A soil moisture detection method based on a forest-farmland compound management mode

Abstract

The application belongs to the technical field of soil moisture detection, and specifically discloses a soil moisture detection method based on a forest-farmland combined management mode, which comprises the following steps: arranging and collecting the soil moisture content, micro-terrain parameters and forest characteristic parameters of initial detection points, respectively calculating terrain and forest hydrology correction coefficients, cooperatively correcting the soil moisture content, judging the corrected moisture content based on the reference moisture threshold of each depth layer, identifying abnormal detection points, spatially clustering the abnormal points to divide abnormal areas, calculating the moisture abnormality degree in combination with the corrected moisture content, dynamically adjusting the detection point arrangement density according to the moisture abnormality degree, determining the final arrangement position, and completing the arrangement of the detection points.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of soil moisture detection and relates to a soil moisture detection method based on a forest-farmland compound management mode. BACKGROUND

[0002] In a forest-farmland compound system, the spatial differentiation of soil moisture is jointly influenced by topographic hydrological processes and physiological water consumption of trees, and is highly heterogeneous. This leads to inherent defects of low precision and poor coverage of traditional uniform point detection methods. Developing precise soil moisture sensing technology that can cope with such complexity has become a core requirement to improve the efficiency of system water management.

[0003] For example, the Chinese invention patent with publication number CN119667096A discloses a soil moisture detection device for plant monitoring, which realizes the synchronous collection, integration and abnormal state early warning of multi-source soil data through the cooperation of the data collection unit, the synchronous control module, the data integration analysis module and the monitoring and early warning module. This technical solution makes overall analysis by constructing a soil data model, and has made beneficial progress in ensuring the generality of data synchronization and abnormality identification.

[0004] The above existing technologies have the following deficiencies: 1. The current data integration fails to quantify the core environmental driving factors in the forest-farmland compound system. Neither hydrological correction coefficients based on terrain parameters such as slope and aspect, nor forest correction coefficients reflecting the effects of canopy and root system are introduced, and further lack of such collaborative correction based on physical and biological mechanisms, resulting in that the processing of soil moisture data stays at the level of post-hoc explanation, and cannot realize prior correction, so the measurement accuracy is significantly reduced in the highly heterogeneous forest-farmland interlaced area.

[0005] 2. The current monitoring and early warning function terminates at abnormality identification and alarm, constituting an open-loop system, and fails to convert the diagnosis results into feedback signals for optimizing the layout of the detection network, and further cannot dynamically adjust the layout density and position of the detection points according to the spatial clustering characteristics and water anomaly degree of the abnormal areas, leading to rigid allocation of detection resources, and thus cannot realize adaptive investment in key areas, restricting the efficiency and effectiveness of large-scale detection. SUMMARY

[0006] In view of this, in order to solve the problems raised in the background art, a soil moisture detection method based on a forest-farmland compound management mode is proposed.

[0007] The object of the present application can be achieved by the following technical solutions: The present application provides a soil moisture detection method based on a forest-farmland compound management mode, comprising: S1, arranging initial detection points in the vertical gradient direction of the forest belt and farmland in the target area, collecting the initial soil moisture content of each initial detection point at different depth layers, and synchronously acquiring the microtopography parameters and forest characteristics parameters of each initial detection point.

[0008] S2, calculating the terrain hydrology correction coefficient and the forest hydrology correction coefficient based on the microtopography parameters and the forest characteristics parameters respectively, and cooperatively correcting the initial soil moisture content according to the terrain hydrology correction coefficient and the forest hydrology correction coefficient to obtain the corrected soil moisture content.

[0009] S3, based on the dynamic suitable moisture range of different depth layers, performing anomaly determination on the corrected soil moisture content to identify each abnormal detection point.

[0010] S4, based on the spatial clustering of each abnormal detection point, dividing each abnormal area, and calculating the moisture anomaly degree combining the corrected soil moisture content in each abnormal area.

