Environmental evaluation device and environmental evaluation method

Abstract

To evaluate conservation activities with high accuracy in relation to ecosystems that respond non-linearly to such activities. [Solution] The environmental assessment device of the present invention approximates the time-series relationship between the dependent variable and the explanatory variables to be analyzed, which serve as the basis for calculating biodiversity credits, using a nonlinear model; evaluates the degree of fluctuation and the symmetry of the fluctuations of the dependent variable based on the approximated nonlinear model; estimates the interaction between the explanatory variables to be analyzed based on the approximated nonlinear model; estimates the time at which the value of the dependent variable converges in the nonlinear model and the value of the dependent variable at the time of convergence based on the value of the dependent variable at the time of convergence; and evaluates the amount of change in the dependent variable that changes nonlinearly with respect to the explanatory variables to be analyzed based on the value of the dependent variable at the time of convergence.

Description

Technical Field

[0001] The present invention relates to an environmental assessment apparatus and an environmental assessment method.

Background Art

[0002] Recently, the impact of global warming caused by carbon dioxide and the like emitted along with socioeconomic activities has become prominent, and decarbonization efforts to mitigate this are being considered globally. These efforts include efforts to reduce the amount of greenhouse gases such as carbon dioxide emitted into the atmosphere (emission-side efforts), and efforts to increase the amount of greenhouse gases once emitted recovered from the atmosphere (recovery-side efforts).

[0003] Among these, the recovery-side efforts are called DAC (Direct Air Capture), and as representative examples, DAC by forest ecosystems and DAC by marine ecosystems are known. In DAC by forest ecosystems, companies, etc. obtain biodiversity credits corresponding to the increase by increasing the forest area through conservation actions such as afforestation. In DAC by marine ecosystems, companies, etc. obtain biodiversity credits corresponding to the increase by increasing the amount of algae habitat through conservation actions such as water quality improvement.

[0004] Biodiversity credits are the rightsization of the consideration contributed by companies, etc. to the recovery of greenhouse gases. Companies, etc. can recover the funds invested for the recovery of greenhouse gases by selling that right (biodiversity credits) through the market. In that sense, biodiversity credits are similar to carbon credits in emission-side efforts.

[0005] There are attempts to accurately predict or calculate the value of such biodiversity credits based on conservation actions by companies, etc. The environmental monitoring system of Patent Document 1 estimates the true value regarding past water quality by analyzing data dispersion through time-series data analysis of environmental data (specifically water quality). [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] International Publication No. 2017 / 088040 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] In emission-side efforts, there is a proportional relationship between greenhouse gas emissions and the amount of carbon credits. Here, "amount" is a hypothetical unit to which a monetary value is multiplied. Even if the two are not exactly proportional, there is a simple increasing relationship where an increase in greenhouse gas emissions leads to an increase in the amount of carbon credits. This kind of relationship is called "linear." On the other hand, in capture-side efforts, the relationship between the instrumental variable of conservation activities (e.g., afforestation area) and the amount of biodiversity credits is not linear.

[0008] There is a "nonlinearity" inherent to living organisms at play. For example, after afforestation, it takes a long time for trees to grow sufficiently and be able to perform sufficient photosynthesis, and the natural conditions during that time (temperature, rainfall, pest damage, etc.) are not uniform. For instance, while the target land has the appearance of a "savanna" after afforestation, the level of greenhouse gas capture is extremely low, and the rate of increase is also low. However, once a sufficient number of years have passed and the target land has the appearance of a "forest," the level of greenhouse gas capture becomes extremely high, and the rate of increase also increases. In other words, there is a period in which the relationship between the instrumental variables of conservation activities and the amount of biodiversity credits changes in a complex and random manner, and there is a period before and after which the level of greenhouse gas capture changes dramatically.

[0009] Even if natural conditions were uniform, the ecosystem's response to human conservation activities is not linear. For example, if conservation activities such as spreading fertilizer on a savanna are carried out, the savanna's condition will remain constant for a certain period, after which it will dramatically change into a forest. The term "nonlinear" includes this meaning as well.

[0010] The environmental monitoring system described in Patent Document 1 does not take the aforementioned nonlinearity into consideration. Therefore, the present invention aims to evaluate conservation activities with high accuracy in relation to ecosystems that exhibit nonlinear responses to such conservation activities (equilibrium 1: savanna → equilibrium 2: forest, etc.). [Means for solving the problem]

[0011] The present invention provides an environmental assessment device comprising: an input processing unit that accepts environmental information relating to the environment of a site to be analyzed, and a target variable from the environmental information that serves as a criterion for calculating biodiversity credits; a nonlinear model that takes the target variable as output and a plurality of explanatory variables from the environmental information that indicate the quantity or quality of conservation activities performed by humans on ecosystems as input; a nonlinear model that accepts the user's extraction of explanatory variables to be analyzed from the plurality of explanatory variables; and approximates the time-series relationship between the target variable and the explanatory variables to be analyzed using the nonlinear model by learning from past time-series values ​​of the target variable and the explanatory variables to be analyzed. The system is characterized by comprising: an environmental evaluation processing unit that evaluates the degree of fluctuation and the symmetry of the fluctuation of the objective variable based on the formula, estimates the interaction between the explanatory variables under analysis based on the approximated nonlinearity model, estimates the time at which the value of the objective variable converges in the nonlinearity model and the value of the converged objective variable based on the evaluation results of the degree of fluctuation and the symmetry of the fluctuation, and the estimated interaction, and evaluates the amount of change of the objective variable that changes nonlinearly with respect to the explanatory variables under analysis based on the value of the converged objective variable; and an output processing unit that displays the amount of change of the objective variable that has changed up to the time of convergence. Other means will be described in the section on embodiments for carrying out the invention. [Effects of the Invention]

