simulation.