Basin landscape pattern change and water quality effect analysis method

Abstract

The invention discloses a watershed landscape pattern change and water quality effect analysis method, which belongs to the technical field of landscape ecology, and comprises the following steps: step 1, data collection and preprocessing; step 2, landscape pattern analysis; 3, water quality analysis; 4, analyzing the correlation between the landscape pattern and the water quality; and 5, formulating a targeted drainage basin management strategy according to an analysis result. According to the method, a plurality of technical means such as a geographic information system (GIS), a remote sensing technology and statistical analysis are combined, the landscape pattern change and the water quality effect of the watershed can be comprehensively and systematically analyzed, and the comprehensive analysis method is beneficial to deeply understanding the complexity and the dynamic nature of the watershed ecosystem; and powerful support is provided for formulating a scientific and reasonable drainage basin management strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of landscape ecology, and specifically to a method for analyzing the changes in the landscape pattern of a basin and the water quality effect. Background Art

[0002] The landscape pattern controls various biogeochemical and physical processes in the basin, and can interfere with the migration and transformation of nutrients by changing the hydrological cycle processes such as rainfall, evapotranspiration, and runoff, thereby affecting the quality of the surface water environment. Especially under the dual influence of climate change and human activities, the profound impact of the landscape pattern on water quality has attracted extensive attention. With the rapid development of modern information technology, experts and scholars have conducted a large number of studies on the relationship between landscape patterns and water quality at various scales using different methods, analyzed the impact of the landscape pattern of the basin on water quality changes, and the research results have provided a scientific basis for the optimal allocation of regional land resources and water resource protection.

[0003] In the current research on the impact of landscape pattern on water quality changes, on the one hand, restricted by the multicollinearity of landscape pattern indices and the small number of observation samples, the interpretation effect of numerical analysis is poor. On the other hand, most studies are based on different spatial scale division methods, using the measured water quality data of monitoring stations as the response variable and the landscape pattern index as the explanatory variable. Therefore, the results only explain the degree of the impact of the landscape pattern on water quality changes, and do not well explain the direction of water quality changes, that is, it is not clear whether the landscape patterns of which landscape types improve or deteriorate the water quality at an appropriate scale. Therefore, how to comprehensively and systematically analyze the changes in the landscape pattern of the basin and its water quality effect has become a difficult problem that needs to be solved urgently.

[0004] Therefore, the technical personnel in this field have provided a method for analyzing the changes in the landscape pattern of a basin and the water quality effect to solve the problems raised in the above background art. Summary of the Invention

[0005] Aiming at the defects in the prior art, the present invention provides a method for analyzing the changes in the landscape pattern of a basin and the water quality effect, including the following steps:

[0006] Step 1: Data collection and preprocessing, specifically:

[0007] S101: Collect basic geographic information data covering the terrain and landform, water system distribution, and soil type of the basin;

[0008] S102: Obtain multi-phase remote sensing image data for analyzing the changes in the landscape pattern, and classify and interpret the land use types in combination with aerial photography and field surveys;

[0009] S103: Collect the water quality monitoring data of dissolved oxygen, chemical oxygen demand, ammonia nitrogen, and total phosphorus at each monitoring site in the basin at different times, and record the sampling time and location information simultaneously;

[0010] S104: Perform preprocessing operations such as radiometric correction and geometric correction on the remote sensing images, and conduct outlier tests and processing on the water quality data;

[0011] Step Two: Landscape pattern analysis, specifically,

[0012] S201: Use remote sensing image interpretation and geographic information system technology to divide the basin landscape into different types such as forest land, grassland, cultivated land, water area, and construction land;

[0013] S202: Calculate landscape pattern indices including patch number, patch density, average patch area, landscape fragmentation degree, diversity index, and aggregation index to quantitatively describe the landscape pattern characteristics;

[0014] S203: By comparing the landscape pattern indices at different times, analyze the change trend of the landscape pattern over time, and use the spatial analysis function of the geographic information system to display the spatial distribution characteristics of the landscape pattern change;

