Method and system for monitoring soil degradation of dry-hot valley plain land

Abstract

This invention provides a method and system for monitoring soil degradation in plains of arid-hot river valleys, belonging to the field of soil degradation monitoring technology. The method involves dividing the area to be monitored into grids to obtain multiple monitoring sub-regions, and monitoring each sub-region separately. The steps include: periodically collecting multiple soil attribute data from the monitoring sub-regions to form a multi-soil attribute data vector at a unified time scale; identifying multiple soil degradation problems and establishing a mapping relationship matrix and coupling coefficients between the soil degradation problems; constructing a barrier degree model for each soil degradation problem; outputting initial risk assessment parameters for various soil degradation problems in the monitoring sub-regions of the current collection period through a risk evolution model; and performing soil degradation trend analysis based on the initial risk assessment parameters. The advantages of this invention are improved foresight, accuracy, and interpretability of soil degradation monitoring.

Description

Technical Field

[0001] This invention relates to the field of soil degradation monitoring technology, and more specifically, to a method and system for monitoring soil degradation in plains of arid and hot river valleys. Background Technology

[0002] The dry-hot river valley plains are the core geomorphic units with the best water, heat, soil, and topographic conditions within the basin, and also the core carriers for regional population agglomeration and large-scale agricultural development. Under long-term cultivation and natural evolution, the soil ecosystems of these plains exhibit various types of soil degradation. Therefore, to ensure the quality of cultivated conditions, it is necessary to monitor soil degradation.

[0003] Current monitoring methods for soil degradation largely rely on threshold judgments based on indicators, such as soil moisture, nutrients, and salinity. While these methods can reflect local degradation to some extent, they fail to comprehensively depict the degradation process under the combined effects of multiple factors. They also lack a detailed characterization of the mechanistic relationships between different data points and different soil degradation problems, resulting in coarse monitoring results that are difficult to pinpoint specific causes of degradation. More detailed monitoring methods for different soil degradation problems generally treat them as independent processes, monitoring them separately. This ignores the coupling relationships between different soil degradation problems and lacks a comprehensive analysis of the risks and evolution paths of different soil degradation problems, leading to insufficient foresight, accuracy, and interpretability of the monitoring results.

[0004] Therefore, there is an urgent need to improve the monitoring methods for soil degradation, integrate multiple soil property data and degradation mechanisms, and achieve coupled analysis of the risks of various soil degradation problems, thereby improving the foresight, accuracy and interpretability of soil degradation monitoring. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for monitoring soil degradation in plains of arid and hot river valleys, which can improve the foresight, accuracy and interpretability of soil degradation monitoring.

[0006] This invention is achieved through the following technical solution: A method for monitoring soil degradation in plains of arid-hot river valleys involves dividing the area to be monitored into grids to obtain multiple monitoring sub-regions, and monitoring each sub-region separately, including the following steps: Periodically collect and monitor multiple soil attribute data for sub-regions to form a data vector of multiple soil attributes at a unified time scale; To identify multiple soil degradation problems, a mapping matrix is ​​established between the various soil attribute data and the various soil degradation problems. The element in the i-th row and j-th column of the mapping matrix is... For the first The mapping coefficient between soil attribute data and soil degradation problem j represents the first type of soil degradation problem. The degree of influence of soil property data on the j-th type of soil degradation problem is determined, and a coupling coefficient is established between soil degradation problems. The coupling coefficient represents the degree of coupling effect between soil degradation problems. For each type of soil degradation problem, a barrier degree model is constructed. The input of the barrier degree model is a vector of multiple soil attribute data and a mapping relationship matrix, and the output is the contribution degree of degradation factor for each type of soil degradation problem in the current collection period. Based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, the initial risk assessment parameters of various soil degradation problems in the monitoring sub-region of the current collection period are output through the risk evolution model. Based on the risk evolution model output of each monitoring sub-region, soil degradation trend analysis is performed based on the initial risk assessment parameters, and the soil degradation problems and corresponding final risk assessment parameters of the monitoring sub-regions are output.

[0007] Preferably, the method for constructing the barrier degree model for each type of soil degradation problem is as follows: The various soil attribute data are standardized to obtain standardized multiple soil attribute data. The value range of the standardized multiple soil attribute data is [0,1] and all of them are positive indicators. Principal component analysis is performed on standardized multi-soil attribute data from N periods up to the current collection period. Based on the degree of variation of each soil attribute data in the time dimension and the principal component analysis results of the multi-soil attribute data from N periods, the first weight and the second weight of each soil attribute data are calculated respectively. The quality participation weight of the corresponding data is obtained by weighted summation of the first weight and the second weight. Based on the mapping matrix, the quality participation weights of each soil attribute data, and the standardized multiple soil attribute data, the degradation factor contribution of each soil degradation problem is calculated.