[0012] According to the present invention, conservation activities can be evaluated with high accuracy on ecosystems that respond nonlinearly to such conservation activities. [Brief explanation of the drawing]

[0013] [Figure 1] This is a diagram showing the configuration of the environmental evaluation device. [Figure 2] This diagram illustrates the transitional period for environmental indicators. [Figure 3] This diagram illustrates the relationships between explanatory variables. [Figure 4] This diagram illustrates the changes in the dependent variable during a transitional period. [Figure 5] This diagram illustrates the degree of fluctuation and the symmetry of the fluctuation. [Figure 6] This is a diagram illustrating the transition of the equilibrium point. [Figure 7] This is a flowchart of the processing procedure. [Modes for carrying out the invention]

[0014] Hereafter, embodiments of the present invention ("this embodiment") will be described in detail with reference to the drawings.

[0015] (term) An ecosystem is a collection of organisms that contribute to the capture of greenhouse gases such as carbon dioxide. Ecosystems produce oxygen and glucose from carbon dioxide and water through photosynthesis. Environmental information refers to time-series physical quantities obtained from ecosystems and their surrounding environments by measuring instruments (sensors, etc.). Environmental information is often associated with a specific location being analyzed, such as a demarcated forest or ocean. Conservation activities are efforts undertaken by individuals or corporations to protect ecosystems with the aim of capturing greenhouse gases.

[0016] The dependent variable is a time-series indicator from environmental information that shows the ecosystem's ability to capture greenhouse gases. A high dependent variable indicates a superior quantity and variety of organisms in the ecosystem. As a result, the amount of biodiversity credits is also large. Generally, the indicators used as the basis for calculating biodiversity credits are called "environmental indicators." These environmental indicators can become the dependent variables in the models described below. The dependent variable is another name for the environmental indicator when the model is used. Explanatory variables are time-series indicators that show the quantity or quality of conservation activities performed by humans on an ecosystem. A single dependent variable may correspond to multiple explanatory variables. Explanatory variables may be environmental information or other management indicators.

[0017] A model is a function that takes a value of an explanatory variable as input and outputs a value of a dependent variable. It is possible to have a one-dimensional output for a two-dimensional input. For example, "Y=F(X1,X2)" is a function (model) where Y is the dependent variable and X1 and X2 are the explanatory variables. A typical model in this embodiment is a nonlinearity model. Since the dependent and explanatory variables are time-series indicators, as mentioned above, it is possible to partially differentiate the dependent and explanatory variables of the model with respect to time.

[0018] Interactions between explanatory variables refer to the correlations between multiple explanatory variables when several such variables exist. The fluctuation of the dependent variable is the variance of the dependent variable as it changes over time. Variance is defined for each time window. Note that variance can be replaced by the amplitude of the fluctuation (waveform amplitude). Fluctuation symmetry refers to the appearance of sample points in a coordinate plane with the value of a certain dependent variable at time t and the value of the same dependent variable at time t+1 as the axis. When sample points are distributed along a 45-degree line, the symmetry is low, and when sample points are concentrated around a point on the 45-degree line, the symmetry is high (details are described later in Figure 5). The 45-degree line is a line (symmetry line) that serves as a guideline for the symmetry of the appearance of sample points in the coordinate plane. In this embodiment, the symmetry line is a line passing through the origin with a slope of 45 degrees and is referred to as the 45-degree line, but the slope and other parameters are not limited to this.

[0019] In marine ecosystems, for example, reactive nitrogen concentration and phosphate concentration are explanatory variables, while pH (hydrogen ion concentration), dissolved oxygen level, population size of specific surrogate species, and DNA (Deoxyribo-Nucleic Acid) content of specific surrogate species are dependent variables. In forest ecosystems, for example, the Normalized Difference Vegetation Index (NDVI) of land and atmospheric carbon dioxide concentration are explanatory variables, while land carbon reserves are dependent variables.

[0020] (Biodiversity credits) First, let's briefly explain the values ​​that should underlie biodiversity credits. Ecosystems provide benefits to human economic activities. For example, without a supply of fresh air, humans cannot conduct economic activities. These so-called "gifts of nature" are called "ecosystem services." Biodiversity credits can be said to be the payment for maintaining these ecosystem services.

[0021] Ecosystem services include supply services, regulation services, cultural services, and habitat services. Forest carbon sequestration and water quality maintenance are examples of regulation services. The supply of food, raw materials, fuel, and freshwater are examples of supply services. Biodiversity credits are used to assess these ecosystem services.