[0011] S5, based on the moisture anomaly degree, dynamically adjusting the layout density of the reference detection points, determining the layout position of each detection point according to the layout density, completing the deployment of the detection points and collecting the soil moisture data.

[0012] Compared with the prior art, the present application has the following advantages: (1) The present application calculates the terrain hydrology correction coefficient and the forest hydrology correction coefficient, and cooperatively corrects the initial soil moisture content based on the two, which improves the physical and biological hydrological effects of terrain and forest from the posteriori explanation to the priori correction level, thereby significantly improving the accuracy and reliability of soil moisture detection in the forest-farmland compound area.

[0013] (2) The present application divides the abnormal area based on the spatial clustering algorithm, and calculates the moisture anomaly degree combining the corrected soil moisture content and the dynamic suitable moisture range, which realizes the upgrading of the soil moisture anomaly condition from point identification to regionalization and quantitative evaluation, and provides a direct basis for accurately positioning the problem area and evaluating its severity.

[0014] (3) The present application dynamically adjusts the layout density of the detection points in each abnormal area according to the moisture anomaly degree, and sets an upper limit value of the layout density for constraint, so that the detection network has self-adaptive optimization capability, and then realizes the configuration transformation of the detection resources from static uniform distribution to dynamic focusing, thereby significantly improving the detection efficiency and cost-effectiveness in a large-scale scene.

[0015] (4) The present application establishes a collaborative correction rule based on the terrain and the hydrological correction coefficient of the forest, so that the comprehensive correction coefficient can dynamically adapt to different hydrological scenarios, overcoming the limitations of traditional linear weighting in accurately reflecting the interaction between factors in complex ecological hydrological processes, thereby realizing the fidelity of the simulation of the water movement law of the forest-farmland complex system. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The schematic diagram for connecting each step of the method of the present application is shown.

[0018] Figure 2 The schematic diagram for connecting the step of dividing the abnormal area of the present application is shown.

[0019] Figure 3 The schematic diagram for connecting the step of determining the position of the detection point of the present application is shown. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] Please refer to Figure 1 As shown in the drawings, the present application provides a soil moisture detection method based on a forest-farmland complex management mode, which comprises the following steps: S1, arranging initial detection points in the vertical gradient direction of the forest belt and farmland in the target area, collecting the initial soil moisture content of each initial detection point at different depth layers, and synchronously acquiring the microtopography parameters and forest characteristic parameters of each initial detection point, wherein the vertical gradient direction refers to the direction perpendicular to the forest belt and farmland intersection zone, and the different depth layers include but are not limited to the surface layer, the main root activity layer and the deep layer.

[0022] S2, calculating the terrain hydrological correction coefficient and the forest hydrological correction coefficient based on the microtopography parameters and the forest characteristic parameters respectively, and collaboratively correcting the initial soil moisture content according to the terrain hydrological correction coefficient and the forest hydrological correction coefficient to obtain the corrected soil moisture content.

[0023] Illustratively, the calculating terrain hydrology correction coefficient comprises: S2-1, obtaining the slope of each initial detection point from the micro-topographic parameters, and matching the slope with the slope effect factor corresponding to each slope interval to obtain the slope effect factor of each initial detection point.

[0024] It should be noted that the slope effect factor corresponding to each slope interval is a quantitative parameter reflecting the influence of a specific slope on the surface runoff speed and the soil moisture infiltration capacity. The theoretical basis is that the greater the slope, the faster the surface runoff speed, and the shorter the soil moisture infiltration time, thereby causing the soil surface water loss to intensify and the water effect to weaken. Therefore, the factor is a dimensionless value positively correlated with the slope.

[0025] The slope effect factor is obtained by table lookup. The system pre-stores a mapping relationship table of slope intervals and slope effect factors as shown in Table 1, and the slope effect factor corresponding to the slope of the detection point can be obtained by querying the table.