[0008] Preferably, the method for calculating the first weight based on the degree of variation of each soil attribute data in the time dimension is as follows:

[0009]

[0010]

[0011] in, For the first The first weight of soil attribute data, Representing the The degree of variation of soil property data over time. This represents the total number of data types for various soil properties. For the t-th collection period up to the current collection period, the first... Soil property data The standardized value of N is the total number of collection cycles for the various soil attribute data when performing the principal component analysis, and is also the number of the current collection cycle. The method for calculating the second weight based on the principal component analysis results of each of the multiple soil attribute data is as follows:

[0012]

[0013] in, For the first The second weight of soil attribute data, Let K be the eigenvalue of the k-th principal component, and K be the total number of principal components. Representing the The contribution of soil attribute data to the k-th principal component. It is the i-th element of the eigenvector of the k-th principal component.

[0014] Preferably, the method for calculating the contribution of the degradation factor to each soil degradation problem is as follows: Obtain the initial degradation factor contribution for each of the aforementioned soil degradation problems:

[0015] in, The initial degradation factor contribution for the j-th type of soil degradation problem in the current data collection period. For the first The quality of soil attribute data is used as a weighting factor; Normalization yields the contribution of each degradation factor to each of the aforementioned soil degradation problems:

[0016] Where L represents the total number of soil degradation problems. The contribution of the degradation factor to the j-th type of soil degradation problem in the current collection period.

[0017] Preferably, the method for establishing the risk evolution model is as follows: Based on the data vectors of various soil attributes in the current collection period and the contribution of degradation factors for each soil degradation problem, based on the contribution of degradation factors for the j-th soil degradation problem in the current collection period. Multi-source driving items and type coupling items Output the initial risk assessment parameters for the j-th type of soil degradation problem. , :

[0018]

[0019] ; Where L represents the total number of soil degradation problems. The contribution of the degradation factor to the j-th type of soil degradation problem in the current collection period, where N is the number of the current collection period. This represents the total number of data types for various soil properties. The current collection cycle is the [number]th ... Soil property data Standardized values, The number of the previous collection cycle in the current collection cycle Soil property data Standardized values, To prevent positive numbers with a denominator of 0, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The worsening of soil degradation problems has a significant impact on the first The degree of coupling effect of soil degradation problems, where max represents finding the maximum value. , and These are the weighting coefficients.

[0020] Preferably, the method for establishing the mapping relationship matrix is ​​as follows: Construct a soil degradation mechanism relationship diagram. The soil degradation mechanism relationship diagram includes index elements and directed edges. The index elements include all data of the various soil attribute data and all the soil degradation problems. If the change of the p-th index element will cause the change of the q-th index element, then establish a directed edge from the p-th index element to the q-th index element. The element in the i-th row and j-th column of the mapping relationship matrix is ​​calculated based on the soil degradation mechanism relationship diagram. The value:

[0021] in, The diagram showing the relationship between soil degradation mechanisms is shown in Figure 1. The minimum path between the index element corresponding to a certain type of soil attribute data and the index element corresponding to the j-th type of soil degradation problem. The path is defined as the number of directed edges connecting two index elements along a directed edge. The value range is greater than 0 and less than 1.

[0022] Preferably, the final risk assessment parameters are calculated as follows: The final risk assessment parameters for the j-th type of soil degradation problem in the c-th monitoring sub-region during the current data collection period. for: ; in, The initial risk assessment parameters represent the type u soil degradation problem in the c-th monitoring sub-region during the current data collection period. Let V be the spatially distributed influence parameter of the v-th type of soil degradation problem in the g-th monitoring sub-region on the u-th type of soil degradation problem in the c-th monitoring sub-region. The weights are values ​​greater than 0 and less than 1, and N is the number of the current collection period.

[0023] Preferably, the method for obtaining the influence parameters based on spatial distribution is as follows: Construct a spatial weight matrix among the monitored sub-regions, where the first... Line number Column elements The distance evaluation parameter between the c-th monitoring sub-region and the g-th monitoring sub-region is:

[0024] in, Let c be the center distance between the c-th monitoring sub-region and g-th monitoring sub-region. This is the distance attenuation coefficient; Calculate the :

[0025] in, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The degree of coupling effect of the exacerbation of the first type of soil degradation problem on the second type of soil degradation problem.

