A wetland ecological environment monitoring method based on multi-dimensional data analysis

Abstract

The present invention relates to the field of environmental monitoring technology, specifically disclosing a wetland ecological environment monitoring method based on multi-dimensional data analysis. The method comprises the following steps: selecting a number of height values ​​within a preset height range, setting sampling points in a wetland lake based on the height values, and obtaining the resistance of the water body at a single sampling point in real time; when a resistance greater than a preset value exists, verifying the resistance based on the resistance change within a monitoring period and obtaining a verification result; when the verification result is a failure, replacing the resistance greater than the preset value with the average resistance within a preset supplementary period; plotting a curve of the resistance of the water body at the sampling point over time, determining whether the water quality at the sampling point is abnormal based on the curve, and assessing the wetland ecological environment status based on the spatiotemporal information of the abnormal water quality. The present invention improves the accuracy of wetland ecological environment monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a wetland ecological environment monitoring method based on multi-dimensional data analysis. Background Art

[0002] A wetland ecosystem is a transitional ecosystem dominated by hydrological conditions, characterized by long-term or periodic soil saturation, and nurturing unique plant and animal communities. It combines the characteristics of both land and water, forming a unique ecological pattern with distinct structures and functions.

[0003] Currently, wetland ecological monitoring often focuses on numerical water quality indicators. Real-time monitoring and statistical analysis of parameters such as conductivity, dissolved oxygen, and pH are used to determine whether the ecosystem is experiencing unusual fluctuations. However, in actual monitoring, the impact of wildlife activities on water conductivity is often overlooked. For example, when waterfowl forage, fish swim, or other large terrestrial animals wade, they often stir up sediment and organic matter at the bottom of lakes or rivers, causing a transient increase in suspended matter concentrations. This increase in dissolved ions in sediment and organic matter causes a temporary increase in water conductivity. If not discriminated, these conductivity changes caused by biological activity can easily be misinterpreted as signals of pollutant intrusion or environmental deterioration, triggering false warnings or unnecessary interventions. This bias not only undermines the accuracy of monitoring data but also hinders scientific assessments and decision-making support for wetland ecological status. Summary of the Invention

[0004] The purpose of the present invention is to provide a wetland ecological environment monitoring method based on multi-dimensional data analysis to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A wetland ecological environment monitoring method based on multi-dimensional data analysis includes the following steps:

[0007] Select several height values ​​within a preset height range, set sampling points in the wetland lake based on the height values, and obtain the water resistance at each sampling point in real time;

[0008] When there is a resistance greater than a preset value, verification is performed based on the resistance change during the monitoring period to obtain a verification result, wherein the verification result includes verification pass or verification fail;

[0009] When the verification result is failure, the resistance greater than the preset value is replaced by the average resistance within the preset replenishment period;

[0010] A curve showing the change in resistance of the water body at the sampling point over time is drawn, and based on the curve, whether the water quality at the sampling point is abnormal is determined, and the wetland ecological environment status is evaluated based on the spatiotemporal information of the abnormal water quality.

[0011] As a further solution of the present invention: setting the sampling point includes:

[0012] Set a height range. Starting from the starting point of the height range, set several height values ​​at intervals of the preset height range. For height value a, set n sampling points in a single lake, where n is the preset number and the distance from the sampling point to the bottom of the lake is height value a.

[0013] As a further solution of the present invention: the verification process includes:

[0014] Obtain the time point t1 when the resistance greater than the preset value appears, and set the monitoring period [t1-(Δt / 2), t1+Δt], where Δt represents the preset monitoring period length;

[0015] Marking sampling points where resistance greater than a preset value appears as verification points, periodically obtaining the resistance at the verification points within the monitoring period, and sorting the resistances in timeline order to obtain a resistance ranking;

[0016] Calculate the resistance difference ΔR between the resistor at position i and the resistor at position i+1 in the resistor ranking i+1 =R i+1 -R i , the statistical resistance difference is greater than the ratio of the preset resistance difference threshold value A1;

[0017] Counting the proportion A2 of resistance greater than a preset value during the monitoring period;

[0018] When the ratio A1 is less than 0.3 and / or the ratio A2 is less than 0.3, the verification result is verification failure;

[0019] Except for the above cases, the verification result is passed.