[0015] Step Three: Water quality analysis, specifically,

[0016] S301: Conduct descriptive statistical analysis on the water quality monitoring data, and calculate statistical parameters such as the mean, standard deviation, maximum value, and minimum value of each water quality index;

[0017] S302: Evaluate the water quality of the basin using one or more of the single-factor evaluation method, comprehensive pollution index method, and water quality category evaluation method;

[0018] S303: Draw curves of the water quality index changing over time, analyze the annual and seasonal change laws, and use the spatial interpolation function of the geographic information system to draw the water quality spatial distribution map;

[0019] Step Four: Correlation analysis between landscape pattern and water quality, specifically,

[0020] S401: Standardize the landscape pattern index and water quality index data;

[0021] S402: Use one or more methods such as Pearson correlation analysis and Spearman rank correlation analysis to analyze the correlation between the landscape pattern index and the water quality index;

[0022] S403: Select appropriate landscape pattern indices as independent variables and water quality indices as dependent variables, and construct one or more of linear regression models and non-linear regression models to quantitatively analyze the impact degree of landscape pattern change on water quality;

[0023] S404: Validate and evaluate the accuracy of the regression model using cross - validation and the leave - one - out method;

[0024] Step Five: According to the analysis results, formulate targeted watershed management strategies.

[0025] As a further solution of the present invention: In the said Step One, the specific process of performing outlier test and processing on water quality data is as follows:

[0026] Arrange the data in ascending order, calculate the first quartile and mark it as Q1, calculate the third quartile and mark it as Q3;

[0027] Calculate the target reference value IQR = Q3 - Q1, and consider the values lower than (Q1 - 1.5 * IQR) or higher than (Q3 + 1.5 * IQR) as outliers;

[0028] Mark the monitoring stations corresponding to the outliers as target stations, and extract the water quality data of the adjacent upstream and downstream monitoring stations of the target stations;

[0029] If the differences between the water quality data of the two extracted monitoring stations and the outliers are both within the preset range, then retain the outliers. If any one of the differences between the water quality data of the two extracted monitoring stations and the outliers is not within the preset range, then delete the outliers and mark the monitoring stations corresponding to the outliers as abnormal stations;

[0030] Formulate a personnel scheduling plan and inspection routes for all abnormal stations.

[0031] As a still further solution of the present invention: The specific formulation process of the said personnel scheduling plan and inspection routes is as follows:

[0032] Mark all on - duty maintenance personnel as Ai, i = 1···k, where k is a positive integer, and then mark the number of all abnormal stations as B, and calculate the average maintenance quantity b_avg=(B / k);

[0033] Compare the average maintenance quantity b_avg with the preset value b_preset. If b_avg ≤ b_preset, then do not transfer the resting maintenance personnel. If b_avg > b_preset, then transfer one resting maintenance personnel to work and return to the previous step;

[0034] Mark the total number of maintenance personnel finally as a. Among them, if the average of b < 1, then let a = B; then connect all the abnormal sites on the map along the basin in sequence to generate a target line segment, and divide the target line segment into a sub-line segments connected end to end; if (B / a) is an integer, each sub-line segment includes (B / a) abnormal sites; if (B / a) is not an integer, the first (a - 1) sub-line segments include N1 abnormal sites, and the last sub-line segment includes N2 abnormal sites, where N1 is the quotient of (B / a) and N2 is the remainder of (B / a).

[0035] Sort each sub-line segment in descending order of its length to generate a first list, and then sort each maintenance personnel in ascending order of their monthly contribution degree to generate a second list;

[0036] Allocate the sub-line segments in the first list to the maintenance personnel in the second list one by one in order, and the allocated sub-line segments are the maintenance routes of the corresponding maintenance personnel.