[0026] Preferably, the coupling coefficient is calculated as follows: A symbol matrix is ​​constructed based on the degradation mechanism. The element in the u-th row and v-th column of the symbol matrix... The value represents the effect of the worsening of the u-th type of soil degradation on the v-th type of soil degradation, with 1 representing promotion, -1 representing inhibition, and 0 representing no effect. Calculate the absolute value of the correlation coefficient between the exacerbation of type u soil degradation and the contribution of the degradation factors to type v soil degradation using historical samples. The selection criterion for historical samples is that the variance of the contribution of the degradation factor of the w-th soil degradation problem at each collection time in the historical samples is not greater than a preset variance threshold. , and ; Calculate the coupling coefficient between the exacerbation of the u-th type of soil degradation problem and the v-th type of soil degradation problem. : .

[0027] This invention also provides a soil degradation monitoring system for plains in arid and hot river valleys, which applies the above-mentioned soil degradation monitoring method for plains in arid and hot river valleys, including: The data acquisition module is used to periodically collect various soil property data of the monitored sub-regions; The degradation mechanism module is used to identify various soil degradation problems, establish a mapping relationship matrix between the various soil attribute data and the various soil degradation problems, and establish coupling coefficients between soil degradation problems; The obstacle degree model construction module is used to build obstacle degree models for each type of soil degradation problem. The initial assessment module is used to output initial risk assessment parameters for various soil degradation problems in the monitoring sub-region of the current collection period based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, through the risk evolution model. The final assessment module is used to analyze soil degradation trends based on the risk evolution model output of each monitoring sub-region and the initial risk assessment parameters, and output the soil degradation problems and corresponding final risk assessment parameters of the monitoring sub-region.

[0028] The technical solution of the present invention has at least the following advantages and beneficial effects: This invention improves the accuracy of the analysis area by dividing the monitoring sub-regions, making the monitoring results more targeted; This invention constructs a mapping matrix between various soil attribute data and various soil degradation problems, thereby achieving a refined expression of the degree of influence between different data and different soil degradation problems. This can more accurately reflect the differentiated contributions of various soil attribute data to various soil degradation problems, thus improving the accuracy of soil degradation identification. This invention establishes coupling coefficients between soil degradation problems and introduces the promoting or inhibiting relationships between different soil degradation problems into the risk analysis process based on the mechanism. This can characterize the interaction between soil degradation problems and further improve the reliability of monitoring. This invention constructs a risk evolution model by combining the coupling relationship between soil degradation problems. This enables the calculation of soil degradation risk not only to be based on the static evaluation of the current monitoring status, but also to reflect and predict the future dynamic evolution process based on the different monitoring results between monitoring sub-regions. This improves the dynamism and foresight of risk assessment and is beneficial for providing a basis for subsequent governance and decision-making. Attached Figure Description

[0029] Figure 1 This is a schematic flowchart of the method for monitoring soil degradation in flatlands of arid and hot river valleys provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the soil degradation monitoring system for dry-hot valley plains provided in Embodiment 2 of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0031] Example 1 This embodiment provides a method for monitoring soil degradation in plains of arid and hot river valleys. First, the area to be monitored is divided into grids to obtain multiple monitoring sub-regions. Monitoring is then conducted in each sub-region separately. (See reference...) Figure 1 This includes the following steps: Step S1: Periodically collect multiple soil attribute data of the monitoring sub-region to form a multiple soil attribute data vector at a unified time scale.

[0032] Step S2: Identify various soil degradation problems, establish a mapping relationship matrix between the various soil attribute data and the various soil degradation problems, and establish coupling coefficients between the soil degradation problems.

[0033] In the application examples of this embodiment, soil degradation problems include soil organic matter deficiency, insufficient nitrogen storage capacity, poor aggregate structure stability, weak soil nutrient retention capacity, and secondary salinization. Various soil property data may include soil water holding capacity, saturated hydraulic conductivity, geometric mean diameter of soil aggregates, soil organic matter and various nutrient contents, soil microbial biomass carbon and nitrogen, cation exchange capacity, and electrical conductivity, etc.