[0026] Table 1: Mapping table of slope interval and slope effect factor

[0027]

[0028] The matching process of the slope effect factor is realized by pre-establishing the mapping relationship of the slope interval-effect factor. In specific implementation, the mapping relationship can be stored in the system memory in the form of a lookup table. When the slope data of the detection point is obtained, the system automatically queries the interval range to which the slope of the detection point belongs, and extracts the corresponding slope effect factor value.

[0029] In actual application, the mapping relationship of the slope interval and the slope effect factor can be adaptively adjusted according to the topographic and climatic characteristics of different regions. For example, in arid and semi-arid areas, due to the large water evaporation potential, the effect factor baseline value of each slope interval can be increased based on local hydrology observation data, so as to more accurately quantify the effect of water loss intensified by slope in the region.

[0030] S2-2, obtaining the slope direction of each initial detection point from the micro-topographic parameters, and calculating the slope direction effect factor of each initial detection point.

[0031] Further, the calculating the slope direction effect factor of each initial detection point comprises: S2-2-1, calculating the direction angle cosine value between each initial detection point and the dominant environmental direction based on the slope direction of each initial detection point and the dominant environmental direction.

[0032] The main environmental direction is any of the following: a geographic south direction, as a default reference direction, applicable to a general scenario where solar radiation is the main factor of evapotranspiration; and a prevailing wind direction determined based on meteorological data of a target region, applicable to a specific region where wind-driven evapotranspiration is significant.

[0033] It should be noted that the calculation of the direction angle cosine value is based on the principle of a space vector. The slope direction of the detection point is converted into a unit vector, and a dot product operation is performed with the main environmental direction vector. The result is the direction angle cosine value. The cosine value ranges from -1 to 1, and is converted into a slope direction effect factor in the interval [0, 1] through linear mapping. The main environmental direction can be selected according to the specific characteristics of the detection region.

[0034] S2-2-2, linearly mapping the direction angle cosine value to obtain the slope direction effect factor of each initial detection point.

[0035] The calculation process of the slope direction effect factor is as follows: first, calculate the direction angle cosine value of all initial detection points in the target region, and form a sample set. Then, sort the sample set from small to large, and map the direction angle cosine value of each detection point to the interval [0, 1] based on the actual range of the sample through the following linear normalization formula. The linear mapping formula is , where is the slope direction effect factor, is the direction angle cosine value, and are the maximum and minimum values of all direction angle cosine values in the sample set, respectively. When , the slope direction effect factor of all detection points is uniformly taken as 0.5.

[0036] S2-3, weighted fusion calculation of the slope effect factor and the slope direction effect factor to obtain the topographic hydrological correction coefficient of each initial detection point.

[0037] It should be noted that the calculation formula of the topographic hydrological correction coefficient is: , where is the topographic hydrological correction coefficient, and are the slope effect factor and the slope direction effect factor, respectively, and are the weights of the slope effect factor and the slope direction effect factor, respectively, used to quantify the different dimensional effects of the slope effect factor and the slope direction effect factor on the topographic hydrological correction, for example , .

[0038] The weighted fusion calculation reflects the differentiated influence of the slope and the aspect on the terrain hydrological correction on one hand, and directly fuses the information of the two dimensions to comprehensively consider the contribution of the two to the terrain hydrological correction on the other hand. The weight can be set according to the regional terrain hydrological characteristics and actual engineering experience, or can be obtained through statistical analysis of historical data: collect historical terrain data and corresponding hydrological effect data, calculate the correlation coefficients of the slope, aspect factor and hydrological correction amount, determine the respective contribution degrees through regression analysis, and convert them into weights after normalization processing, and the sum of the weights is 1, so as to accurately quantify the terrain hydrological correction coefficient.

[0039] Illustratively, the calculation of the forest hydrological correction coefficient comprises: obtaining the forest coverage area and root water absorption area of each initial detection point from the forest feature parameters.

