Slope surface interflow monitoring method combining geophysical detection and soil moisture sensor network

Abstract

The invention discloses a slope interflow monitoring method combining geophysical exploration and a soil moisture sensor network. The slope interflow monitoring method comprises the following steps: acquiring rainfall data, resistivity space dynamic distribution data, multi-depth and multi-slope soil volumetric water content data, soil temperature data and multi-depth runoff data of a target runoff plot; processing the resistivity data; establishing a quantitative relation model between the resistivity and the soil water content; determining rainfall data and resistivity data for identifying an interflow path and a dynamic change process; calculating to obtain soil moisture spatial distribution dynamic states of monitoring ranges at different moments; the inversion result is verified; and visualizing output based on the dynamic process. The method is focused on multi-source data fusion and monitoring system structure innovation, breaks through the technical bottleneck of interflow dynamic path identification and quantitative analysis under the in-situ condition, and provides high-precision and non-intrusive method support for slope rainfall runoff relation analysis and mountainous area water and soil process in-situ monitoring.

Description

Technical Field

[0001] The invention relates to a slope soil midflow monitoring method combining geophysical detection with a soil moisture sensor network, and belongs to the field of environmental science and technology. Background Art

[0002] As a key transporter of slope hydrological processes, midstream flow is not only a crucial component of mountain runoff but also a core factor in regulating soil erosion and the evolution of flood disasters. In recent years, driven by the dual effects of global climate change, altered rainfall patterns, and increased land development in mountainous areas, the coupling effect between midstream flow and surface erosion has become increasingly pronounced. Developing a precise midstream flow monitoring system has become a cutting-edge issue in the development of ecological security barriers in mountainous areas.

[0003] The current soil flow monitoring technology system suffers from a significant gap in observation scales. Traditional point-scale monitoring methods (such as time-domain reflectometry (TDR) and neutron moisture meters) can obtain local soil moisture time-series data but struggle to reconstruct the soil flow path network. While cross-section excavation dye tracing methods can visually visualize soil flow paths during specific rainfall events, they disrupt soil structure and fail to capture dynamic processes. Notably, electrical resistivity tomography (ERT) technology, through the deployment of high-density electrode arrays, can non-invasively acquire the spatiotemporal distribution of soil resistivity. Its strong correlation with soil moisture provides a new approach for inverting soil flow dynamics. However, single ERT data suffer from multi-resolution bottlenecks. Resistivity anomalies may correspond to multiple factors, such as water migration, changes in soil density, or root distribution. Therefore, a verification mechanism for multi-source data fusion is urgently needed.

[0004] It is worth noting that the standardized runoff plot, as a classic experimental unit for hydrological research, has closed boundary conditions and a complete outlet flow monitoring system, providing an ideal monitoring site for ERT technology. However, existing technologies have not yet formed a complete solution that organically integrates ERT continuous monitoring with runoff plot system verification, which restricts the depth of research on the mechanism of soil flow processes and the reliability of engineering applications.

[0005] Therefore, there is an urgent need for a monitoring method that uses runoff plots as sites and ERT technology as the main monitoring method to accurately and reliably capture the soil flow path and dynamic changes. Summary of the Invention

[0006] The purpose of the present invention is to address the problems existing in the prior art and to provide a method for monitoring soil midflow on a slope by combining geophysical detection with a soil moisture sensor network.

[0007] The present invention provides a technical solution to solve the above technical problems: a method for monitoring mid-slope soil flow by combining geophysical detection and a soil moisture sensor network, comprising the following steps:

[0008] Step 1: Obtain rainfall data, resistivity data, spatial dynamic distribution data, soil volume moisture content data and soil temperature data at multiple depths and slopes, and multi-depth runoff data for the target runoff area;

[0009] Step 2: Process the resistivity data by terrain and temperature correction, soil moisture inversion, etc.

[0010] Step 3: Establish a quantitative relationship model between resistivity and soil moisture content;

[0011] Step 4: Determine rainfall data and resistivity data for identifying soil flow paths and dynamic change processes;

[0012] Step 5: Obtain the spatial distribution dynamics of soil moisture in the monitoring range at different times through time-lapse geophysical scanning and inversion calculation;

[0013] Step 6: Verify the inversion results by dynamic water content verification, point scale verification, and soil outflow flow verification;

[0014] Step 7: Visualize the dynamic process output based on the inversion results to achieve in-situ high-precision monitoring of the slope soil flow path and dynamic change process during rainfall events.