[0034] First, the element in the i-th row and j-th column of the mapping relation matrix For the first The mapping coefficient between soil attribute data and soil degradation problem j represents the first type of soil degradation problem. The method for establishing the mapping relationship matrix to determine the influence of soil attribute data on the degradation problem of the j-th type of soil is as follows: To establish the relationship between various soil attribute data and soil degradation problems, a soil degradation mechanism relationship graph is constructed. The soil degradation mechanism relationship graph includes index elements and directed edges. The index elements include all data of the various soil attribute data and all the soil degradation problems. If a change in the p-th index element will cause a change in the q-th index element, then a directed edge is established from the p-th index element to the q-th index element. The element in the i-th row and j-th column of the mapping relationship matrix is ​​calculated based on the soil degradation mechanism relationship diagram. The value:

[0035] in, The diagram showing the relationship between soil degradation mechanisms is shown in Figure 1. The minimum path between the index element corresponding to a certain type of soil attribute data and the index element corresponding to the j-th type of soil degradation problem. The path is defined as the number of directed edges connecting two index elements along a directed edge. The value range is greater than 0 and less than 1.

[0036] In establishing the mapping relationship matrix, by introducing a mechanism relationship diagram, the relationship between various soil attribute data and soil degradation problems is transformed from "empirical judgment" to "structured modeling". This not only distinguishes between direct and indirect influences, but also avoids the insufficient accuracy caused by simple weighting in traditional methods, thereby improving the scientificity and accuracy of soil degradation problem identification.

[0037] It is important to note that when constructing the soil degradation mechanism relationship graph, it is not simply about establishing direct correlations (directed edges) between multiple soil attribute data and soil degradation problems. Instead, all indicator elements are uniformly incorporated into the same directed graph structure to establish directed edges. This method does not require mining the length of the data's impact path on degradation problems based on complex mechanism chains. It can automatically mine the shortest potential impact path based on the graph structure and using a min function, thereby ensuring the integrity of the model while reducing the complexity of model construction. For example, data A may only affect degradation problem B, but degradation problem B affects degradation problem C. If only direct correlations are established between multiple soil attribute data and soil degradation problems, this step would require in-depth analysis based on mechanism mining of such indirect impact relationships. However, according to the method in this embodiment, only the direct impact relationships between indicator elements need to be considered.

[0038] Secondly, to characterize the interactions between different soil degradation problems, it is necessary to establish a coupling coefficient. The coupling coefficient represents the degree of coupling between soil degradation problems, and the calculation method for the coupling coefficient is as follows: A symbol matrix is ​​constructed based on the degradation mechanism. The element in the u-th row and v-th column of the symbol matrix... The value represents the effect of the worsening of the u-th type of soil degradation on the v-th type of soil degradation, with 1 representing promotion, -1 representing inhibition, and 0 representing no effect. Calculate the absolute value of the correlation coefficient between the exacerbation of type u soil degradation and the contribution of the degradation factors to type v soil degradation using historical samples. The selection criterion for historical samples is that the variance of the contribution of the degradation factor of the w-th soil degradation problem at each collection time in the historical samples is not greater than a preset variance threshold. , and The purpose of the selection criteria for historical samples is to control for interference from other variables; Calculate the coupling coefficient between the exacerbation of the u-th type of soil degradation problem and the v-th type of soil degradation problem. : .

[0039] The core design of the above coupling coefficients lies in reflecting the types of influences between soil degradation problems through mechanisms, and reflecting the degree of influence through the absolute value of the correlation coefficients. This ensures that the direction of the calculated coupling coefficients originates from clear physical mechanisms, and the strength comes from objective statistical data. By combining mechanistic analysis with historical data statistics, the limitations of relying solely on mechanisms or historical data are avoided. This results in coupling coefficients that possess both clear and mechanistic relationships and good numerical adaptability, thereby improving the model's ability to characterize complex degradation processes and more stably and reliably depicting the coupling relationships between different degradation problems.

[0040] Step S3: Construct a barrier degree model for each type of soil degradation problem. The input to the barrier degree model is a vector of various soil attribute data and a mapping relationship matrix. The output is the degradation factor contribution of each soil degradation problem in the current data collection period. The purpose of this step is to quantify the degree to which various soil attribute data represent different degradation problems, thereby obtaining the degradation factor contribution of each type of soil degradation problem.

[0041] Based on the above scheme, the method for constructing the barrier degree model for each type of soil degradation problem in this embodiment is as follows: Step S301: Standardize the various soil attribute data to obtain standardized multiple soil attribute data. The value range of the standardized multiple soil attribute data is [0,1] and all are positive indicators.