[0022] (Configuration of the environmental evaluation device) Figure 1 shows the configuration of the environmental evaluation device 1. The environmental evaluation device 1 is a general-purpose computer. The environmental evaluation device 1 has a central control unit 11, input devices 12 such as a mouse and keyboard, output devices 13 such as a display, main memory 14, auxiliary storage 15, and communication device 16. The input processing unit 21, environmental evaluation processing unit 22, output processing unit 23, data acquisition unit 24, and credit granting unit 25 of the main memory 14 are programs. In the following description, when the subject is indicated as "the ○○ unit," it means that the central control unit 11 reads the program from the auxiliary storage 15 to the main memory 14 and executes the processing (details described later) that is written in the program beforehand. The auxiliary storage 15 stores the environmental information database 31 and model 32.

[0023] External device 2 is connected to environmental assessment device 1 via network 3. External device 2 stores environmental information measured by users of environmental assessment device 1 and other businesses. Measuring instruments 43a, 43b, and 43c are placed at locations where environmental information should be acquired. Of these, measuring instrument 43a is a ground sensor that measures, for example, the carbon dioxide concentration in a forest over time. Measuring instrument 43b is an optical measuring instrument that measures, for example, the number of trees and the thickness of trees in a forest over time. Measuring instrument 43c is a satellite measuring instrument that measures, for example, the phase and density of a forest from a geostationary satellite over time. Each of measuring instruments 43a, 43b, and 43c transmits the environmental information it has acquired to environmental assessment device 1. External device 2 may also store environmental information acquired by the company itself. In this sense, external device 2 means any device other than environmental assessment device 1.

[0024] Measurement condition modulators 42a, 42b, and 42c exist corresponding to the measuring instruments 43a, 43b, and 43c. The measurement condition controller 41 receives control commands for the measuring instruments 43a, 43b, and 43c from the environmental evaluation device 1 and transmits these control commands to the measuring instruments 43a, 43b, and 43c via the measurement condition modulators 42a, 42b, and 42c at appropriate timings. The control commands include the timing and duration of environmental information acquisition.

[0025] As a result of the data exchange described above, the environmental information database 31 of the auxiliary storage device 15 will accumulate various environmental information from various locations in chronological order.

[0026] (Transitional period for environmental indicators) Figure 2 illustrates the transitional period of environmental indicators. The horizontal axis in Figure 2 represents time, and the vertical axis represents environmental indicators. Before the start of conservation activities (t0), the environmental index is stable at a certain level Y0. After time t0, conservation activities are carried out systematically and continuously. However, after time t0, the environmental index fluctuates in a complex manner. For a period after time t0, the environmental index irregularly experiences rapid increases and decreases. This period is called the "transitional period." Environmental indicators during the transitional period are significantly unstable and cannot be used as a basis for calculating biodiversity credits. After a period following the transitional period, the environmental index returns to another level Y * It then stabilizes (converges) again. The timing at which the results of the preservation action appear correctly is at time t * From here on. And the result of the conservation action (increase) is ΔY = Y * The value is -Y0. The amount of biodiversity credits is calculated based on ΔY.

[0027] (Relationships between explanatory variables) Figure 3 illustrates the relationships between explanatory variables. Suppose we have a single dependent variable and three explanatory variables X1, X2, and X3. Graph 51 shows the time-series change of explanatory variable X1. Graph 52 shows the time-series change of explanatory variable X2. Graph 53 shows the time-series change of explanatory variable X3.

[0028] Focusing on graphs 51 and 52, the explanatory variables X1 and X2 change in almost the same way and at the same level over time. Focusing on graphs 51 and 53, the explanatory variables X1 and X3 change in completely different ways and at different levels over time. The same can be said for graphs 52 and 53. From the above, it can be said that the interaction between explanatory variables X1 and X2 is strong, while the interaction between explanatory variables X1 and X2 and explanatory variable X3 is weak. The strength of the interaction between two explanatory variables can be determined by calculating the correlation between the two explanatory variables. Furthermore, if the above model is a dynamical system model, calculating the correlation is nothing more than calculating the coefficients of the differential equation of the dynamical system model. In other words, calculating the interaction (correlation) between two explanatory variables is a factor that determines the shape of the model "Y=F(X1,X2)" in coordinate space.

[0029] (Changes in the dependent variable during the transitional period) Figure 4 illustrates the change in the dependent variable during the transition period. Graph 61 shows the relationship between the explanatory variable X and the dependent variable Y. When the value of the explanatory variable X is sufficiently small, the value of the dependent variable Y is stable ((a) Equilibrium 1). When the value of the explanatory variable X is sufficiently large, the value of the dependent variable Y is also stable ((a) Equilibrium 2). However, the value of the dependent variable Y itself is much larger in Equilibrium 2 than in Equilibrium 1.

[0030] When the value of the explanatory variable X lies between equilibrium 1 and equilibrium 2 (b), the value of the dependent variable Y increases in the long term, irregularly alternating between increases and decreases. The S-shaped portion of the curve illustrates this.

[0031] Graph 62 shows the time series of the environmental indicator Y in the vicinity of equilibrium 1 or equilibrium 2 (a). In these vicinitys, the dependent variable Y is stable at a certain level. That is, the degree of fluctuation of the dependent variable (variance or amplitude of the waveform) is small.

[0032] Graph 63 shows the time series of the dependent variable Y in the region (b) between the vicinity of equilibrium 1 and equilibrium 2. In this region, the dependent variable Y fluctuates with a short period and a large variance (amplitude). In other words, the degree of fluctuation of the dependent variable is large.