[0015] A further technical solution is to obtain rainfall data, resistivity data, soil volume moisture data, soil temperature data, and runoff data in step one through a rain gauge, an electrical resistivity tomography (ERT), a time domain reflectometry (TDR) system, and a flow meter.

[0016] A further technical solution is to use the trigonometric linear interpolation method in step 1 to interpolate the inverted resistivity value at the location of the TDR sensor; based on the end time of each ERT measurement, the TDR measurement results are linearly interpolated; through this process, the inverted resistivity, soil volumetric moisture content and soil temperature data at the same spatial location and the same time point can be obtained.

[0017] A further technical solution is to use the commercial software Res2DInv in step 2 to perform terrain and temperature correction, soil moisture inversion and other processing on the resistivity data.

[0018] A further technical solution is that the specific processing process of step 2 is:

[0019] Step 1. First, add terrain elevation data: open the original data file ending with "rd.dat" and start entering elevation information after the last line of resistivity data;

[0020] Step 2: Then open the Res2DInv software and set the inversion parameters;

[0021] Step 3: After the inversion parameters are set successfully, start the inversion using the software;

[0022] Step 4: After the inversion is completed, open the inversion file and manually remove the data with a data mismatch percentage exceeding 100. After removal, you need to save a new resistivity data file and then perform a second inversion with the same inversion parameters to obtain the final inverted resistivity distribution result.

[0023] A further technical solution is that the step 3 adopts the least squares inversion algorithm, and the mathematical expression is as follows:

[0024] (J T J+λF)Δq k =J T g-λFq k

[0025]

[0026] Where: C x is the horizontal roughness filter coefficient, C z is the vertical roughness filter coefficient, J is the Jacobian matrix of partial derivatives, J T is the transposed matrix of J, λ is the damping coefficient, q is the model change vector, and g is the data mismatch vector.

[0027] A further technical solution is that the specific process of step three is:

[0028] Step S1: firstly perform temperature correction on the inverted resistivity;

[0029] Step S2: Based on the long-term inversion resistivity and soil volumetric water content data, a power function fitting method is used to construct a quantitative relationship model between the temperature-corrected inversion resistivity and the soil volumetric water content.

[0030] A further technical solution is that the temperature correction formula in step S1 is:

[0031] ρ0=ρ[1+δ(T-25)]

[0032] Where: ρ0 is the inverted resistivity correction value at the standard temperature of 25℃; ρ is the inverted resistivity value at the soil temperature T; δ is the temperature correction coefficient, which is 0.025.

[0033] The beneficial effects of the present invention are as follows: the present invention reveals the coupling law between the dynamic transport process of subsoil flow and surface runoff-erosion under specific rainfall events through the spatiotemporal correlation analysis of ERT continuous monitoring and runoff volume data at the outlet of the runoff plot, breaking through the static and fragmented constraints of traditional methods on mechanism research; and focuses on overcoming the technical bottlenecks of subsoil flow dynamic path identification and quantitative analysis through structural innovation of multi-source data fusion and monitoring system, providing high-precision, non-invasive monitoring support for the study of soil and water loss mechanism and the construction of ecological safety barriers in mountainous areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the runoff plot monitoring system;

[0035] Figure 2 This is a diagram of the instrument layout;

[0036] Figure 3 is the inverted resistivity distribution result. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0038] The present invention provides a method for monitoring mid-slope soil flow by combining geophysical detection with a soil moisture sensor network, comprising the following steps:

[0039] Step 1: Use Figure 1 The monitoring system shown monitors the target runoff area and obtains rainfall data, resistivity spatial dynamic distribution data, soil volume moisture content data and soil temperature data at multiple depths and slopes, and multi-depth runoff data of the target runoff area;

[0040] The monitoring system includes four instruments;

[0041] The first type is a rain gauge, located Figure 1 The upper left panel is used to measure rainfall within the runoff area. A tipping bucket rain gauge is installed in this runoff area.