[0042] Different types of data differ in their units of measurement, value range, and evaluation direction. For example, some data are "the larger the better," while others are "the smaller the better." Some data may also have a standard value range, where being too large or too small is considered detrimental. By unifying standardization and positive transformation, the influence of units of measurement can be eliminated and the evaluation direction can be unified, providing a consistent data foundation for subsequent obstacle degree calculations. Standardization can be achieved using conventional normalization methods.

[0043] Step S302: Perform principal component analysis based on standardized multi-soil attribute data from N periods up to the current collection period. Based on the degree of variation of each soil attribute data in the time dimension and the principal component analysis results of the multi-soil attribute data from the N periods, calculate the first weight and the second weight of each soil attribute data respectively. Obtain the quality participation weight of the corresponding data by weighted summing of the first weight and the second weight.

[0044] Specifically, the method for calculating the first weight based on the degree of variation of each soil attribute data over time is as follows:

[0045]

[0046]

[0047] in, For the first The first weight of soil attribute data, Representing the The degree of variation of soil property data over time. This represents the total number of data types for various soil properties. For the t-th collection period up to the current collection period, the first... Soil property data The standardized value is N, which is the total number of collection cycles for the various soil attribute data when performing the principal component analysis, and is also the number of the current collection cycle.

[0048] Furthermore, the method for calculating the second weight based on the principal component analysis results of each of the various soil attribute data is as follows:

[0049]

[0050] in, For the first The second weight of soil attribute data, Let K be the eigenvalue of the k-th principal component, and K be the total number of principal components. Representing the The contribution of soil attribute data to the k-th principal component. It is the i-th element of the eigenvector of the k-th principal component.

[0051] This embodiment calculates information entropy based on the information distribution of each data point over time, thus obtaining the first weight. Data with significant temporal variation (i.e., uneven information distribution) has stronger discriminative power and is therefore assigned a higher weight, as data with small fluctuations provide fewer features. On the other hand, Principal Component Analysis (PCA) is used to determine the contribution of various soil attribute data to the overall information structure, measuring the importance of the data within the overall information structure and using this to obtain weights. Finally, the first and second weights are weighted and merged to obtain the corresponding data's quality participation weight. Both weights can be set to 0.5 during the weighting process.

[0052] Step S303: Based on the mapping matrix, the quality participation weights of each soil attribute data, and standardized multi-soil attribute data, calculate the degradation factor contribution of each soil degradation problem. The method is as follows: Obtain the initial degradation factor contribution for each of the aforementioned soil degradation problems:

[0053] in, The initial degradation factor contribution for the j-th type of soil degradation problem in the current data collection period. For the first The quality of soil attribute data is used as a weighting factor; Normalization yields the contribution of each degradation factor to each of the aforementioned soil degradation problems:

[0054] Where L represents the total number of soil degradation problems. The contribution of the degradation factor to the j-th type of soil degradation problem in the current collection period.

[0055] Due to the standardization of the preceding steps, All are positive indicators with values ​​in the range [0,1] (i.e., the larger the better). The processing can cause the data to deviate further from the ideal state, the greater the impact on degradation (i.e., the smaller the data, the worse the degradation, and thus...). The larger the value, the better. Assigning two coefficients to the multiplication and This approach incorporates both data weights and the strength of the mapping between data and soil degradation issues, avoiding both redundant and insufficient data considerations. The final result... It can clearly consider the dominant influencing factors of various degradation problems and the information content of the data itself.

[0056] Overall, step S3 transformed data from various soil properties into the contribution of degradation factors to different types of soil degradation problems. During the calculation process, it extracted the importance of various soil degradation problems relative to specific soil degradation problems, providing a foundation for subsequent risk evolution analysis. Step S4: Based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, the initial risk assessment parameters of various soil degradation problems in the monitoring sub-region of the current collection period are output through the risk evolution model.

[0057] As the preferred approach, the method for establishing the risk evolution model is as follows: Based on the data vectors of various soil attributes in the current collection period and the contribution of degradation factors for each soil degradation problem, based on the contribution of degradation factors for the j-th soil degradation problem in the current collection period. Multi-source driving items and type coupling items Output the initial risk assessment parameters for the j-th type of soil degradation problem. , :

[0058]

[0059] ; Where L represents the total number of soil degradation problems. The contribution of the degradation factor to the j-th type of soil degradation problem in the current collection period, where N is the number of the current collection period. This represents the total number of data types for various soil properties. The current collection cycle is the [number]th ... Soil property data Standardized values, The number of the previous collection cycle in the current collection cycle Soil property data Standardized values, To prevent positive numbers with a denominator of 0, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The worsening of soil degradation problems has a significant impact on the first The degree of coupling effect of soil degradation problems, where max represents finding the maximum value. , and These are the weighting coefficients.