[0040] In an optional implementation, the forest coverage area can be obtained through remote sensing image interpretation or field measurement. For example, high-resolution images can be obtained by unmanned aerial vehicle aerial photography, and single-tree crown width can be identified through image segmentation algorithm, and then the forest coverage area can be calculated. The root water absorption area can be estimated by a forest root distribution model, and can be calculated based on the growth parameters such as tree diameter at breast height and tree height, and the root distribution characteristic function of different tree species. In actual implementation, a parameter database of main tree species can be established to improve the accuracy of the estimation of the root water absorption area.

[0041] The forest coverage area and the root water absorption area of each initial detection point are sorted in descending order respectively, and the middle values are taken as the reference forest coverage area and the reference root water absorption area respectively, wherein when the data quantity is even, the arithmetic mean of the middle two values after sorting is taken as the middle value.

[0042] It should be noted that the middle value is used as the reference, which can effectively avoid the interference of abnormal samples and enhance the robustness of the model due to its insensitivity to extreme values, and can reflect the typical level of the data set, so that the reference is more representative, and the calculation result is closer to the actual situation of most detection points.

[0043] The forest coverage area and the root water absorption area of each initial detection point are divided by the reference forest coverage area and the reference root water absorption area respectively, to obtain the canopy coverage relative index and the root water absorption relative index of each initial detection point.

[0044] The canopy coverage relative index and the root water absorption relative index are multiplied to obtain the forest hydrological correction coefficient of each initial detection point.

[0045] Exemplarily, the performing the collaborative correction comprises: comparing the terrain hydrology correction coefficient and the forest hydrology correction coefficient with the reference correction coefficient respectively, and selecting different collaborative correction rules to calculate the comprehensive correction coefficient based on the size relationship between the terrain hydrology correction coefficient, the forest hydrology correction coefficient and the reference correction coefficient.

[0046] The reference correction coefficient is an empirical threshold for judging the significant degree of terrain or forest influence, and can be set according to the statistical distribution of regional historical data or the experience of field experts, for example, set to 0.5, and the implementer can adjust it according to the specific application scene.

[0047] Further, the calculating the comprehensive correction coefficient comprises: when the terrain hydrology correction coefficient and the forest hydrology correction coefficient are both greater than the reference correction coefficient, multiplying the two to obtain the comprehensive correction coefficient.

[0048] When the two are both less than or equal to the reference correction coefficient, the arithmetic mean of the two is taken as the comprehensive correction coefficient.

[0049] When one of the terrain hydrology correction coefficient and the forest hydrology correction coefficient is greater than the reference correction coefficient, and the other is less than or equal to the reference correction coefficient, the maximum value of the two is taken as the comprehensive correction coefficient.

[0050] Based on this, when both are greater than the reference value, it indicates that the terrain and the forest both significantly enhance the hydrological effect, and the multiplication rule can reflect the synergistic amplification of the effect. When both are small, the arithmetic mean can avoid excessive correction. When one is significant and the other is not, taking the maximum value can ensure that the effect of the dominant factor is fully reflected. Compared with the fixed weight fusion method, this dynamic fusion strategy based on comparison can more accurately reflect the actual change rule of the hydrological effect in a complex environment.

[0051] The comprehensive correction coefficient is multiplied by the initial soil moisture content of each initial detection point at different depth layers to obtain the corrected soil moisture content.

[0052] S3, based on the dynamic appropriate moisture range of different depth layers, the corrected soil moisture content is abnormally judged, and each abnormal detection point is identified.

[0053] Exemplarily, the abnormally judging the corrected soil moisture content comprises: comparing the corrected soil moisture content with the dynamic appropriate moisture range of the corresponding depth layer.

[0054] As an example, the dynamic suitable moisture range is a reasonable moisture range dynamically updated according to seasons, vegetation growth stages and meteorological conditions. The dynamic suitable moisture range is usually obtained based on historical data. For example, long-term detected soil moisture data at each depth layer can be collected, and a moisture statistical model can be constructed for different growth periods or seasons to define the dynamic range of the period with the median ± 1.5 times the interquartile range.