[0020] As a further solution of the present invention: when the verification result is that the verification is passed, a prompt message is sent to the preset manager, indicating that there is an abnormality in the corresponding lake.

[0021] As a further solution of the present invention: when the verification result is verification passed, the method further includes:

[0022] Mark the verification points as outliers, and obtain the volume of the minimum circumscribed sphere containing all outliers;

[0023] The volume is acquired periodically, a volume sequence is generated, and the volume sequence is visualized.

[0024] As a further solution of the present invention: determining whether the water quality is abnormal includes:

[0025] Obtaining a functional relationship f(t) of the curve, where t represents time;

[0026] Calculating water quality scores , [ta, ts] represents the definition domain of the curve, and R represents a preset standard resistance;

[0027] When the water quality score is greater than a preset water quality score threshold K, it is determined that the water quality at the sampling point is abnormal.

[0028] As a further solution of the present invention: evaluating the wetland ecological environment status includes:

[0029] For a single lake, periodically obtain the locations of sampling points corresponding to abnormal water quality, and obtain the volume V of the minimum circumscribed sphere that includes all locations;

[0030] Calculate the average water quality score K1 of all sampling points in a single minimum circumscribed sphere;

[0031] Calculate the initial score ,η is the preset correction coefficient;

[0032] Calculating Assessment Scores , p j represents the initial score corresponding to the j-th lake, m represents the total number of lakes, and the mean is calculated as the evaluation score P.

[0033] As a further solution of the present invention, the evaluation of the wetland ecological environment status further includes:

[0034] Set evaluation score thresholds P1 and P2, with P1 < P2;

[0035] When the evaluation score P'<P1, it is determined that the wetland ecological environment state is poor;

[0036] When P1≤P'<P2, the wetland ecological environment is judged to be in general state;

[0037] When P2≤P', it is determined that the wetland ecological environment is in poor condition.

[0038] The beneficial effects of the present invention are as follows:

[0039] 1) The present invention sets up multi-point sampling at different depths and spatial locations, and combines it with a dynamic resistance verification mechanism to effectively identify transient conductivity fluctuations caused by biological activities such as waterfowl foraging and fish swimming. By analyzing the resistance sorting and difference ratio within the monitoring period, it can automatically determine whether it is a short-term anomaly caused by biological disturbance. If the verification fails, the data will be replaced with the average resistance of the normal period, thereby avoiding mistaking biological activity interference for pollution events. This mechanism not only ensures the continuity of monitoring data, but also significantly improves the accuracy of water quality assessment, prevents frequent false alarms, and ensures the scientific nature and effectiveness of ecological decision-making. At the same time, the minimum circumscribed sphere volume analysis of the spatial distribution of abnormal points is performed and the volume sequence is visually displayed, providing monitoring personnel with intuitive spatial disturbance information, further enhancing the multi-dimensional understanding of the ecological operation status of the wetland.

[0040] 2) This invention establishes an adaptive exception handling process that automatically sends an alarm to a preset management personnel after verification, enabling timely response to real-world anomalies. If verification fails, the anomaly point is marked and replaced with the average resistance of the normal measurement value within the replenishment period to ensure data integrity. If verification passes, combined with the periodic calculation and visualization of the volume sequence, monitoring personnel can clearly grasp the evolution trajectory of the anomaly and conduct targeted on-site inspections or interventions accordingly. In addition, by scoring water quality based on the functional relationship of the resistance change curve, this invention can accurately depict water quality fluctuation trends, providing a reliable basis for wetland maintenance, pollution tracing, and scientific remediation, thereby enhancing the comprehensive response capabilities of the monitoring system.