[0060] In the above schemes, It is a static representation of multiple soil property data. Alternatively, it could be a dynamic representation based on historical event segments using multiple soil property data. It is a coupling term between different soil degradation problems.

[0061] Specifically The extraction focuses on the driving effects of changes in various soil properties on the current degradation problem. The relative change rate is obtained based on the changing trend of the i-th soil property data. multiplied by The impact of the i-th soil attribute data on the j-th soil degradation problem is taken into account, and then multiplied by the current... According to the current The magnitude of the value determines The degree of magnification of the item. For example, if the current... The fluctuations are small, and the system still has a buffering capacity even if data indicators fluctuate. The larger the value, the greater the risk from amplified fluctuations in the indicator. Simply put... This reflects that the more severe the degradation, the more sensitive it is to external changes.

[0062] in addition, It depicts the impact of the worsening of other soil degradation problems within the monitored sub-regions on the current soil degradation problem. Among them, To measure whether other soil degradation problems are aggravated, the max function is set to consider only deterioration and ignore improvement, and then a coupling relationship is introduced. To depict the extent to which other soil degradation problems are exacerbated.

[0063] Therefore, step S4 characterizes the degradation-driving process through the static features and change rates of various soil attribute data, and also considers the interactions between different degradation problems, enhancing the responsiveness to environmental changes and the characterization of the coupling of different internal soil degradation problems, thereby improving... Information content and expression skills.

[0064] After obtaining the risk evolution model output for each monitoring sub-region, the following steps can be performed: Step S5: Based on the risk evolution model output of each monitoring sub-region, conduct soil degradation trend analysis based on the initial risk assessment parameters, and output the soil degradation problems and corresponding final risk assessment parameters for each monitoring sub-region. This step mainly considers the spatial interaction between different monitoring sub-regions and the coupling relationship between different degradation problems to conduct spatial propagation analysis and trend assessment of soil degradation risk, thereby obtaining the final risk assessment parameters.

[0065] The final risk assessment parameters are calculated as follows: The final risk assessment parameters for the j-th type of soil degradation problem in the c-th monitoring sub-region during the current data collection period. for: ; in, The initial risk assessment parameters represent the type u soil degradation problem in the c-th monitoring sub-region during the current data collection period. Let V be the spatially distributed influence parameter of the v-th type of soil degradation problem in the g-th monitoring sub-region on the u-th type of soil degradation problem in the c-th monitoring sub-region. The weights are values ​​greater than 0 and less than 1, and N is the number of the current collection period.

[0066] exist In the calculation, This indicates the degradation status of the u-th type of soil degradation problem in the c-th monitoring sub-region. This represents the degradation state of the u-th type of soil degradation problem in other monitoring sub-regions, and spatial distribution influence parameters are introduced into this term. This is used to measure the impact of the degradation status of various soil degradation problems in other monitoring sub-regions on the degradation status of the u-th soil degradation problem in the current c-th monitoring sub-region itself. Influence parameters The establishment of the model takes into account the spatial distance between monitoring sub-regions and the degree of influence of various soil degradation problems on the currently calculated u-th type of soil degradation problem.

[0067] Based on this, the method for obtaining the influence parameters based on spatial distribution is as follows: Construct a spatial weight matrix among the monitored sub-regions, where the first... Line number Column elements The distance evaluation parameter between the c-th monitoring sub-region and the g-th monitoring sub-region is:

[0068] in, Let c be the center distance between the c-th monitoring sub-region and g-th monitoring sub-region. This is the distance attenuation coefficient; Calculate the :

[0069] in, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The degree of coupling effect of the exacerbation of the first type of soil degradation problem on the second type of soil degradation problem.

[0070] The result obtained in step S5 This can characterize the risk level of the j-th type of soil degradation problem in the c-th monitoring sub-region during the current data collection period. In specific implementation, it can be... Substitute the data into the judgment model to determine whether there is a risk of soil degradation of type j. The judgment model can be a threshold judgment or a trained multilayer perceptron model. The parameters of the judgment model can be obtained by training with labeled historical data.