[0042] The second type is electrical resistivity tomography (ERT), which monitors soil resistivity using electrodes buried in the ground (red dots indicate electrode locations). Two ERT lines were installed in this runoff plot, each 14.5 meters long and containing 30 electrodes, with a 0.5-meter spacing between them. They use a Wenner-Schlumber setup.

[0043] The third type is time-domain reflectometry (TDR), which monitors soil volumetric moisture and temperature using sensors buried at varying depths. Three TDRs were installed in this runoff plot, located at the upper, middle, and lower slopes of the plot. The probes for monitoring soil volumetric moisture and temperature were located at depths of 10, 20, 30, 40, and 50 cm.

[0044] The fourth type is a flow meter, located Figure 1 This instrument is standard equipment for runoff plots and is used to measure runoff at different depths of the runoff plot outlet section. The flowmeters in this runoff plot are installed at depths of 20, 60, and 100 cm at the outlet section.

[0045] Figure 2 The specific placement of the TDR sensor in the soil profile is presented. This paper uses trigonometric linear interpolation to interpolate the inverted resistivity values ​​at the TDR sensor locations. Based on the end time of each ERT measurement, the TDR measurement results (soil volumetric moisture content and soil temperature) are linearly interpolated. This process allows for the acquisition of inverted resistivity, soil volumetric moisture content, and soil temperature data at the same spatial location and time point, laying the foundation for the subsequent establishment of a quantitative relationship between inverted resistivity and soil volumetric moisture content.

[0046] Step 2: Process the inverted resistivity data;

[0047] First, add terrain elevation data. Open the raw data file ending in "rd.dat" and, after the last line of resistivity data, begin entering elevation information: ① First line: 1 represents horizontal distance, 2 represents slope distance; ② Second line: Number of terrain elevation points; ③ From the third line onwards: Enter each terrain elevation point in the order of "distance" and "elevation." Enter four lines of "0" as the terminator after the end, and save the data.

[0048] Then open the Res2DInv software and set the inversion parameters.

[0049] In terms of terrain correction, the distorted finite element method with a terrain attenuation factor of 0.75 is selected. Since the resistivity data has already been input with elevation information, the software's inversion numerical method will automatically select the finite element method, which is more suitable for hillside environments with undulating terrain than the finite difference method. In terms of the inversion data constraint type, enhanced constraints are selected rather than smooth constraints, because enhanced constraints use L1 norm regularization, and through sparsity constraints, some parameters are allowed to be zero, retaining the edges and mutation features of the geological body, making the generated resistivity distribution map clearer. Due to the fragmentation of the hillside surface and the large heterogeneity of the soil resistivity distribution, the model refinement setting is enabled, that is, when the inversion model is established, the width of the model is half the electrode spacing. This setting can optimize the inversion results of the resistivity in the near-surface area and improve data reliability.

[0050] After the above settings are successful, the software is used to start the inversion. The software uses the least squares inversion algorithm, and the mathematical expression is as follows:

[0051] (J T J+λF)Δq k =J T g-λFq k

[0052]

[0053] Where: C x is the horizontal roughness filter coefficient, C z is the vertical roughness filter coefficient, J is the Jacobian matrix of partial derivatives, J T is the transposed matrix of J, λ is the damping coefficient, q is the model change vector, and g is the data mismatch vector.

[0054] After the inversion is completed, open the inversion file and manually remove the data with a data mismatch percentage exceeding 100. After removal, you need to re-save a new resistivity data file and then perform a second inversion with the same inversion parameters to obtain the final inversion resistivity distribution result (such as Figure 3 shown).

[0055] Step 3: Establish a quantitative relationship model between resistivity and soil moisture content based on the processed inverted resistivity data and soil volume moisture content data;

[0056] First, the inverted resistivity is temperature corrected;

[0057] ρ0=ρ[1+δ(T-25)]

[0058] Where: ρ0 is the inverted resistivity correction value at the standard temperature of 25℃; ρ is the inverted resistivity value at the soil temperature T; δ is the temperature correction coefficient, which is 0.025;

[0059] Based on long-term (recommended at least 1 month) inverted resistivity and soil volumetric moisture content data, a power function fitting method is used to construct a quantitative relationship model between the temperature-corrected inverted resistivity and soil volumetric moisture content. On this basis, the soil volumetric moisture content is backcalculated based on the temperature-corrected inverted resistivity.