[0041] 3) This invention introduces a unified comprehensive ecological and environmental assessment model at the multi-lake level, integrating the spatial range of abnormal sample points of each lake with the water quality score to generate an overall wetland status assessment score, and realizes intelligent classification of ecological health status based on preset classification thresholds. This model takes into account the two key dimensions of spatial distribution and water quality intensity, combined with dynamic adjustment of correction coefficients to adapt to the needs of different wetland types and seasonal changes. By regularly updating the assessment scores and classification results of each lake, the monitoring system can intuitively reflect the spatiotemporal evolution of the wetland ecological environment, provide scientific support for management departments to accurately formulate protection, restoration and regulation plans, and promote the long-term sustainable management and optimization of wetlands. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below with reference to the accompanying drawings.

[0043] Figure 1 It is a flow chart of a wetland ecological environment monitoring method based on multi-dimensional data analysis of the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0045] See also Figure 1 As shown, the present invention is a wetland ecological environment monitoring method based on multi-dimensional data analysis, comprising the following steps:

[0046] Select several height values ​​within a preset height range, set sampling points in the wetland lake based on the height values, and obtain the water resistance at each sampling point in real time;

[0047] For example, a water depth range is first identified in the bathymetric map or sonar bathymetry results of a wetland lake, such as the range from shallow to deep water. Then, based on the overall span of this depth range, for example, from slightly shallow waters along the shore to deeper waters in the center of the lake, a number of specific depth values ​​are extracted as sampling layers at preset intervals (e.g., divided into several equal sections). Subsequently, during field deployment, representative points are selected at each depth value, such as near reeds on the shore, at the outer edge of aquatic weeds, or in the center of open water. The sample pile or sensor probe is precisely positioned to the predetermined depth using a buoy or tether casting method. The probe is usually equipped with a quadrupole electrode structure, which can be fixed in the water column by a cable, and the resistance measurement results are displayed in real time on the sensor instrument or data acquisition module. During the monitoring process, the system automatically records the resistance value of each probe at a set frequency (e.g., every few minutes or seconds), and uploads it synchronously to the background server via a wireless network or wired communication channel for subsequent time series analysis and visualization.

[0048] In a preferred embodiment of the present invention, setting the sampling points includes:

[0049] Set a height range, starting from the starting point of the height range, set a number of height values ​​at intervals of the preset height range, for a height value a, set n sampling points in a single lake, n is a preset number, and the distance from the sampling point to the bottom of the lake is the height value a;

[0050] For example, with the help of sonar sounding or existing bathymetric maps, the overall depth range of the lake from the nearshore shallows to the central deep water area is determined; then, based on the span of this depth range, it is evenly divided into several sections, such as from the edge of the reed bed to the center of the open water along the lake shoreline, and the representative depth value of each section is calculated in turn; for the depth value a (i.e., any depth), the staff will select multiple equally distributed points on the lake plan, such as near the water inlet, the center of the lake, and the water outlet, and then slowly lower the sampler with the electrode probe on the spot through a float or a tether until the vertical distance between the probe and the lake bottom is exactly equal to a; to ensure data stability, multiple samplers will be deployed at each point at the same time, and the buoy and counterweight above the probe can keep the vertical position of the probe in the water column from deviating;

[0051] When there is a resistance greater than a preset value, verification is performed based on the resistance change during the monitoring period to obtain a verification result, wherein the verification result includes verification pass or verification fail;

[0052] In another preferred embodiment of the present invention, the verification process includes:

[0053] Obtain the time point t1 when the resistance greater than the preset value appears, and set the monitoring period [t1-(Δt / 2), t1+Δt], where Δt represents the preset monitoring period length;

[0054] Marking sampling points where resistance greater than a preset value appears as verification points, periodically obtaining the resistance at the verification points within the monitoring period, and sorting the resistances in timeline order to obtain a resistance ranking;

[0055] Calculate the resistance difference ΔR between the resistor at position i and the resistor at position i+1 in the resistor ranking i+1 =R i+1 -R i , the statistical resistance difference is greater than the ratio of the preset resistance difference threshold value A1;