[0055] When the corrected soil moisture content continuously falls below the lower limit of the dynamic suitable moisture range, it is determined to be a water stress abnormal type. In a preferred embodiment, the continuous refers to that the corrected soil moisture content meets the corresponding abnormal condition in the continuous 3 set periods, and the set period can be determined according to the specific detection requirement, such as 1 day or 1 week.

[0056] When the corrected soil moisture content continuously exceeds the upper limit of the dynamic suitable moisture range, it is determined to be a water saturation abnormal type.

[0057] When the fluctuation amplitude of the corrected soil moisture content in the set period exceeds the water dynamic abnormal threshold, it is determined to be a water dynamic abnormal type.

[0058] In addition, the water dynamic abnormal threshold is a critical change rate value for determining whether the soil moisture content is in an abnormal and violent fluctuation state. The threshold does not focus on the absolute value of the moisture content, but on the change speed and amplitude thereof in the set period, which is used to capture abnormal conditions that are not continuously deviated from the static moisture range but have extremely unstable change process.

[0059] The water dynamic abnormal threshold is usually obtained based on statistical analysis of historical data and regional soil physical properties. For example, the historical change rate data of the soil moisture content at each depth layer in the target region in the past 3-5 years in the same or similar set period is collected, and the higher quantile of the change rate data sequence, such as the 90th or 95th percentile, is taken as the water dynamic abnormal threshold of the period, to ensure effective identification and early warning of the violent dynamic fluctuation in history.

[0060] The initial detection points determined to be the water stress abnormal type, the water saturation abnormal type and the water dynamic abnormal type are marked as abnormal detection points.

[0061] S4, based on each abnormal detection point, spatial clustering is performed to divide each abnormal region, and the water abnormality degree is calculated based on the corrected soil moisture content in each abnormal region.

[0062] Please refer to Figure 2 As shown, for example, the dividing each abnormal region includes: based on the spatial distribution position of each abnormal detection point and the abnormal type thereof, an abnormal region is divided by using a spatial clustering algorithm.

[0063] In one specific embodiment, the spatial clustering algorithm can employ the DBSCAN algorithm. This algorithm can effectively identify clusters of arbitrary shape and automatically distinguish noise points, and is particularly suitable for scenarios where soil moisture anomaly points may be irregularly distributed. The neighborhood threshold and minimum point number in the DBSCAN algorithm can be optimized and determined according to the characteristics of the detection area by the profile coefficient method or the elbow method to obtain the best clustering effect.

[0064] The anomaly detection points with a spatial distance less than a preset neighborhood threshold and consistent anomaly types are aggregated into the same anomaly cluster, and then each anomaly cluster is obtained.

[0065] The preset neighborhood threshold refers to the maximum distance used to determine whether two anomaly detection points are close enough in space to be classified into the same anomaly cluster. This threshold can be directly set according to business prior knowledge or detection point placement scheme, such as taking 50%-70% of the average distance between sensors as the threshold.

[0066] If the anomaly cluster contains only one anomaly detection point, the anomaly region is demarcated with the anomaly detection point as the center and a preset influence radius as the boundary.

[0067] If the anomaly cluster contains multiple anomaly detection points, a spatial convex hull algorithm is used to generate a minimum convex polygon containing all anomaly detection points as the anomaly region.

[0068] Illustratively, the calculation of the moisture anomaly degree includes: calculating the area of each anomaly region, and dividing the number of anomaly detection points contained in each anomaly region by the area of the anomaly region to obtain the anomaly detection point distribution density.

[0069] Specifically, the calculation method of the area of the anomaly region adopts different strategies according to the shape characteristics of the region: a single anomaly detection point is taken as a radius with a preset influence radius, and the area is accurately calculated by the circular area formula. For an anomaly cluster containing multiple anomaly detection points, a spatial convex hull algorithm is used to calculate the area of the minimum convex polygon. These two methods are respectively suitable for typical spatial distribution patterns, ensuring the accuracy and applicability of area calculation. The circular shape is approximately suitable for anomalies with uniform diffusion of influence range, while the polygonal depiction can accurately capture irregular anomaly regions distributed along specific terrain or features, providing a reliable area benchmark for subsequent calculation of anomaly distribution density.