[0037] As a further solution of the present invention: The specific process of transferring the resting maintenance personnel to work is as follows:

[0038] Mark all the resting maintenance personnel as Ci, i = 1···n, where n is a positive integer;

[0039] Mark the duration from the end of work to the current time point of maintenance personnel Ci as T1, and then mark the preset first duration as t1;

[0040] Mark the total duration of maintenance personnel Ci's external maintenance this month as T2, and then mark the preset second duration as t2;

[0041] Calculate the weight value V of maintenance personnel Ci = 60%*(T1 / t1) + 40%*(t2 / T2);

[0042] Take the maintenance personnel Ci with the largest weight value V as the transferred maintenance personnel.

[0043] As a further solution of the present invention: The specific calculation process of the contribution degree is as follows:

[0044] Mark each maintenance personnel going to repair abnormal sites as Di, i = 1···a, where a is a positive integer;

[0045] Extract the number of times maintenance personnel Di go out for maintenance this month and mark it as d1, extract the number of times Di works overtime on duty this month and mark it as d2, extract the number of abnormal sites repaired by maintenance personnel Di this month and mark it as d3, extract the number of successfully repaired abnormal sites by maintenance personnel Di this month and mark it as d4, and extract the total duration of maintenance personnel Di's overtime on duty this month and mark it as d5;

[0046] Taking d1, d2, d3, d4, d5 as variables, a radar chart P is generated. Then, the area of the radar chart P is calculated and marked as Q, and the area Q is output as the contribution degree of the maintenance personnel Di. Among them, the larger the area Q, the greater the contribution degree of the corresponding maintenance personnel Di.

[0047] As a further solution of the present invention: in the first step, the basic geographic information data is sourced from a geographic information system database, a geological survey department, or a relevant government agency; the remote sensing image data comes from Landsat and Sentinel satellite images.

[0048] As a further solution of the present invention: in the second step, when displaying the spatial distribution characteristics of landscape pattern changes, the buffer analysis and overlay analysis functions of the geographic information system are used to further explore the spatial correlation between landscape pattern changes and water systems and topographic elements.

[0049] As a further solution of the present invention: in the third step, when using the spatial interpolation function of the geographic information system to draw the water quality spatial distribution map, one or more of the inverse distance weighted interpolation method and the Kriging interpolation method are used for interpolation calculation.

[0050] As a further solution of the present invention: in the fourth step, when constructing the regression model, a stepwise regression algorithm is introduced to screen the independent variables to optimize the regression model and improve the model's explanatory ability for the relationship between landscape pattern and water quality.

[0051] The beneficial effects of the present invention are reflected in:

[0052] This application combines various technical means such as geographic information system (GIS), remote sensing technology, and statistical analysis, and can comprehensively and systematically analyze the changes in the landscape pattern of the basin and its water quality effects. This comprehensive analysis method helps to more deeply understand the complexity and dynamics of the basin ecosystem and provides strong support for formulating scientific and reasonable basin management strategies. This application can deeply understand the impact of changes in the landscape pattern of the basin on water quality, and then formulate effective management strategies to improve and protect the water quality and ecological environment of the basin. This helps to promote the sustainable development of the basin and achieve the coordinated development of economy, society, and environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0054] Figure 1It is a flowchart of a method for analyzing the change of watershed landscape pattern and its water quality effect. Detailed implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0056] As mentioned in the background art of this application, through research, it is found that in the existing research on the impact of landscape pattern on water quality change, on the one hand, restricted by the multicollinearity of landscape pattern indices and the small number of observation samples, the interpretation effect of numerical analysis is poor. On the other hand, most studies are based on different spatial scale division methods, using the measured water quality data of monitoring stations as the response variable and the landscape pattern indices as the explanatory variables. Therefore, the results only explain the degree of the impact of landscape pattern on water quality change, and do not well explain the direction of water quality change, that is, it is not clear whether the landscape pattern of which landscape types improves or deteriorates the water quality at an appropriate scale. Generally speaking, the existing technology cannot comprehensively and systematically analyze the change of watershed landscape pattern and its water quality effect, and there are certain defects.

[0057] To solve the above defects, this application discloses a method for analyzing the change of watershed landscape pattern and its water quality effect, which combines various technical means such as geographic information system (GIS), remote sensing technology, and statistical analysis, and can comprehensively and systematically analyze the change of watershed landscape pattern and its water quality effect. This comprehensive analysis method helps to more deeply understand the complexity and dynamics of the watershed ecosystem and provides strong support for formulating scientific and reasonable watershed management strategies.