[0056] Counting the proportion A2 of resistance greater than a preset value during the monitoring period;

[0057] When the ratio A1 is less than 0.3 and / or the ratio A2 is less than 0.3, the verification result is verification failure;

[0058] Except for the above cases, the verification result is passed;

[0059] In actual operation, when the resistance value of a certain sampling point exceeds the preset threshold for the first time, the system automatically records this moment as t1 and defines the monitoring period as a continuous interval extending from t1 by Δt / 2 forward to Δt backward. The platform then marks the sampling point as a "verification point" and automatically captures the resistance reading of this point at a fixed frequency throughout the monitoring period. All readings are then arranged in chronological order, just like arranging photos one by one on a timeline. The differences between two adjacent measurements are then calculated, such as the difference ΔR6 between the fifth and sixth measurements. After calculating all adjacent differences within the entire period, the system calculates the proportion of these differences that exceed the preset difference threshold, A1, and the ratio, A2, of the number of times all resistance values ​​remain above the threshold to the total number of times within the period. Finally, if one or both of these ratios are below the judgment criteria, the system judges the verification as "failed," otherwise it is considered "passed."

[0060] It should be noted that the original intention of this design is to distinguish between occasional interference and real environmental problems, and to judge whether the sudden increase signal will quickly fall back or remain high by comparing the fluctuation amplitude and duration of the resistance reading. If it falls back to the normal range, it means that the high value at that time was more likely to be a short-term fluctuation caused by wild animals disturbing the sediment in the water; if it remains at a high value, it indicates that there may be real water quality changes. Through this verification process, false alarms caused by natural behaviors such as birds foraging or fish swimming can be avoided, while ensuring that any real environmental anomalies will not be masked, thereby providing more accurate and reliable decision support for wetland management;

[0061] When the verification result is failure, the resistance greater than the preset value is replaced by the average resistance within the preset replenishment period;

[0062] If the verification result is determined to be "failed," the system will first determine a subsequent replenishment period, such as a period of time extending from the time the verification point occurred. Then, during this replenishment period, the platform will collect resistance readings from the same sampling point again at a predetermined frequency. All these newly collected resistance values ​​are then aggregated and their arithmetic average is calculated. Finally, the original abnormal reading that exceeded the threshold is uniformly replaced with the calculated average resistance.

[0063] It is important to note that the method of replacing the average value within the supplementary period can smooth out short-term abnormal fluctuations caused by biological disturbances while maintaining the continuity and integrity of the data series. If these points are directly discarded or ignored, "gaps" will be left on the curve or cause deviations in subsequent analysis. Filling the gaps with the average results of actual measurements at the same location under similar conditions can maximize the retention of the true trend of water quality changes and avoid false alarms caused by occasional sediment disturbances. This makes subsequent time series analysis, water quality scoring, and spatial distribution assessment more reliable, and also provides a robust data foundation for wetland ecological status assessment.

[0064] In another preferred embodiment of the present invention, when the verification result is that the verification is passed, a prompt message is sent to a preset manager, indicating that an abnormality exists in the corresponding lake;

[0065] When the system determines that a sampling point has abnormal resistance and passes verification, the backend automatically integrates the abnormal event with key information such as the name of the corresponding lake, sampling point number, and occurrence time into a preset notification template in the data center. It then uses SMS, email, or mobile push interfaces to send the formatted alarm information to the communication devices of the maintenance team or management personnel. For example, immediately after the abnormal period ends, the system will generate a text message containing the fields "lake name - sampling point number - abnormal start time - resistance continuous abnormality" and send it through the operator gateway or the company's intranet email server.

[0066] By proactively issuing alerts after verification, this invention can quickly communicate real environmental changes to relevant personnel, ensuring that any genuine water quality or ecological anomalies are not missed. Timely push notifications allow managers to respond quickly, arrange on-site inspections, or implement emergency measures to prevent the problem from escalating or worsening. At the same time, this mechanism eliminates false alarms caused by biological interference, focusing attention on areas that truly require attention, improving the efficiency of emergency resource utilization, and providing strong support for subsequent ecological restoration and management decisions, further promoting the sustainable and healthy development of the wetland ecosystem.