[0070] The absolute difference between the corrected soil moisture content and the corresponding depth layer dynamic suitable moisture range reference value is calculated, and the absolute difference is averaged to obtain the average deviation of each depth layer, wherein the reference value is the arithmetic mean of the upper and lower limits of the corresponding depth layer dynamic suitable moisture range.

[0071] The overall average deviation degree of the abnormal region is obtained by weighted average of the average deviation degrees of each depth layer based on the abnormal type corresponding to the abnormal region and preset weights of the abnormal type.

[0072] Specifically, the calculation formula of the overall average deviation degree is: .

[0073] In the formula, D is the overall average deviation degree, N is the number of depth layers, is the average deviation degree of the i-th depth layer, is the preset weight of the abnormal type corresponding to the abnormal region, which is used to quantify the relative severity of the influence of different abnormal types on the regional hydrological system. The higher the weight, the greater the potential impact of the abnormal type. For example, the water stress and water saturation abnormal types have greater direct harm to crop or forest growth, and the weight can be set to 1.2 when the abnormal type corresponding to the abnormal region is the water stress or water saturation abnormal type. The water dynamic abnormal type may represent a potential risk, and the weight can be set to 0.8 when the abnormal type corresponding to the abnormal region is the water dynamic abnormal type.

[0074] By introducing the preset weight of the abnormal type, the differentiated influence of different types of water anomalies is realized. The weight can be set according to the experience of experts in the field, or can be obtained through historical data modeling analysis. For example, the historical occurrence frequency, duration and actual damage data of each abnormal type to the ecological environment or crop yield are collected, a regression model of abnormal characteristics and damage degree is established, and the model coefficients are standardized and converted into the weight of each abnormal type, so as to ensure that the potential influence severity of each abnormal type can be objectively quantified.

[0075] Based on the distribution density and the overall average deviation degree, the water anomaly degree of each abnormal region is calculated by geometric mean method.

[0076] Therefore, the water anomaly degree is calculated by geometric mean method, and the specific calculation formula is: .

[0077] In the formula, D is the overall average deviation degree, ​​​The distribution density is the density of distribution. The core of the method is to combine the distribution density which describes the spatial clustering characteristics of outliers with the overall average deviation which reflects the intensity characteristics of outliers. The main advantages are as follows: first, the distribution density and the overall average deviation usually have different dimensions and orders of magnitude, and the geometric mean can effectively balance the difference and avoid the situation that one index dominates the final result due to its large absolute value. Second, the inherent multiplication characteristics of the method emphasize the synergistic effect of the two indexes, that is, an area with high distribution density and high overall deviation will be identified as a serious problem area with extremely high anomaly degree. Compared with the traditional weighted average, the geometric mean method reduces the subjectivity of setting weights, making the evaluation result more objective and stable.

[0078] S5, dynamically adjusting the reference detection point layout density based on the water anomaly degree, determining the layout positions of the detection points according to the reference detection point layout density, completing the detection point deployment and collecting soil moisture data.

[0079] Referring to Figure 3 As shown in the figure, the determination of the layout positions of the detection points includes: multiplying the water anomaly degree of each abnormal area by the reference detection point layout density to obtain a preliminary layout density.

[0080] The preliminary layout density is compared with the layout density upper limit value. If the preliminary layout density is greater than the layout density upper limit value, the layout density upper limit value is taken as the target layout density, otherwise the preliminary layout density is taken as the target layout density.

[0081] It should be noted that the layout density upper limit value refers to the maximum density of detection points allowed in a specific detection area after considering the economic cost, equipment performance and detection demand. When setting, the smaller value between the technically feasible density and the economically feasible density should be taken to ensure that the detection network is technically feasible and economically reasonable.