[0058] The following will introduce in detail how the solution of this application solves the above technical problems in conjunction with the accompanying drawings.

[0059] Please refer to Figure 1 , in the embodiments of the present invention, a method for analyzing the change of watershed landscape pattern and its water quality effect includes the following steps:

[0060] Step 1: Data collection and preprocessing. Specifically, S101: Collect basic geographic information data covering the terrain and geomorphology, water system distribution, and soil types of the basin; S102: Obtain multi-temporal remote sensing image data for landscape pattern change analysis, and classify and interpret land use types in combination with aerial photography and field surveys; S103: Collect water quality monitoring data such as dissolved oxygen, chemical oxygen demand, ammonia nitrogen, and total phosphorus at different monitoring stations in the basin, and record the sampling time and location information at the same time; S104: Perform preprocessing operations such as radiometric correction (Radiometric correction refers to the processing of remote sensing image data to eliminate the errors of the sensor itself and the influence of atmospheric conditions, so that it can more accurately reflect the true radiation characteristics of the surface) and geometric correction (Geometric correction is to eliminate the geometric distortion in the remote sensing image so that it can more accurately reflect the true shape and position of the surface) on the remote sensing images, and conduct outlier tests and processing on the water quality data;

[0061] Step 2: Landscape pattern analysis. Specifically, S201: Use remote sensing image interpretation and geographic information system technology to divide the basin landscape into different types such as forest land, grassland, cultivated land, water area, and construction land; S202: Calculate landscape pattern indices including patch number, patch density, average patch area, landscape fragmentation degree, diversity index, and aggregation index to quantitatively describe the characteristics of the landscape pattern; S203: Analyze the change trend of the landscape pattern over time by comparing the landscape pattern indices in different periods, and use the spatial analysis function of the geographic information system to display the spatial distribution characteristics of the landscape pattern change;

[0062] Step 3: Water quality analysis. Specifically, S301: Conduct descriptive statistical analysis on the water quality monitoring data, and calculate statistical parameters such as the mean, standard deviation, maximum value, and minimum value of each water quality index; S302: Use one or more of the single-factor evaluation method (The single-factor evaluation method is a method for evaluating water quality based on a single water quality parameter. It determines whether the water quality meets the standard by comparing the monitored concentration of each water quality parameter with the corresponding water quality standard), the comprehensive pollution index method (The comprehensive pollution index method is a method for comprehensively evaluating water quality based on multiple water quality parameters. It calculates the relative pollution index of each water quality parameter and obtains the comprehensive pollution index to reflect the overall pollution degree of the water body), and the water quality category evaluation method (The water quality category evaluation method is a method for classifying water quality according to the comprehensive evaluation results of water quality parameters. It usually classifies water quality based on the results of the single-factor evaluation method or the comprehensive pollution index method, combined with water quality standards) to evaluate the water quality of the basin; S303: Draw curves of water quality indices changing over time, analyze the annual and seasonal variation laws, and use the spatial interpolation function of the geographic information system to draw the water quality spatial distribution map;

[0063] Step 4: Correlation analysis between landscape pattern and water quality. Specifically, S401: Standardize the landscape pattern index and water quality index data; S402: Use one or more of Pearson correlation analysis (Pearson correlation analysis is a statistical method for evaluating the linear correlation between two continuous variables. It calculates the correlation coefficient based on the covariance and standard deviation of the sample data, and the value of this coefficient ranges from -1 to 1), Spearman rank correlation analysis (Spearman rank correlation analysis is a statistical method for evaluating the rank correlation between two variables. It calculates the correlation coefficient based on the ranks of the sample data (i.e., the positions of the data after sorting), and the value of this coefficient also ranges from -1 to 1) to analyze the correlation between the landscape pattern index and the water quality index; S403: Select appropriate landscape pattern indices as independent variables and water quality indices as dependent variables, and construct one or more of linear regression models and non-linear regression models to quantitatively analyze the impact of landscape pattern changes on water quality; S404: Use methods such as cross-validation (Cross-validation is a commonly used method for evaluating the performance of a model. Its core idea is to divide the dataset into multiple subsets, and then train the model multiple times. Each time, a different subset is used as the test set, and the remaining subsets are used as the training set, so as to more comprehensively evaluate the generalization ability of the model. Among them, K-fold cross-validation is the most commonly used cross-validation method), leave-one-out method (Leave-one-out cross-validation is an extreme form of K-fold cross-validation, where the value of K is equal to the number of samples in the dataset. That is, in each iteration, n-1 samples are used to train the model, and the remaining one sample is used for testing) to verify and evaluate the accuracy of the regression model;