[0067] In a preferred embodiment of this embodiment, when the verification result is verification passed, the process further includes:

[0068] Mark the verification points as outliers, and obtain the volume of the minimum circumscribed sphere containing all outliers;

[0069] Periodically obtain volume, generate volume sequences, and visualize the volume sequences;

[0070] After the verification is determined to be passed, the platform automatically marks the verification point and its corresponding latitude and longitude or lake plane coordinates as an outlier. It then calls the minimum circumscribed sphere algorithm for the three-dimensional point set to calculate the coordinates of all marked outlier points, determine the minimum sphere radius covering these points, and calculate the volume of the sphere accordingly. The system then continuously re-retrieves the latest outlier point set at preset time intervals, repeatedly performs the minimum circumscribed sphere calculation, and stores the sphere volumes obtained each time in the time series database. In the front-end visualization module, these volume values ​​are presented in the form of a broken line or bar chart along with the corresponding timestamp. Users only need to select the time range on the interactive interface to intuitively see the changing trend of the spatial range of the outlier area over time.

[0071] Minimum circumscribed sphere volume analysis can intuitively reflect the spatial spread or contraction of abnormal water quality or disturbance ranges. When the volume expands, it means the abnormal impact area is spreading, while when the volume contracts, it may indicate that intervention or restoration measures are taking effect. This combination of spatial quantitative indicators and time series not only elevates monitoring results from single-point data to area analysis, but also helps managers quickly grasp the geographic coverage of abnormal events, thereby accurately dispatching inspection forces and governance resources, ultimately providing solid spatial decision-making support for timely response and scientific management of wetland ecological environments.

[0072] Draw a curve showing the change in electrical resistance of the water body at the sampling point over time, determine whether the water quality at the sampling point is abnormal based on the curve, and evaluate the wetland ecological environment status based on the spatiotemporal information of the abnormal water quality;

[0073] In a preferred embodiment of the present invention, determining whether water quality is abnormal includes:

[0074] Obtaining a functional relationship f(t) of the curve, where t represents time;

[0075] Calculating water quality scores , [ta, ts] represents the definition domain of the curve, and R represents a preset standard resistance;

[0076] When the water quality score is greater than a preset water quality score threshold K, it is determined that the water quality at the sampling point is abnormal;

[0077] It should be noted that the discrete resistance values ​​recorded at each sampling point during the monitoring period are first converted into a continuous function f(t) through a curve fitting method (such as spline interpolation or polynomial regression), with the start time ta and end time ts as the curve definition domain. The platform then samples f(t) at a set step size within the time interval and calculates the absolute value of the difference between f(t) and the preset standard resistance R at each time point. These differences are then accumulated or integrated, and the cumulative result is finally divided by the time span (ts–ta) to generate a water quality score representing the degree of resistance deviation during the period. When this score exceeds a preset threshold, the water quality at the sampling point is deemed abnormal.

[0078] Condensing complex temporal fluctuations into a comparable quantitative indicator can not only reflect the overall deviation between the resistance curve and the standard value, but also avoid the interference of single extreme values ​​in judgment. Through a unified water quality score, horizontal comparison and vertical trend tracking at different depths and locations can be achieved, and standardized input can be provided for subsequent ecological assessment models, thus supporting objective evaluation of wetland water quality and scientific decision-making.