[0082] The target layout density is multiplied by the area of the abnormal area to obtain the total number of detection points that should be laid out in the abnormal area.

[0083] Based on the total number of detection points, the layout positions of the detection points are determined in the abnormal area range according to the equidistant grid layout method.

[0084] Specifically, the equidistance grid arrangement can be implemented by the following steps: firstly, calculating the theoretical grid size according to the shape and area of the abnormal area; then, extending the grid points outward from the geometric center of the area according to the grid size; finally, removing or adjusting the points falling outside the area boundary. To ensure uniform arrangement, an optimization algorithm can be used to fine-tune the positions of the grid points, and for the grid points falling outside the boundary of the abnormal area, they can be adjusted to the nearest boundary position or determined by an interpolation method according to the terrain features. In implementation, the grid spacing can be determined according to the detection accuracy requirements and device performance, and in a preferred embodiment, the grid spacing is between 10-50 meters.

[0085] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product.

[0086] Those skilled in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized by hardware or software depends on the specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0087] In addition, the functional modules in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0088] The above is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0089] Finally, the above is merely a preferred embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for soil moisture detection based on an agroforestry management model, characterized in that: The method comprises: S1, arranging initial detection points in the vertical gradient direction of forest belts and farmland in the target area, collecting initial soil moisture content of each initial detection point at different depth layers, and synchronously acquiring microtopographic parameters and tree characteristic parameters of each initial detection point; S2, calculating terrain hydrological correction coefficients and tree hydrological correction coefficients based on the microtopographic parameters and the tree characteristic parameters, and cooperatively correcting the initial soil moisture content according to the terrain hydrological correction coefficients and the tree hydrological correction coefficients to obtain corrected soil moisture content; S3, performing abnormality judgment on the corrected soil moisture content based on dynamic suitable moisture ranges of different depth layers to identify each abnormal detection point; S4, performing spatial clustering based on each abnormal detection point to divide each abnormal area, and calculating a moisture abnormality degree in combination with the corrected soil moisture content in each abnormal area; S5, dynamically adjusting a reference detection point arrangement density based on the moisture abnormality degree, determining arrangement positions of each detection point according to the reference detection point arrangement density, completing detection point arrangement, and collecting soil moisture data; The tree coverage area and the root water absorption area of each initial detection point are obtained from the tree characteristic parameters; the middle values of the tree coverage areas and the root water absorption areas of the initial detection points are taken as reference tree coverage areas and reference root water absorption areas after the tree coverage areas and the root water absorption areas of the initial detection points are respectively sorted in descending order; the tree coverage areas and the root water absorption areas of the initial detection points are respectively divided by the reference tree coverage areas and the reference root water absorption areas to obtain crown layer coverage relative indexes and root water absorption relative indexes of the initial detection points; the crown layer coverage relative indexes and the root water absorption relative indexes are multiplied to obtain tree hydrological correction coefficients of the initial detection points; The terrain hydrological correction coefficients and the tree hydrological correction coefficients are compared with reference correction coefficients, and based on the size relationship between the terrain hydrological correction coefficients, the tree hydrological correction coefficients and the reference correction coefficients, different cooperative correction rules are selected to calculate comprehensive correction coefficients; the comprehensive correction coefficients are multiplied by the initial soil moisture content of each initial detection point at different depth layers to obtain corrected soil moisture content; The area of each abnormal area is calculated, and the number of abnormal detection points contained in each abnormal area is divided by the area of the abnormal area to obtain an abnormal detection point distribution density; the absolute difference between the corrected soil moisture content and the reference value of the dynamic suitable moisture range of the corresponding depth layer is calculated, and the absolute difference is averaged to obtain the average deviation of each depth layer; the average deviation of each depth layer is weighted and averaged based on the preset weight for different abnormal types to obtain the overall average deviation of the abnormal area; the distribution density and the overall average deviation are used to calculate the moisture abnormality degree of each abnormal area by a geometric mean method. The moisture abnormality degree of each abnormal area is multiplied by the reference detection point arrangement density to obtain a preliminary arrangement density; the preliminary arrangement density is compared with the arrangement density upper limit value, if the preliminary arrangement density is greater than the arrangement density upper limit value, the arrangement density upper limit value is taken as the target arrangement density, otherwise the preliminary arrangement density is taken as the target arrangement density; the target arrangement density is multiplied by the area of the abnormal area to obtain the total number of detection points that should be arranged in the abnormal area; based on the total number of detection points, the arrangement positions of each detection point are determined in the abnormal area range according to the equidistance grid arrangement mode.