[0060] In order to verify the reliability of the model, it is necessary to select measured soil water data outside the modeling period as an independent validation set, and evaluate the accuracy and effect of the model by calculating the coefficient of determination and root mean square error.

[0061] Step 4: Determine rainfall data and resistivity data for identifying soil flow paths and dynamic change processes;

[0062] Check the rain gauge data (hourly data, for example) and select the start and end times of the rainfall event. The criteria are as follows: During periods without rainfall, the rain gauge reading is always 0 mm. After rainfall occurs, the rain gauge reading is significantly greater than 0 mm. The rain gauge reading at the start of the rainfall event must not be 0 mm, and the reading must be always 0 mm for at least 12 hours before this time. The rain gauge reading at the end of the rainfall event must not be 0 mm, and the reading must be always 0 mm for at least 12 hours after this time. The duration of the entire rainfall event must be at least 4 hours.

[0063] Since it is assumed that the rain gauge data is at the hourly level and the TDR data is at the 10-minute level, the TDR data (soil moisture content and soil temperature) within 1 hour before and after the start of the rainfall event are checked, and the moment when the soil moisture content begins to increase is taken as the actual start time of the rainfall event. The same is true for the end time of the rainfall event. The TDR data within 1 hour before and after the end time are checked, and the moment when the soil moisture content begins to decrease is taken as the actual end time of the rainfall event.

[0064] The ERT data is hourly. The ERT data that is rounded down to 6 hours before the actual start time of the rainfall event is selected as the t0 data (for example, if the actual start time is 12:10 or 12:50, the t0 data time is 6:00). The ERT data that is rounded down to 1 hour is the t1 data. The data of all times within the rainfall event are deduced in this way. The ERT data that is rounded up to 1 hour after the actual end time of the rainfall event is selected as the t n Data (e.g., the actual end time is 18:10 or 18:50, and the time of tn data is 19:00), round up the data of 6 hours and 24 hours as t n+1 Data and t n+2 data.

[0065] The runoff meter data from 12 h before the start time to 48 h after the end time of the rainfall event were selected as auxiliary reference.

[0066] Step 5: Obtain the spatial distribution dynamics of soil moisture in the monitoring range at different times through time-lapse geophysical scanning and inversion calculation;

[0067] Set the inversion parameters: For terrain correction, select the distorted finite element method with a terrain attenuation factor of 0.75, the number of iterations is 3, the finite element method is selected for the inversion numerical method, the enhanced constraint is selected for the inversion data constraint type, the model refinement setting is enabled, and the extended model setting is disabled.

[0068] Perform inversion calculation: generate the inversion result of the relative resistivity change △ρ / ρ0, where ρ0 corresponds to the inverted resistivity value of the t0 data.

[0069] Through the inversion result display function of Res2DInv, a percentage change model of the resistivity of the inversion results was established to identify the resistivity drop area, and the drop threshold was △ρ / ρ0≤-12% as the potential soil flow path;

[0070] Step 6: Verify the inversion results by dynamic water content verification, point scale verification, and soil outflow flow verification;

[0071] Dynamic verification of moisture content: Substitute the resistivity change △ρ obtained by inversion into the quantitative relationship model between resistivity and soil moisture content established in step 3, calculate the moisture content change rate θ(t) of each spatial node, construct a two-dimensional soil moisture change model, and verify the soil flow path.

[0072] Point scale verification: Extract the model of the spatial coordinate point corresponding to the TDR sensor to calculate the water content θ model , and the error analysis was performed with the TDR measured value θTDR, using the root mean square error RMSE≤0.05cm 3 cm -3 as a verification standard.

[0073] Verification of outflow flow in soil: The total runoff Q monitored by the flow meter obs The soil flow flux Q calculated by the model cal Perform mass balance calibration and require the absolute value of relative error |(Q obs -Q cal ) / (Q obs |≤10%.

[0074] Step 7: Visualize the dynamic process based on the inversion results to achieve in-situ high-precision monitoring of the soil flow path and dynamic change process on the slope during rainfall events;

[0075] Construct a spatiotemporal visualization model: Parameters such as resistivity changes, water content distribution, and dynamic changes in soil flow paths within the selected rainfall event are integrated and displayed by drawing inversion results.