[0079] In a preferred embodiment of this invention, the assessment of the wetland ecological environment status includes:

[0080] For a single lake, periodically obtain the locations of sampling points corresponding to abnormal water quality, and obtain the volume V of the minimum circumscribed sphere that includes all locations;

[0081] Calculate the average water quality score K1 of all sampling points in a single minimum circumscribed sphere;

[0082] Calculate the initial score ,η is the preset correction coefficient;

[0083] Calculating Assessment Scores , p j represents the initial score corresponding to the jth lake, m represents the total number of lakes, and the mean is calculated as the evaluation score P;

[0084] It can be understood that in each evaluation cycle, all sampling points that are judged to have abnormal water quality are automatically retrieved and their coordinate information on the lake plan is read. For example, a point is located in the reed belt area on the left side of the water inlet, and another point is located in the shallows near the water outlet. Subsequently, the background integrates these three-dimensional coordinates and calls the minimum circumscribed sphere algorithm. The convex hull of the point set is first calculated and then the minimum sphere radius that can contain all abnormal points is determined. The volume V of the circumscribed sphere is calculated based on the radius and recorded. Next, the system queries the water quality score records corresponding to these abnormal points in the same cycle, accumulates all the score values ​​and divides them by the number of abnormal points to obtain the average score K1 within the circumscribed sphere area. The average water quality score is compared with the impact range. By multiplying the volume and the correction coefficient η, the degree of water quality deviation and the affected spatial range can be superimposed, thereby taking into account both the risk brought by the pollution intensity and the breadth of the affected area; at the same time, η is also used to correct the impact of V on the initial score, which can not only prevent the deviation caused by the decision-making based solely on the score or volume, but also balance the relationship between the intensity of water quality changes and the affected area, so that the initial score can better reflect the overall risk magnitude; the evaluation score p' reflects the quantitative score of the wetland ecological environment when the minimum circumscribed sphere is obtained in a single time, and the calculated mean reflects the average degree, which is more robust than a single evaluation; spatiotemporal information refers to time and space information, time is reflected by the mean, and space is reflected by the minimum circumscribed sphere;

[0085] This assessment process organically combines the spatial distribution of single-point water quality anomalies with their intensity, not only quantifying the spread of anomalies within a lake but also grading lake ecological risks based on intensity differences. By introducing correction coefficients, the method accounts for differences in lake size and topographical characteristics, ensuring comparability of scoring results. This integrated spatial-intensity assessment model helps managers intuitively determine which lakes are most in need of governance or focused monitoring, thereby enabling precise allocation and tiered management of wetland resources, ultimately enhancing the scientific nature and pertinence of overall ecological and environmental protection and decision-making.

[0086] It's important to note that correction factors can be used to compensate for differences in specific ecological characteristics or monitoring parameters across lakes, ensuring comparability and fairness across diverse wetland types. For example, consider two typical wetland water bodies, Lake A and Lake B. Lake A, located in a low-lying area at the foot of a mountain, is rich in humus and organic acids released by plant roots year-round, resulting in a slightly lower background resistivity. Lake B, on the other hand, is an open, shallow depression with strong sunlight and high water temperatures, resulting in consistently high ion concentrations in the water and a higher background resistivity. Without distinguishing between these two, the same resistance deviation in the two lakes would receive the same weight in the scoring. In reality, the same deviation in Lake A indicates a more serious pollution risk, while in Lake B, it may simply be a fluctuation in the natural background. In this case, a larger correction factor could be set for Lake A, so that the final initial score reflects a higher risk given the same deviation; a smaller correction factor could be set for Lake B to reduce oversensitivity to the naturally high resistivity background.

[0087] In another preferred embodiment of this embodiment, the assessment of the wetland ecological environment status further includes:

[0088] Set evaluation score thresholds P1 and P2, with P1 < P2;

[0089] When the evaluation score P'<P1, it is determined that the wetland ecological environment is in good condition;

[0090] When P1≤P'<P2, the wetland ecological environment is judged to be in general state;

[0091] When P2≤P', it is determined that the wetland ecological environment is in poor condition.

[0092] The larger the volume V, the wider the affected area, the larger the water quality score, and the higher the average severity of the sampling point. The increase in volume V or K1 will lead to an increase in the initial score, and then an increase in the evaluation score;

[0093] Therefore, the larger the assessment score, the wider the affected area or the higher the average severity of the sampling points, both of which reflect that the ecological environment of the wetland is worse. Therefore, the larger the assessment score, the worse the ecological environment of the wetland, and the relationship between the assessment score and the assessment score threshold is set accordingly.