[0064] Step 5: According to the analysis results, formulate targeted watershed management strategies. This application combines various technical means such as geographic information system (GIS), remote sensing technology, and statistical analysis, and can comprehensively and systematically analyze the changes in the watershed landscape pattern and its water quality effects. This comprehensive analysis method helps to more deeply understand the complexity and dynamics of the watershed ecosystem, and provides strong support for formulating scientific and reasonable watershed management strategies. This application can deeply understand the impact of changes in the watershed landscape pattern on water quality, and then formulate effective management strategies to improve and protect the water quality and ecological environment of the watershed. This helps to promote the sustainable development of the watershed and achieve the coordinated development of economy, society, and environment.

[0065] In this embodiment, in step one, the specific process of performing outlier detection and processing on water quality data is as follows: The data is sorted in ascending order, the first quartile is calculated and marked as Q1, and the third quartile is calculated and marked as Q3; the target reference value IQR = Q3 - Q1 is calculated, and values lower than (Q1 - 1.5 * IQR) or higher than (Q3 + 1.5 * IQR) are regarded as outliers; the monitoring stations corresponding to the outliers are marked as target stations, and the water quality data of the adjacent upstream and downstream monitoring stations of the target stations are extracted; if the differences between the water quality data of the two extracted monitoring stations and the outliers are both within the preset range, the outliers are retained, and if any one of the differences between the water quality data of the two extracted monitoring stations and the outliers is not within the preset range, the outliers are deleted, and the monitoring stations corresponding to the outliers are marked as abnormal stations; a personnel scheduling plan and a maintenance route are formulated for all abnormal stations. This setting can effectively eliminate significantly abnormal water quality data, and retaining those relatively reasonable abnormal data can further improve the accuracy of data analysis.

[0066] In this embodiment, the specific process of formulating a personnel scheduling plan and a maintenance route is as follows: All on-duty maintenance personnel are marked as Ai, i = 1 ··· k, where k is a positive integer, and then the number of all abnormal stations is marked as B, and the average maintenance quantity b average = (B / k) is calculated; the average maintenance quantity b average is compared with the preset value b preset. If b average ≤ b preset, no off-duty maintenance personnel are transferred. If b average > b preset, one off-duty maintenance personnel is transferred to work and the previous step is returned; the total number of final maintenance personnel is marked as a, where if b average < 1, then a = B; then all abnormal stations are sequentially connected along the river basin on the map to generate a target line segment, and the target line segment is divided into a sub-line segments connected end to end; if (B / a) is an integer, each sub-line segment includes (B / a) abnormal stations; if (B / a) is not an integer, the first (a - 1) sub-line segments include N1 abnormal stations, and the last sub-line segment includes N2 abnormal stations, where N1 is the quotient of (B / a) and N2 is the remainder of (B / a); the sub-line segments are sorted in descending order of their lengths to generate a first list, and then the maintenance personnel are sorted in ascending order of their monthly contributions to generate a second list; the sub-line segments in the first list are sequentially assigned to the maintenance personnel in the second list, and the assigned sub-line segments are the maintenance routes of the corresponding maintenance personnel. This setting can effectively allocate suitable maintenance personnel to all abnormal stations, improve the maintenance efficiency, and ensure the fairness of work distribution to a certain extent.