[0071] Example 2 This invention also provides a soil degradation monitoring system for plains in arid and hot river valleys, applying the aforementioned soil degradation monitoring method for plains in arid and hot river valleys, see reference. Figure 2 ,include: The data acquisition module is used to periodically collect various soil property data of the monitored sub-regions; The degradation mechanism module is used to identify various soil degradation problems, establish a mapping relationship matrix between the various soil attribute data and the various soil degradation problems, and establish coupling coefficients between soil degradation problems; The obstacle degree model construction module is used to build obstacle degree models for each type of soil degradation problem. The initial assessment module is used to output initial risk assessment parameters for various soil degradation problems in the monitoring sub-region of the current collection period based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, through the risk evolution model. The final assessment module is used to analyze soil degradation trends based on the risk evolution model output of each monitoring sub-region and the initial risk assessment parameters, and output the soil degradation problems and corresponding final risk assessment parameters of the monitoring sub-region.

[0072] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring soil degradation in dry-hot valley plain land, characterized by, The area to be monitored is divided into grids to obtain multiple monitoring sub-regions. Each monitoring sub-region is monitored separately, including the following steps: Periodically collect and monitor multiple soil attribute data for sub-regions to form a data vector of multiple soil attributes at a unified time scale; determining a plurality of soil degradation problems, establishing a mapping relationship matrix between the plurality of soil property data and the plurality of soil degradation problems, the element in the i-th row and the j-th column of the mapping relationship matrix is a mapping coefficient between the i-th soil property data and the j-th soil degradation problem , representing the influence degree of the i-th soil property data on the j-th soil degradation problem , and establishing a coupling coefficient between the soil degradation problems, the coupling coefficient representing the coupling degree between the soil degradation problems; For each type of soil degradation problem, a barrier degree model is constructed. The input of the barrier degree model is a vector of multiple soil attribute data and a mapping relationship matrix, and the output is the contribution degree of degradation factor for each type of soil degradation problem in the current collection period. Based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, the initial risk assessment parameters of various soil degradation problems in the monitoring sub-region of the current collection period are output through the risk evolution model. Based on the risk evolution model output of each monitoring sub-region, soil degradation trend analysis is performed based on the initial risk assessment parameters, and the soil degradation problems and corresponding final risk assessment parameters of the monitoring sub-regions are output. The method for calculating the contribution of the degradation factor for each soil degradation problem is as follows: Obtain the initial degradation factor contribution for each of the aforementioned soil degradation problems: in, The initial degradation factor contribution for the j-th type of soil degradation problem in the current data collection period. For the first The quality of soil attribute data is included in the weighting. This is the [number]th [period] of the current data collection cycle. Soil property data Standardized values, The total number of data types for various soil properties; Normalization yields the contribution of each degradation factor to each of the aforementioned soil degradation problems: wherein L is the total number of soil degradation problems, is the degradation factor contribution of the jth soil degradation problem for the current collection cycle; The method for establishing the risk evolution model is as follows: based on the plurality of soil property data vectors of the current acquisition cycle and the degradation factor contribution of each soil degradation problem, the degradation factor contribution of the jth soil degradation problem based on the current acquisition cycle , multi-source driving terms and type coupling terms , output the initial risk evaluation parameter of the jth soil degradation problem , : ； Where L represents the total number of soil degradation problems. The contribution of the degradation factor to the j-th type of soil degradation problem in the current collection period, where N is the number of the current collection period. The number of the previous collection cycle in the current collection cycle Soil property data Standardized values, To prevent positive numbers with a denominator of 0, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The worsening of soil degradation problems has a significant impact on the first The degree of coupling effect of soil degradation problems, where max represents finding the maximum value. , and These are weighting coefficients; The calculation method for the final risk assessment parameters is as follows: The final risk assessment parameters for the j-th type of soil degradation problem in the c-th monitoring sub-region during the current data collection period. for: ； in, The initial risk assessment parameters represent the type u soil degradation problem in the c-th monitoring sub-region during the current data collection period. Let V be the spatially distributed influence parameter of the v-th type of soil degradation problem in the g-th monitoring sub-region on the u-th type of soil degradation problem in the c-th monitoring sub-region. The weights are values ​​greater than 0 and less than 1, and N is the number of the current collection period.

2. The method for monitoring soil degradation in plains of arid and hot river valleys according to claim 1, characterized in that, The method for constructing the barrier degree model for each type of soil degradation problem is as follows: The various soil attribute data are standardized to obtain standardized multiple soil attribute data. The value range of the standardized multiple soil attribute data is [0,1] and all of them are positive indicators. Principal component analysis is performed on standardized multi-soil attribute data from N periods up to the current collection period. Based on the degree of variation of each soil attribute data in the time dimension and the principal component analysis results of the multi-soil attribute data from N periods, the first weight and the second weight of each soil attribute data are calculated respectively. The quality participation weight of the corresponding data is obtained by weighted summation of the first weight and the second weight. Based on the mapping matrix, the quality participation weights of each soil attribute data, and the standardized multiple soil attribute data, the degradation factor contribution of each soil degradation problem is calculated.