[0094] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the latest real situation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0095] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A wetland ecological environment monitoring method based on multi-dimensional data analysis, characterized in that: The following steps are involved: Select several height values ​​within a preset height range, set sampling points in the wetland lake based on the height values, and obtain the water resistance at each sampling point in real time; When there is a resistance greater than a preset value, verification is performed based on the resistance change during the monitoring period to obtain a verification result, wherein the verification result includes verification pass or verification fail; When the verification result is failure, the resistance greater than the preset value is replaced by the average resistance within the preset replenishment period; Draw a curve showing the change in electrical resistance of the water body at the sampling point over time, determine whether the water quality at the sampling point is abnormal based on the curve, and evaluate the wetland ecological environment status based on the spatiotemporal information of the abnormal water quality; The verification process includes: Obtain the time point t1 when the resistance greater than the preset value appears, and set the monitoring period [t1-(Δt / 2), t1+Δt], where Δt represents the preset monitoring period length; Marking sampling points where resistance greater than a preset value appears as verification points, periodically obtaining the resistance at the verification points within the monitoring period, and sorting the resistances in timeline order to obtain a resistance ranking; Calculate the resistance difference ΔR between the resistor at position i and the resistor at position i+1 in the resistor ranking i+1 =R i+1 -R i , the statistical resistance difference is greater than the ratio of the preset resistance difference threshold value A1; Counting the proportion A2 of resistance greater than a preset value during the monitoring period; When the ratio A1 is less than 0.3 and / or the ratio A2 is less than 0.3, the verification result is verification failure; Except for the above cases, the verification result is passed; Determining whether there is abnormality in water quality includes: Obtaining a functional relationship f(t) of the curve, where t represents time; Calculating water quality scores , [ta, ts] represents the definition domain of the curve, and R represents a preset standard resistance; When the water quality score is greater than a preset water quality score threshold K, it is determined that the water quality at the sampling point is abnormal.

2. A wetland ecological environment monitoring method based on multi-dimensional data analysis according to claim 1, characterized in that: Setting the sampling points includes: Set a height range. Starting from the starting point of the height range, set several height values ​​at intervals of the preset height range. For height value a, set n sampling points in a single lake, where n is the preset number and the distance from the sampling point to the bottom of the lake is height value a.

3. The wetland ecological environment monitoring method based on multi-dimensional data analysis according to claim 1 is characterized in that: When the verification result is that the verification is passed, a prompt message is sent to the preset administrator, indicating that there is an abnormality in the corresponding lake.

4. The wetland ecological environment monitoring method based on multi-dimensional data analysis according to claim 1 is characterized in that: When the verification result is verification passed, the method further includes: Mark the verification points as outliers, and obtain the volume of the minimum circumscribed sphere containing all outliers; The volume is acquired periodically, a volume sequence is generated, and the volume sequence is visualized.

5. The wetland ecological environment monitoring method based on multi-dimensional data analysis according to claim 1 is characterized in that: Assessment of wetland ecological environment status includes: For a single lake, periodically obtain the locations of sampling points corresponding to abnormal water quality, and obtain the volume V of the minimum circumscribed sphere that includes all locations; Calculate the average water quality score K1 of all sampling points in a single minimum circumscribed sphere; Calculate the initial score ,η is the preset correction coefficient; Calculating Assessment Scores , p j represents the initial score corresponding to the j-th lake, m represents the total number of lakes, and the mean is calculated as the evaluation score P.

6. The wetland ecological environment monitoring method based on multi-dimensional data analysis according to claim 1 is characterized in that: The assessment of wetland ecological environment status also includes: Set evaluation score thresholds P1 and P2, with P1 < P2; When the evaluation score P'<P1, it is determined that the wetland ecological environment is in good condition; When P1≤P'<P2, the wetland ecological environment is judged to be in general state; When P2≤P', it is determined that the wetland ecological environment is in poor condition.

Images