[0033] By calculating the degree of fluctuation in the dependent variable, it is possible to determine whether or not the dependent variable is in equilibrium (converging). Furthermore, if the aforementioned model is a dynamical system model, calculating the degree of fluctuation is equivalent to calculating the magnitude of the change in the time derivative of the differential equation of the dynamical system model.

[0034] (Degree of fluctuation and symmetry of fluctuation) Figure 5 illustrates the degree and symmetry of fluctuations. Graph 71 shows the time series of the dependent variable Y. Graph 71 is essentially the same as graphs 62 and 63 in Figure 4. Specifically, in graph 71, the value of the dependent variable Y is centered around "20" and falls within a range of approximately ±5. In that sense, the degree of fluctuation in graph 71 is small (similar to the example in graph 62).

[0035] The vertical axis of graphs 72a and 72b represents time t. n The horizontal axis represents the value of the dependent variable Y at time t. n+1 This is the value of the dependent variable Y. Furthermore, a "line of symmetry (45-degree line)" is drawn on this graph, connecting points where the values ​​on the vertical axis and horizontal axis are the same.

[0036] Now, for example, in Graph 71, the waveform is trending upwards to the right, and the level at the left end of the waveform is "10", and the level at the right end of the waveform is "30". In this case, time t n The value of the dependent variable Y and time point t n+1 The points representing combinations of values ​​for the dependent variable Y are clustered around the line segment connecting the points "(10,10)" and "(30,30)" on the line of symmetry, as shown in Graph 72a. For the sake of explanation, this transition of the dependent variable Y is called "modulation pattern 1".

[0037] As another example, in Graph 71, assume that the waveform is shifting horizontally around a certain level (e.g., "20"). In this case, the point indicating the combination of the value of the target variable Y at time point t n and the value of the target variable Y at time point t n+1 collects around the point "(20, 20)" on the symmetry line, as shown in Graph 72b. For convenience of explanation, such a transition of the target variable Y is called "modulation pattern 2".

[0038] More generally, four modulation patterns can be assumed by combining the magnitude of the fluctuation and the level of the symmetry of the fluctuation as a pair.

[0039] 〈Modulation pattern 1: The degree of fluctuation is small and the symmetry of the fluctuation is low.〉 The target variable Y fluctuates stably in a short period, while its level fluctuates in a longer period. 〈Modulation pattern 2: The degree of fluctuation is small and the symmetry of the fluctuation is high.〉 The target variable Y fluctuates stably in a short period, while its level is stable in a longer period. 〈Modulation pattern 3: The degree of fluctuation is large and the symmetry of the fluctuation is low.〉 The target variable Y fluctuates greatly in a short period, while its level fluctuates in a longer period. 〈Modulation pattern 4: The degree of fluctuation is large and the symmetry of the fluctuation is high.〉 The target variable Y fluctuates greatly in a short period, while its level is stable in a longer period.

[0040] It can be said that the target variable Y in modulation pattern 2 is suitable as a criterion for calculating biodiversity credits. Conversely, it can be said that the target variable Y in modulation pattern 3 is the least suitable.

[0041] (Equilibrium point transition) Figure 6 illustrates the transition of the equilibrium point. When conservation activities are carried out on a forest, the value of the dependent variable Y increases as the value of the explanatory variable X1 (or X2), which indicates the quantity or quality of the conservation activities, increases. Graphs 81 and 82 succinctly show this transition. As mentioned above, the relationship between the explanatory variable X1 (or X2) and the dependent variable Y is by no means linear. Initially, when the forest is in a "savanna" state, the rate at which the dependent variable Y increases (the increase in the dependent variable in response to a small increase in the explanatory variable) is small. However, this rate increases rapidly at a certain point in time (the forest growth period). Then, at a later point in time, this rate returns to almost its original level.

[0042] The problem lies in the way the dependent variable Y changes (equilibrium point change) between the two boundary time points. If the change between the two time points is gradual, the relationship between the explanatory variable X1 (or X2) and the dependent variable Y will be as shown in Graph 81. A curve like that in Graph 81 is generally called a logistic curve (c).

[0043] When the change between two time points is drastic, the relationship between the explanatory variable X1 (or X2) and the dependent variable Y will look like Graph 82. A curve like the one in Graph 82 is generally called a hysteresis curve (d). Note that, due to limitations in plotting techniques, the hysteresis curve is represented as two straight lines in Figure 6. The two dashed arrows pointing upward and downward indicate that the value of the dependent variable Y is influenced not only by the current value of the explanatory variable X1 (or X2), but also by past values ​​of the explanatory variable X1 (or X2).

[0044] Graph 83 plots a model "Y=F(X1,X2)" representing the relationship between three variables, X1, X2, and Y, within a coordinate space with three axes representing these variables. The model here is a three-dimensional figure resembling a "flying carpet." When this figure is projected onto a plane with X1 and Y as its axes, the relationship between X1 and Y appears on that plane. Similarly, when this figure is projected onto a plane with X2 and Y as its axes, the relationship between X2 and Y appears on that plane. In Figure 6, due to construction constraints, these relationships are not depicted.