[0067] In this embodiment, the specific process of transferring the maintenance personnel on rest to work is as follows: all the maintenance personnel on rest are marked as Ci, where i = 1···n, and n is a positive integer; the duration from the end of work to the current time point of the maintenance personnel Ci is marked as T1, and the preset first duration is marked as t1; the total duration of the maintenance personnel Ci for external maintenance this month is marked as T2, and the preset second duration is marked as t2; calculate the weight value V of the maintenance personnel Ci = 60%*(T1 / t1) + 40%*(t2 / T2); the maintenance personnel Ci with the maximum weight value V is taken as the maintenance personnel to be transferred. This setting ensures that the maintenance personnel with relatively low workload this month can be preferentially transferred, further improving the fairness of work assignment.

[0068] In this embodiment, the specific calculation process of the contribution degree is as follows: the maintenance personnel going to repair the abnormal sites are marked as Di, where i = 1···a, and a is a positive integer; extract the number of times the maintenance personnel Di goes out for maintenance this month and mark it as d1, extract the number of times Di works overtime on duty this month and mark it as d2, extract the number of abnormal sites repaired by the maintenance personnel Di this month and mark it as d3, extract the number of successfully repaired abnormal sites by the maintenance personnel Di this month and mark it as d4, extract the total duration of the maintenance personnel Di working overtime on duty this month and mark it as d5; use d1, d2, d3, d4, d5 as variables to generate a radar chart P, and then calculate the area of the radar chart P and mark it as Q, and output the area Q as the contribution degree of the maintenance personnel Di. The larger the area Q, the greater the contribution degree of the corresponding maintenance personnel Di. This setting ensures that on the premise that the maintenance workloads are not much different, the maintenance routes with shorter distances are preferentially assigned to the maintenance personnel with greater contribution degrees, further improving the fairness of work assignment.

[0069] In this embodiment, in step one, the basic geographic information data is sourced from the geographic information system database, geological survey department or relevant government agencies; the remote sensing image data comes from satellite images such as Landsat and Sentinel. The Landsat series of satellites is operated by the United States Geological Survey (USGS) and has continuously acquired satellite remote sensing data for many years globally since 1972. These data have very good inheritance and have become the most widely used data in the field of global resource and environmental surveys. The Sentinel series of satellites is part of the Copernicus program of the European Space Agency (ESA), which aims to use multi-source data (mainly remote sensing data) to obtain timely and high-quality information, services and knowledge.

[0070] In this embodiment, in step two, when displaying the spatial distribution characteristics of the landscape pattern changes, the buffer analysis and overlay analysis functions of the Geographic Information System (GIS) are used to further explore the spatial correlation between the landscape pattern changes and elements such as water systems and terrain.

[0071] In this embodiment, in step three, when using the spatial interpolation function of the Geographic Information System (GIS) to draw the spatial distribution map of water quality, one or more of the following interpolation methods are used for interpolation calculation: the Inverse Distance Weighting (IDW) interpolation method (the IDW interpolation method is an interpolation method that calculates the value of the interpolation point by weighting the values and distances of surrounding points. It assigns higher weights to closer data points and lower weights to farther data points, and the weights are proportional to the reciprocal of the distance), and the Kriging interpolation method (Kriging interpolation is a method of estimating the value of an unknown point by weighted summation of the data of all known points in space. It is based on the assumption of the homogeneity of spatial attributes and takes into account the autocorrelation of spatial data).

[0072] In this embodiment, in step four, when constructing the regression model, the stepwise regression algorithm is introduced to screen the independent variables in order to optimize the regression model and improve the model's explanatory ability for the relationship between landscape pattern and water quality. Introducing the stepwise regression algorithm to screen the independent variables when constructing the regression model is an effective optimization strategy, especially when studying complex problems such as the relationship between landscape pattern and water quality. The stepwise regression algorithm can automatically select the most important variables from a large number of candidate independent variables, thereby establishing a more concise and better-predicting regression model. The stepwise regression method is a feature screening and optimization method in the construction of a multiple linear regression model. It automatically screens out the independent variables (explanatory variables) that have a significant impact on the target variable (dependent variable) through forward selection, backward elimination, or bidirectional selection steps, in order to reduce the model complexity and improve the interpretability. The core principle of the stepwise regression method is to gradually optimize the variable combination in the model through iteration until the optimal variable combination is found.