[0076] Generate a set of dynamic process characteristic parameters, including but not limited to: subsoil flow initiation time (the time difference between the start of rainfall and the appearance of the first path), subsoil flow path development rate, maximum infiltration depth, lateral expansion range, etc.

[0077] Based on the three verification results in step six, as well as the visual mapping of resistivity changes, moisture content distribution, and soil flow paths, the effects of the soil flow paths and dynamic change processes identified in the selected rainfall events are evaluated.

[0078] The above description does not limit the present invention in any form. Although the present invention has been disclosed through the above embodiments, it is not intended to limit the present invention. Any technician familiar with the profession can use the technical content disclosed above to make some changes or modifications to equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are within the scope of the technical solution of the present invention.

Claims

1. A method for monitoring mid-slope soil flow using a combination of geophysical detection and a soil moisture sensor network, characterized in that: The following steps are involved: Step 1: Obtain rainfall data, resistivity data, spatial dynamic distribution data, soil volume moisture content data and soil temperature data at multiple depths and slopes, and multi-depth runoff data for the target runoff area; Step 2: Process the resistivity data by terrain and temperature correction, soil moisture inversion, etc. Step 3: Establish a quantitative relationship model between resistivity and soil moisture content; Step 4: Determine rainfall data and resistivity data for identifying soil flow paths and dynamic change processes; Step 5: Obtain the spatial distribution dynamics of soil moisture in the monitoring range at different times through time-lapse geophysical scanning and inversion calculation; Step 6: Verify the inversion results by dynamic water content verification, point scale verification, and soil outflow flow verification; Step 7: Visualize the dynamic process output based on the inversion results to achieve in-situ high-precision monitoring of the slope soil flow path and dynamic change process during rainfall events.

2. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 1, characterized in that: In the step 1, rainfall data, resistivity data, soil volumetric water content data, soil temperature data, and runoff data are obtained through a rain gauge, a resistivity tomography instrument, a time domain reflectometry system, and a flow meter.

3. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 2, characterized in that: In the step 1, a trigonometric linear interpolation method is used to interpolate the inverted resistivity value at the location of the TDR sensor; based on the end time of each ERT measurement, the TDR measurement result is subjected to linear interpolation processing; Through this process, the inverted resistivity, soil volumetric moisture content and soil temperature data at the same spatial location and time point can be obtained.

4. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 1, characterized in that: In the second step, the resistivity data is subjected to terrain and temperature correction and soil moisture inversion post-processing using the commercial software Res2DInv.

5. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 4, characterized in that: The specific processing process of step 2 is as follows: Step 1. First, add terrain elevation data: open the original data file ending with "rd.dat" and start entering elevation information after the last line of resistivity data; Step 2: Then open the Res2DInv software and set the inversion parameters; Step 3: After the inversion parameters are set successfully, start the inversion using the software; Step 4: After the inversion is completed, open the inversion file and manually remove the data with a data mismatch percentage exceeding 100. After removal, you need to save a new resistivity data file and then perform a second inversion with the same inversion parameters to obtain the final inverted resistivity distribution result.

6. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 5, characterized in that: The step 3 adopts the least squares inversion algorithm, and the mathematical expression is as follows: (J T J+λF)Δq k =J T g-λFq k Where: C x is the horizontal roughness filter coefficient, C z is the vertical roughness filter coefficient, J is the Jacobian matrix of partial derivatives, J T is the transposed matrix of J, λ is the damping coefficient, q is the model change vector, and g is the data mismatch vector.

7. The method for monitoring mid-slope soil flow by combining geophysical detection and soil moisture sensor network according to claim 1, characterized in that: The specific process of step three is: Step S1: firstly perform temperature correction on the inverted resistivity; Step S2: Based on the long-term inversion resistivity and soil volumetric water content data, a power function fitting method is used to construct a quantitative relationship model between the temperature-corrected inversion resistivity and the soil volumetric water content.

8. The method for monitoring mid-slope soil flow by combining geophysical detection with a soil moisture sensor network according to claim 7, characterized in that: The temperature correction formula in step S1 is: ρ0=ρ[1+δ(T-25)] Where: ρ0 is the inverted resistivity correction value at the standard temperature of 25℃; ρ is the inverted resistivity value at the soil temperature T; δ is the temperature correction coefficient, which is 0.025.

Images