[0045] Graph 83 does not have a time axis. However, it is possible to consider time in relation to each point on the three-dimensional figure. For example, when a company increases the quantity or quality of maintenance activities over time, initially, at the start of the maintenance activities, the value of explanatory variable X1 is close to "0", and the value of explanatory variable X2 is also close to "0". Therefore, the point representing the combination of values ​​of the three variables moves across the surface of the three-dimensional figure from the lower left corner in the foreground to the upper right corner in the background. In other words, if this point is projected onto a horizontal plane, it moves from the lower left to the upper right in the time series. If this point is projected onto a vertical plane, it moves from the bottom to the top in the time series.

[0046] For the sake of simplicity, let's assume that the value of the explanatory variable X2 is fixed at a relatively small first value. Then, when the dependent variable Y is in modulation pattern 2, the points representing the combination of values ​​of the three variables are, for example, points 84 and 87 on curve (d). When the dependent variable Y is in modulation pattern 3, the points representing the combination of values ​​of the three variables are, for example, points 85 and 86 on curve (d). Time progresses in the order of point 84 → point 85 → point 86 → point 87.

[0047] Furthermore, let's assume that the value of the explanatory variable X2 is fixed at a relatively large second value. Then, points 84, 85, 86, and 87 will move along curve (c). Curves (c) and (d) in graph 83 correspond to graphs 81 and 82, respectively. In short, the points representing combinations of values ​​for the three variables rise while moving randomly in the forward, backward, vertical, and horizontal directions on a flying carpet with a horizontal width.

[0048] (Processing procedure) Figure 7 is a flowchart of the processing procedure. As a prerequisite for the processing procedure to begin, the data acquisition unit 24 stores various environmental information acquired from measuring instruments 43a, 43b, and 43c and external device 2 in the environmental information database 31 in chronological order. The environmental information stored here includes one or more dependent variables and one or more independent variables.

[0049] In step S101, the input processing unit 21 of the environmental evaluation device 1 receives environmental information. Specifically, firstly, the input processing unit 21 acquires environmental information stored in the environmental information database 31. Secondly, the input processing unit 21 receives, via the input device 12, the user's designation of one target variable from among one or more target variables as the criterion for calculating biodiversity credits. For the sake of explanation, here we will refer to the target variable Y. p Let's assume that this is specified. The input processing unit 21 may accept the user specifying a location to be protected, propose candidate combinations of the protected object and the target variable corresponding to the accepted location to the user, and accept the user selecting a combination from among the candidates.

[0050] In step S102, the environmental evaluation processing unit 22 of the environmental evaluation device 1 sets up model 32. Specifically, the environmental evaluation processing unit 22 prepares model 32, which is, for example, a differential equation of a dynamical system model. Model 32 has the objective variable Y p It includes multiple explanatory variables and parameters (coefficients for explanatory variables, etc.). The environmental evaluation processing unit 22 processes the target variable Y pFurthermore, Model 32 can be plotted as a three-dimensional shape (manifold) within a multidimensional space with multiple explanatory variables as axes. The shape and position of this three-dimensional shape are determined by the specific values ​​of the parameters.

[0051] A differential equation for a dynamical system model is, for example, the equation of motion. The equation of motion has acceleration (as the second time derivative of the position coordinate), mass, and force (kinetic energy) as variables, as expressed as "force = mass × acceleration". For example, by substituting mass with the mass (or area) of the forest to be conserved, substituting the position coordinate with the dependent variable, and substituting force with the independent variable, the differential equation for a dynamical system model can be used as Model 32 in this embodiment.

[0052] In step S103, the environmental assessment processing unit 22 extracts explanatory variables. Specifically, the environmental assessment processing unit 22 receives a predetermined number (for example, two) of explanatory variables selected by the user from among multiple explanatory variables via the input device 12. For the sake of explanation, the user selects X as the target variable that they particularly want to analyze. q and X r Let's assume we have extracted it.

[0053] In step S104, the environmental assessment processing unit 22 approximates past data with model 32. Specifically, the environmental assessment processing unit 22 approximates the time-series target variable Y in the past. p Value of explanatory variable X q The value of the explanatory variable X r The parameter values ​​of Model 32 are optimized using machine learning (multiple regression analysis) with supervised training data, which is a set of combinations of values ​​for Y. The Model 32 with optimized parameter values ​​is called an "approximate model". The approximate model is based on past data of the target variable Y. p Explanatory variable X q and explanatory variable X r It accurately approximates the time-series relationship. The environmental evaluation processing unit 22 can draw the approximate model in three-dimensional space. The environmental evaluation processing unit 22 assigns "(X q ,X r ,Yp It associates the time with the time.

[0054] In step S105, the environmental evaluation processing unit 22 evaluates the degree of fluctuation and symmetry. Specifically, firstly, the environmental evaluation processing unit 22 creates graph 71 in Figure 5 by analyzing the shape and position of the approximate model, and for each time window, the objective function Y p Obtain the degree of fluctuation of the dependent variable Y. p This is a representative value (maximum value, mean value, etc.) of the variance (amplitude). A time window is a section of time cut out from a continuous period of time, for example, one day, one week, etc., of a predetermined length. The environmental evaluation processing unit 22 determines for each time window whether the degree of fluctuation is less than a predetermined first threshold and stores the result. As described above, if the model is a dynamical system model, calculating the degree of fluctuation is nothing more than calculating the magnitude of the change in the time derivative of the differential equation of the dynamical system model.