2. The soil moisture detection method based on the forest-farmland complex management mode according to claim 1, characterized in that: The calculation of the terrain hydrology correction coefficient comprises: The slope of each initial detection point is obtained from the micro-terrain parameters, and the slope effect factor corresponding to each slope interval is matched to obtain the slope effect factor of each initial detection point; The aspect of each initial detection point is obtained from the micro-terrain parameters, and the aspect effect factor of each initial detection point is calculated; The slope effect factor and the aspect effect factor are weighted and fused to calculate the terrain hydrology correction coefficient of each initial detection point.

3. The soil moisture detection method based on the forest-farmland complex management mode according to claim 2, characterized in that: The calculation of the aspect effect factor of each initial detection point comprises: Based on the aspect of each initial detection point and the dominant environmental direction, the direction included angle cosine value between each initial detection point and the dominant environmental direction is calculated; The direction included angle cosine value is linearly mapped to obtain the aspect effect factor of each initial detection point.

4. The soil moisture detection method based on the forest-farmland complex management mode according to claim 1, characterized in that: The calculation of the comprehensive correction coefficient comprises: When the terrain hydrology correction coefficient and the forest hydrology correction coefficient are both greater than the reference correction coefficient, the product of the two is taken as the comprehensive correction coefficient; When both are less than or equal to the reference correction coefficient, the arithmetic mean of the two is taken as the comprehensive correction coefficient; When one of the terrain hydrology correction coefficient and the forest hydrology correction coefficient is greater than the reference correction coefficient, and the other is less than or equal to the reference correction coefficient, the maximum value of the two is taken as the comprehensive correction coefficient.

5. The soil moisture detection method based on the forest-farmland complex management mode according to claim 1, characterized in that: The abnormal determination of the corrected soil moisture content comprises: The corrected soil moisture content is compared with the dynamic suitable moisture range of the corresponding depth layer; When the corrected soil moisture content continuously falls below the lower limit of the dynamic suitable moisture range, it is determined as a water stress abnormal type; When the corrected soil moisture content continuously exceeds the upper limit of the dynamic suitable moisture range, it is determined as a water saturation abnormal type; When the fluctuation amplitude of the corrected soil moisture content within a set period exceeds the water dynamic abnormal threshold, it is determined as a water dynamic abnormal type; The initial detection points determined as the water stress abnormal type, the water saturation abnormal type and the water dynamic abnormal type are marked as abnormal detection points. 6.The soil moisture detection method based on a forest-farmland complex management mode according to claim 1, characterized in that: The division of each abnormal area comprises: Based on the spatial distribution position and the abnormal type of each abnormal detection point, a spatial clustering algorithm is used for abnormal area division; Abnormal detection points with a spatial distance less than a preset neighborhood threshold and the same abnormal type are aggregated into the same abnormal cluster, and then each abnormal cluster is obtained; If an abnormal cluster only contains one abnormal detection point, the abnormal area is delimited with the abnormal detection point as the center and a preset influence radius as the boundary; If the abnormal cluster contains multiple abnormal detection points, a spatial convex hull algorithm is used to generate a minimum convex polygon containing all abnormal detection points as an abnormal region.

Images