[0073] The present invention combines various technical means such as Geographic Information System (GIS), remote sensing technology, and statistical analysis, and can comprehensively and systematically analyze the changes in the watershed landscape pattern and its water quality effects. This comprehensive analysis method helps to more deeply understand the complexity and dynamics of the watershed ecosystem, and provides strong support for formulating scientific and reasonable watershed management strategies. This application can deeply understand the impact of changes in the watershed landscape pattern on water quality, and then formulate effective management strategies to improve and protect the watershed water quality and ecological environment. This helps to promote the sustainable development of the watershed and achieve the coordinated development of economy, society, and environment.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the claims and the description of the present invention.

Claims

1. A method for analyzing changes in watershed landscape pattern and water quality effects, characterized in that: The following steps are involved: Step 1: Data collection and preprocessing, specifically, S101: Collect basic geographic information data covering basin topography, water system distribution, and soil types; S102: Obtain multiple periods of remote sensing image data for landscape pattern change analysis, and classify and interpret land use types in combination with aerial photography and field surveys; S103: Collect the water quality monitoring data of dissolved oxygen, chemical oxygen demand, ammonia nitrogen and total phosphorus at different monitoring stations in the basin at different times, and record the sampling time and location information; S104: Perform radiation correction and geometric correction preprocessing operations on remote sensing images, and perform outlier inspection and processing on water quality data; Step 2: Landscape pattern analysis, specifically, S201: Use remote sensing image interpretation and geographic information system technology to divide the watershed landscape into different types of forest land, grassland, cultivated land, water area and construction land; S202: Calculate landscape pattern indices including patch number, patch density, average patch area, landscape fragmentation, diversity index, and aggregation index to quantitatively describe landscape pattern characteristics; S203: By comparing the landscape pattern indexes of different periods, analyze the changing trend of landscape pattern over time, and use the spatial analysis function of geographic information system to display the spatial distribution characteristics of landscape pattern changes; Step 3: Water quality analysis, specifically, S301: Perform descriptive statistical analysis on water quality monitoring data and calculate the mean, standard deviation, maximum value, and minimum value statistical parameters of each water quality index; S302: Evaluate the water quality of the basin by using one or more of the single factor evaluation method, comprehensive pollution index method, and water quality category evaluation method; S303: Draw a curve of water quality index changing with time, analyze the annual and seasonal variation patterns, and draw a water quality spatial distribution map using the spatial interpolation function of the geographic information system; Step 4: Correlation analysis between landscape pattern and water quality, specifically, S401: Standardize the landscape pattern index and water quality index data; S402: Use one or more methods of Pearson correlation analysis and Spearman rank correlation analysis to analyze the correlation between landscape pattern index and water quality index; S403: Select appropriate landscape pattern index as independent variable and water quality index as dependent variable, and construct one or more linear regression models and nonlinear regression models to quantitatively analyze the impact of landscape pattern change on water quality; S404: Use cross validation and leave-one-out method to validate and evaluate the accuracy of the regression model; Step 5: Develop targeted watershed management strategies based on the analysis results.

2. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 1, characterized in that: In step 1, the specific process of outlier detection and processing of water quality data is as follows: Arrange the data in ascending order, calculate the first quartile and label it as Q1, calculate the third quartile and label it as Q3; Calculate the target reference value IQR = Q3-Q1, and consider values ​​lower than (Q1-1.5*IQR) or higher than (Q3+1.5*IQR) as outliers; The monitoring station corresponding to the outlier is marked as the target station, and the water quality data of the upstream and downstream monitoring stations adjacent to the target station are extracted; If the difference between the water quality data of the two monitoring stations extracted and the outlier is within the preset range, the outlier is retained; if any of the difference between the water quality data of the two monitoring stations extracted and the outlier is not within the preset range, the outlier is deleted and the monitoring station corresponding to the outlier is marked as an outlier station; Develop personnel scheduling plans and maintenance routes for all abnormal sites.

3. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 2, characterized in that: The specific process of formulating the personnel scheduling plan and maintenance route is as follows: Mark all on-duty maintenance personnel as Ai, i = 1...k, where k is a positive integer, and then mark the number of all abnormal sites as B, and calculate the average maintenance number baverage = (B / k); The average maintenance quantity b average is compared with the preset value b preset. If b average ≤ b preset, the maintenance personnel who are resting will not be transferred. If b average > b preset, a maintenance personnel who are resting will be transferred to work and return to the previous step. The final total number of maintenance personnel is marked as a, where if b is less than 1 on average, a=B; then all abnormal sites are connected in sequence along the watershed on the map to generate a target segment, and the target segment is divided into a sub-segments connected end to end; if (B / a) is an integer, each sub-segment includes (B / a) abnormal sites; if (B / a) is not an integer, the first (a-1) sub-segments include N1 abnormal sites, and the last sub-segment includes N2 abnormal sites, where N1 is the quotient of (B / a) and N2 is the remainder of (B / a); Sort the sub-line segments by length from large to small to generate the first list, and then sort the maintenance personnel by contribution this month from small to large to generate the second list; The sub-line segments in the first list are allocated one by one to the maintenance personnel in the second list in order, and the allocated sub-line segments are the maintenance routes of the corresponding maintenance personnel.

4. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 3, characterized in that: The specific process of transferring the maintenance personnel who are on rest to take up their posts is as follows: Mark all the maintenance personnel who are resting as Ci, i=1···n, where n is a positive integer; Mark the time from when the maintenance personnel Ci gets off work to the current time point as T1, and then mark the preset first time length as t1; The total time that maintenance personnel Ci is out for maintenance this month is marked as T2, and the preset second time is marked as t2; Calculate the weight value V of maintenance personnel Ci = 60%*(T1 / t1)+40%*(t2 / T2); The maintenance personnel Ci with the largest weight value V is selected as the maintenance personnel.

5. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 4, characterized in that: The specific calculation process of the contribution is as follows: Each maintenance personnel who goes to repair the abnormal site is marked as Di, i = 1...a, where a is a positive integer; Extract the number of times maintenance personnel Di went out for maintenance this month and marked it as d1, extract the number of times Di worked overtime this month and marked it as d2, extract the number of abnormal sites maintained by maintenance personnel Di this month and marked it as d3, extract the number of abnormal sites successfully repaired by maintenance personnel Di this month and marked it as d4, extract the total length of time maintenance personnel Di worked overtime this month and marked it as d5; Use d1, d2, d3, d4, and d5 as variables to generate a radar chart P, then calculate the area of ​​the radar chart P and mark it as Q, and output the area Q as the contribution of the maintenance personnel Di, wherein the larger the area Q is, the greater the contribution of the corresponding maintenance personnel Di is.

6. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 5, characterized in that: In the step 1, the basic geographic information data comes from the geographic information system database, geological survey departments or relevant government agencies; the remote sensing image data comes from Landsat and Sentinel satellite images.

7. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 6, characterized in that: In the second step, when displaying the spatial distribution characteristics of landscape pattern changes, the buffer zone analysis and overlay analysis functions of the geographic information system are used to further explore the spatial correlation between landscape pattern changes and water systems and terrain elements.

8. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 7, characterized in that: In the step three, when drawing the water quality spatial distribution map using the spatial interpolation function of the geographic information system, one or more methods including the inverse distance weighted interpolation method and the Kriging interpolation method are used for interpolation calculation.

9. A method for analyzing changes in watershed landscape pattern and water quality effects according to claim 8, characterized in that: In step 4, when constructing the regression model, a stepwise regression algorithm is introduced to screen independent variables to optimize the regression model and improve the model's ability to explain the relationship between landscape pattern and water quality.

Images