[0055] Secondly, the environmental evaluation processing unit 22 creates graph 72 in Figure 5 by analyzing the shape and position of the approximate model, and for each time window, it determines whether the symmetry of the fluctuations is higher than a predetermined second threshold and stores the result. Here, "creating graph 72" means that before creation, it is unknown whether the creation result will be graph 72a or graph 72b. The environmental evaluation processing unit 22 draws an ellipse that encloses the points distributed around the 45-degree line of graph 72, and uses the ratio of the length of the major axis of the ellipse to the length of the minor axis as the symmetry of the fluctuations. Then, if the symmetry falls within the range of the second threshold "1±α", the environmental evaluation processing unit 22 determines that the symmetry is high and stores the result. For example, α = 0.1.

[0056] Thirdly, the environmental evaluation processing unit 22, based on the results of the “first” and “second” steps of step S105, evaluates the evaluation result information “[W n ,A n ,S n Create and temporarily store the following: ]=[W1,A2,S3],[W2,A2,S2],[W3,A3,S3],… nThis is the nth (n=1,2,3,…) time window. n This is the result of comparing the degree of fluctuation in the nth time window with the first threshold, and is either "small" or "large". "Small" indicates that the degree of fluctuation is smaller than the first threshold, and "large" indicates otherwise. n This is the result of comparing the symmetry of the fluctuations in the nth time window with the second threshold, and is either "high" or "low". "High" indicates that the symmetry of the fluctuations falls within the range of the second threshold, and "low" indicates otherwise.

[0057] In step S106, the environmental assessment processing unit 22 estimates the interaction. Specifically, the environmental assessment processing unit 22 uses an approximate model to estimate the target variable X q and the dependent variable X r For each of these, create graph 51 in Figure 3. The environmental evaluation processing unit 22 processes the target variable X q and the objective variable X r The correlation between the two variables is calculated and the result is stored. The calculated correlation takes a value between "-1" and "1", and the larger the absolute value, the greater the degree of correlation (positive or negative correlation). As mentioned above, calculating this correlation is equivalent to calculating the coefficient between the two variables in the differential equation of the dynamical system model.

[0058] In step S107, the environmental evaluation processing unit 22 estimates the convergence of the target variable. Specifically, firstly, the environmental evaluation processing unit 22 temporarily holds the evaluation result “[W n ,A n ,S n Get the ]" and read the first [ ], the second [ ], the third [ ], ... in order. A n Focusing on this, we have empirically observed that as n increases, it starts with a series of "small" values, then a series of "large" values, and finally, another series of "small" values. n Focusing on this, we have empirically observed that as n increases, it starts with a series of "high" values, then a series of "low" values, and finally another series of "high" values. Here, A n It is "large", and S nIf there are consecutive time windows in which the value is "low", then in those time windows, the dependent variable Y p This is a transitional period. This transitional period corresponds to “Transitional Period” in Figure 2 and (b) in Graph 61 of Figure 4.

[0059] Secondly, the environmental evaluation processing unit 22 said, "A n After becoming "large," W became "small" for the first time. n " and "S n After becoming "low," W became "high" for the first time. n Of these, W which is later in time n The “re-equilibrium time window W” * The environmental evaluation processing unit 22 determines the reequilibrium time window W as follows: * In this case, the dependent variable is considered to have converged.

[0060] Thirdly, the environmental evaluation processing unit 22 controls the reequilibrium time window W * "In the subsequent approximation model, the target variable X q and the objective variable X r The correlation between the two, and the dependent variable X q The coefficients and the dependent variable X r We focus on the coefficients. Then, the environmental evaluation processing unit 22 checks whether the following conditions 1 to 3 are met between these.

[0061] <Condition 1> Dependent variable X q and the objective variable X r A correlation is observed between the two. <Condition 2> Dependent variable X q The time series change of the sign of the coefficient is “(t n ,t n+1 ,t n+2 ,t n+3 ,t n+4 ,…) = (positive, negative, positive, negative,…). <Condition 3> Dependent variable X r The time series change of the sign of the coefficient is “(t n ,t n+1 ,t n+2 ,t n+3 ,t n+4 ,…) = (negative, positive, negative, positive,…).

[0062] Conditions 2 and 3 are those in which, within a predetermined number of consecutive time windows, the signs of the two explanatory variables appear in a two-phase transition with respect to the dependent variable, and their signs cancel each other out. In this case, the dependent variable Y p The probability of convergence is higher. Note that the "third" process in step S107 is optional. The environmental evaluation processing unit 22 may determine that the target variable has converged without performing the "third" process in step S107. The environmental evaluation processing unit 22 may determine that the target variable has converged only after evaluating that the target variable has converged in the "second" step of S107, and if conditions 1 to 3 are also met.

[0063] More generally, the environmental assessment processing unit 22 is δY P If the absolute value of / δt falls below a predetermined threshold, it may be determined that the dependent variable has converged.