3. The method for monitoring soil degradation in plains of arid-hot valley areas according to claim 2, characterized in that, The method for calculating the first weight based on the degree of variation of each soil attribute data over time is as follows: in, For the first The first weight of soil attribute data, Representing the The degree of variation of the various soil attribute data over time, where N is the total number of collection periods for the various soil attribute data when performing the principal component analysis, and is also the number of the current collection period; The method for calculating the second weight based on the principal component analysis results of each of the multiple soil attribute data is as follows: ； in, For the first The second weight of soil attribute data, Let K be the eigenvalue of the k-th principal component, and K be the total number of principal components. Representing the The contribution of soil attribute data to the k-th principal component. It is the i-th element of the eigenvector of the k-th principal component.

4. The method for monitoring soil degradation in plains of arid and hot river valleys according to claim 1, characterized in that, The method for establishing the mapping relationship matrix is ​​as follows: Construct a soil degradation mechanism relationship diagram. The soil degradation mechanism relationship diagram includes index elements and directed edges. The index elements include all data of the various soil attribute data and all the soil degradation problems. If the change of the p-th index element will cause the change of the q-th index element, then establish a directed edge from the p-th index element to the q-th index element. The element in the i-th row and j-th column of the mapping relationship matrix is ​​calculated based on the soil degradation mechanism relationship diagram. The value: in, The diagram showing the relationship between soil degradation mechanisms is shown in Figure 1. The minimum path between the index element corresponding to a certain type of soil attribute data and the index element corresponding to the j-th type of soil degradation problem. The path is defined as the number of directed edges connecting two index elements along a directed edge. The value range is greater than 0 and less than 1.

5. The method for monitoring soil degradation in plains of arid and hot river valleys according to claim 1, characterized in that, The method for obtaining the influence parameters based on spatial distribution is as follows: Construct a spatial weight matrix among the monitored sub-regions, where the first... Line 1 Column elements The distance evaluation parameter between the c-th monitoring sub-region and the g-th monitoring sub-region is: ； in, Let c be the center distance between the c-th monitoring sub-region and g-th monitoring sub-region. This is the distance attenuation coefficient; Calculate the : ； in, For the first Soil degradation problems and the first The coupling coefficient between the various soil degradation problems represents the first... The degree of coupling effect between the exacerbation of the first type of soil degradation problem and the vth type of soil degradation problem.

6. The method for monitoring soil degradation in plains of arid and hot river valleys according to claim 5, characterized in that, The coupling coefficient is calculated as follows: A symbol matrix is ​​constructed based on the degradation mechanism. The element in the u-th row and v-th column of the symbol matrix... The value represents the effect of the worsening of the u-th type of soil degradation on the v-th type of soil degradation, with 1 representing promotion, -1 representing inhibition, and 0 representing no effect. Calculate the absolute value of the correlation coefficient between the exacerbation of type u soil degradation and the contribution of the degradation factors to type v soil degradation using historical samples. The selection criterion for historical samples is that the variance of the contribution of the degradation factor of the w-th soil degradation problem at each collection time in the historical samples is not greater than a preset variance threshold. , and ; Calculate the coupling coefficient between the exacerbation of the u-th type of soil degradation problem and the v-th type of soil degradation problem. : .

7. A soil degradation monitoring system for plains in arid and hot river valleys, applied to the soil degradation monitoring method for plains in arid and hot river valleys as described in any one of claims 1-6, characterized in that, include: The data acquisition module is used to periodically collect various soil property data of the monitored sub-regions; The degradation mechanism module is used to identify various soil degradation problems, establish a mapping relationship matrix between the various soil attribute data and the various soil degradation problems, and establish coupling coefficients between soil degradation problems; The obstacle degree model construction module is used to build obstacle degree models for each type of soil degradation problem. The initial assessment module is used to output initial risk assessment parameters for various soil degradation problems in the monitoring sub-region of the current collection period based on the mapping relationship matrix, multiple soil attribute data vectors and degradation factor contribution, and considering the coupling relationship between soil degradation problems, through the risk evolution model. The final assessment module is used to analyze soil degradation trends based on the risk evolution model output of each monitoring sub-region and the initial risk assessment parameters, and output the soil degradation problems and corresponding final risk assessment parameters of the monitoring sub-region.

Images