[0064] In step S107, the environmental evaluation processing unit 22 traces the shape of the approximated nonlinear model in coordinate space and finds the equilibrium (convergence) point of the target variable. The partial (local) shape of the nonlinear model depends on the evaluation results of the degree of fluctuation and the symmetry of the fluctuations, as well as the estimated interactions. In other words, the environmental evaluation processing unit 22 estimates the point in time when the value of the target variable converges in the nonlinear model, and the value of the target variable at which it converges, based on the evaluation results of the degree of fluctuation and the symmetry of the fluctuations, as well as the estimated interactions.

[0065] In step S108, the environmental evaluation processing unit 22 evaluates the change in the target variable. Specifically, the environmental evaluation processing unit 22 evaluates W * Y corresponding to * Obtain ΔY=Y * Calculate -Y0.

[0066] In step S109, the output processing unit 23 of the environmental evaluation device 1 displays the change in the target variable. Specifically, the output processing unit 23 displays the value of ΔY on the output device 13. After that, the processing procedure ends.

[0067] (Variation 1: Future application of the model) As described above, the environmental assessment processing unit 22 retrospectively determines when the dependent variable converged by looking back at the values ​​of the dependent variable in the past. However, the environmental assessment processing unit 22 can also apply in the future an approximate model learned using past values ​​of the dependent variable and explanatory variables. For example, the environmental assessment processing unit 22 can determine the "re-equilibrium time window W" from the start of the maintenance activity. * The length of the "transition period" until the target value is reached is calculated. Then, the environmental assessment processing unit 22 uses the value of the target variable after the transition period has elapsed as the basis for calculating biodiversity credits in future cases having the same target variable and explanatory variables as the approximate model.

[0068] The environmental evaluation processing unit 22 may extract an approximate model from among several approximate models learned using past data in which at least a portion of the point clouds of the dependent and independent variables as specific future (predicted) data are similar. The environmental evaluation processing unit 22 then selects the "re-equilibrium time window W" in the extracted approximate model. * This will determine the corresponding future point in time.

[0069] (Variation 2: Credit Granting Section) Biodiversity credits are tradable. Just as doubts about a company's financial statements result in extremely low market liquidity for its securities, the liquidity of biodiversity credits depends on the "confidence" of the method used to determine ΔY. This confidence can be defined in various ways. One example is a weighted average of sample size and convergence. The larger the sample size and the greater the convergence, the higher the confidence.

[0070] The number of samples refers to the result evaluation information “[W n ,A n ,S n The number of terms in the expression ]=[W1,A2,S3],[W2,A2,S2],[W3,A3,S3],... is the number of time windows. The larger the sample size M, the more statistically significant ΔY is. The degree of convergence is defined as the "re-equilibrium time window W". * This is a weighted average of the degree of fluctuation and the symmetry of the fluctuations in the given variable. The smaller the degree of fluctuation and the higher the symmetry of the fluctuations, the greater the degree of convergence. The greater the degree of convergence, the more statistically significant ΔY is.

[0071] In step S109, the credit allocation unit 25 of the environmental assessment device 1 calculates the increase or decrease in biodiversity credits by multiplying ΔY by a predetermined conversion factor. At this time, the credit allocation unit 25 calculates the confidence level based on the number of samples and the degree of convergence. The output processing unit 23 displays the calculated confidence level along with the increase or decrease in biodiversity credits on the output device 13. If the confidence level is below a predetermined threshold, the credit allocation unit 25 may display a warning on the output device 13. Examples of warnings include, "The liquidity of biodiversity credits appears to be low," and "Biodiversity credits are not easily traded."

[0072] Furthermore, there are various specific methods for calculating biodiversity credits based on the change in the dependent variable. In addition, biodiversity credits may be calculated as monetary value, or as an imputed value (imputed price) that cannot be expressed as a simple monetary value. Incidentally, as an example of calculating biodiversity credits as an imputed price, there is Japanese Patent Application No. 2024-139039, in which some of the applicants and inventors are the same as those of this application.

[0073] (Modification 3: Repeated processing) In step S109 of the modified example 2 described above, if the credit granting unit 25 displays a warning, the process returns to step S101. Subsequently, the process from steps S101 to S109 is repeated until the credit granting unit 25 no longer displays a warning in step S109, which has been passed through again. During this time, the user changes the type of dependent variable and the type of independent variables each time. Furthermore, the user changes the sample size, the length of the time window, etc., each time.

[0074] Then, in step S109 of each repeating loop, the credit granting unit 25 calculates the confidence level. The credit granting unit 25 may also display multiple combinations of "(increase / decrease in biodiversity credits, confidence level)" on the output device 13 in descending order of confidence level. Here, the confidence level also represents the priority for users to market those biodiversity credits.

[0075] (Effects of the embodiment) (1) The environmental assessment device can accurately evaluate the criteria for calculating biodiversity credits by approximating ecosystem changes using a nonlinear model. (2) The environmental assessment device can calculate biodiversity credits. (3) The environmental evaluation device can evaluate the reliability of the objective variable. (4) The reliability of the environmental evaluation device can be evaluated based on the degree of fluctuation and symmetry. (5) The environmental evaluation device can use a general-purpose mechanical model. (6) The environmental evaluation device can accurately define the degree of fluctuation and symmetry. (7) The environmental evaluation device can accurately define the convergence of the target variable.

[0076] While several embodiments of the present invention have been described, these embodiments are presented as examples only and do not limit the scope of the invention. These novel embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0077] 1. Environmental evaluation device 2 External device 3 Network 11 Central Control Unit 12 Input devices 13 Output device 14 Main memory 15 Auxiliary storage 16. Communication equipment 21 Input Processing Unit 22 Environmental Evaluation Processing Unit 23 Output Processing Unit 24 Data Collection Department 25 Credit Granting Department 31 Environmental Information Database 32 Models 41 Measurement Condition Controller 42, 42a, 42b, 42c Measurement condition modulator 43, 43a, 43b, 43c Measuring Instruments

Claims

1. An input processing unit that accepts environmental information about the environment of the site to be analyzed, and a target variable from the said environmental information that serves as the basis for calculating biodiversity credits, A nonlinear model is set up with the aforementioned dependent variable as the output and multiple explanatory variables from the aforementioned environmental information that indicate the quantity or quality of conservation activities performed by humans on ecosystems as inputs. The system accepts the user's selection of the explanatory variables to be analyzed from among the aforementioned multiple explanatory variables. By learning from past time series values ​​of the objective variable and the explanatory variable being analyzed, the relationship between the time series of the objective variable and the explanatory variable being analyzed is approximated by the nonlinear model. Based on the approximated nonlinearity model, the degree of fluctuation of the objective variable and the symmetry of the fluctuations are evaluated. Based on the aforementioned approximate nonlinear model, the interaction between the explanatory variables under analysis is estimated. Based on the evaluation results of the degree of fluctuation and the symmetry of the fluctuation, and the estimated interaction, the time at which the value of the objective variable converges in the nonlinear model and the value of the converged objective variable are estimated. An environmental evaluation processing unit that evaluates the amount of change in the objective variable, which changes non-linearly with respect to the explanatory variable being analyzed, based on the convergent value of the objective variable, An output processing unit that displays the amount of change in the target variable up to the point of convergence, To be equipped, An environmental evaluation device characterized by the following.

2. The system includes a credit allocation unit that calculates an increase or decrease in biodiversity credits corresponding to the change in the aforementioned objective variable, The output processing unit, Display the increase or decrease in the biodiversity credits calculated above. The environmental evaluation apparatus according to claim 1, characterized by the following:

3. The aforementioned credit granting unit, To calculate the confidence level of the change in the target variable up to the point of convergence. The environmental evaluation apparatus according to claim 2, characterized by the following:

4. The aforementioned confidence level is, The weighted average of the degree of fluctuation of the dependent variable and the symmetry of the fluctuations, The environmental evaluation apparatus according to claim 3, characterized by the following:

5. The aforementioned nonlinear model is, It is a dynamical system model. The environmental evaluation apparatus according to claim 1, characterized by the following:

6. The degree of fluctuation in the aforementioned dependent variable is, This is the variance of the aforementioned dependent variable, which fluctuates over time. The symmetry of the fluctuations of the aforementioned dependent variable is, This refers to the appearance of sample points in a coordinate plane with the value of the objective variable at time t and the value of the objective variable at time t+1 as the axes. The aforementioned environmental evaluation processing unit is In the aforementioned coordinate plane, if the sample points are distributed along a line of symmetry that serves as an indicator of the symmetry of how the sample points appear, the symmetry is evaluated as low. If the sample points are concentrated around a certain point on the line of symmetry, the symmetry is evaluated as high. The environmental evaluation apparatus according to claim 1, characterized by the following:

7. The aforementioned environmental evaluation processing unit is The combination of the magnitude of the fluctuation and the degree of symmetry of the fluctuation is obtained in a time series. The objective variable is evaluated as having converged at the later of the following two points in time: the point at which the degree of the fluctuation first becomes small after it has become large, and the point at which the symmetry of the fluctuation first becomes high after it has become low. The value of the aforementioned dependent variable at that point in time shall be used as the basis for calculating the biodiversity credits. The environmental evaluation apparatus according to claim 6, characterized by the following:

8. The input processing unit of the environmental evaluation device is We accept environmental information about the environment of the site to be analyzed, and the target variable from the said environmental information that will be used as the basis for calculating biodiversity credits. The environmental evaluation processing unit of the aforementioned environmental evaluation device is: A nonlinear model is set up with the aforementioned dependent variable as the output and multiple explanatory variables from the aforementioned environmental information that indicate the quantity or quality of conservation activities performed by humans on ecosystems as inputs. The system accepts the user's selection of the explanatory variables to be analyzed from among the aforementioned multiple explanatory variables. By learning from past time series values ​​of the objective variable and the explanatory variable being analyzed, the relationship between the time series of the objective variable and the explanatory variable being analyzed is approximated by the nonlinear model. Based on the approximated nonlinearity model, the degree of fluctuation of the objective variable and the symmetry of the fluctuations are evaluated. Based on the aforementioned approximate nonlinear model, the interaction between the explanatory variables under analysis is estimated. Based on the evaluation results of the degree of fluctuation and the symmetry of the fluctuation, and the estimated interaction, the time at which the value of the objective variable converges in the nonlinear model and the value of the converged objective variable are estimated. Based on the convergence of the dependent variable, the amount of change in the dependent variable, which changes non-linearly with respect to the explanatory variable under analysis, is evaluated. The output processing unit of the aforementioned environmental evaluation device is To display the amount of change in the target variable up to the point of convergence. An environmental assessment method characterized by the following.

Images