Method for lithospheric effective elastic thickness inversion based on intelligent truncation and inflection point matching

Abstract

This invention discloses a method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching, applied to a processor. The method includes the following steps: reading gridded topographic data and Bouguer gravity data stored on a hard disk or in memory to obtain the observation curve of coherence as a function of wavenumber; obtaining the monotonically decreasing coherence spectrum after truncation; solving for the logarithmic wavenumber corresponding to the coherence value at a preset value, denoted as the observed inflection point logarithmic wavenumber; establishing a mapping relationship between the theoretical inflection point logarithmic wavenumber and the effective elastic thickness; obtaining the effective elastic thickness of the lithosphere at the target grid point; traversing all grid points within the study area to generate a full-grid effective elastic thickness distribution array, thus obtaining the inversion result of the effective elastic thickness of the lithosphere. This invention significantly improves computational efficiency, achieves robust noise suppression while preserving the effective signal, and exhibits strong noise robustness.

Description

Technical Field

[0001] This invention relates to the field of electrical digital processing, and in particular to a method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching. Background Technology

[0002] Effective elastic thickness (Te) of the lithosphere is a key parameter characterizing the mechanical strength of the lithosphere. It reflects the lithosphere's ability to resist bending deformation under long-term geological loads and is of great indicative significance for understanding the evolution of foreland basins in orogenic belts, craton stability, earthquake distribution, tectonic-thermal events, and plate boundary interactions. Since the coherence between topographic data and Bouguer gravity data was first proposed, this parameter has become the mainstream parameter for estimating Te because it is less affected by deep loads and the characteristic wavelengths from high coherence value 1 to low coherence value 0 can directly constrain the lithospheric flexural behavior.

[0003] To solve for coherence, the terrain data and Bouguer gravity data are first processed using two-dimensional Fourier transform or wavelet transform to obtain the coherence spectrum. The basic principle of traditional coherence-based methods for estimating Te is to calculate the theoretical coherence curves of the elastic thin plate flexure model under different Te values, and then use least squares fitting of the objective function to obtain the optimal Te.

[0004] However, traditional techniques have the following drawbacks: (1) Full-spectrum fitting requires dozens to hundreds of iterations for each spatial window, resulting in low computational efficiency when processing large-scale grid data; (2) The inversion results are unstable and sensitive to initial values, affecting the reliability of Te estimation; (3) In actual observations, coherence often exhibits irregular oscillations in the high wavenumber region (corresponding to short wavelength bands), and traditional least-squares fitting is easily affected by this noise, reducing the estimation accuracy of Te. Therefore, a fast and robust digital inversion method is urgently needed. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching.

[0006] The objective of this invention is achieved through the following technical solution: A first aspect of the present invention provides a method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching, applied to a processor, comprising the following steps: Step S1: Read the gridded terrain data and Bouguer gravity data stored on the hard disk or in memory, and use two-dimensional continuous wavelet transform to calculate the observation coherence of each grid point to obtain the observation curve of coherence as a function of wave number. Step S2: For the observation curves of the corresponding grid points obtained in Step S1, the non-physical rising part of the high wavenumber segment in the observation curves is automatically identified and removed by using cubic polynomial fitting and iterative truncation in a semi-logarithmic coordinate system to obtain the truncated monotonically decreasing coherence spectrum. Step S3: Perform cubic polynomial fitting again on the truncated monotonically decreasing coherence spectrum obtained in step S2 in a semi-logarithmic coordinate system to solve for the logarithmic wavenumber corresponding to the coherence value when it is preset, and record it as the logarithmic wavenumber of the observation inflection point. Step S4: Based on the elastic plate deflection model, pre-calculate the theoretical coherence spectrum corresponding to a series of effective elastic thickness values, extract the theoretical inflection point log wavenumber corresponding to each effective elastic thickness value using the same method as in step S3, and establish the mapping relationship between the theoretical inflection point log wavenumber and the effective elastic thickness. Step S5: Substitute the observed inflection point logarithmic wavenumber obtained in step S3 into the theoretical inflection point logarithmic wavenumber and into the mapping relationship established in step S4. The effective elastic thickness of the lithosphere at the target grid point is obtained by one-dimensional interpolation. Step S6: Traverse all grid points within the study area and repeat steps S2 to S5 to generate a full-grid effective elastic thickness distribution array, thus obtaining the inversion result of the effective elastic thickness of the lithosphere.

[0007] Furthermore, step S2 specifically includes: Select data points whose coherence in the grid points described in step S1 is within the range of a preset lower threshold to a preset upper threshold, and construct an initial analysis sequence; In a semi-logarithmic coordinate system, a cubic polynomial is fitted to the current sequence, and the coefficients of the fitted polynomial are determined by the least squares method. The minimum point of the fitted curve is then calculated. If the minimum point is located within the current sequence, the data from that minimum point to a higher wavenumber range is truncated and deleted. Repeat the above steps of fitting, calculating the minimum point, and truncating and deleting until the minimum point is located at the highest wavenumber end of the current sequence to obtain the truncated monotonically decreasing coherence spectrum.

[0008] Furthermore, the coherence value mentioned in step S3 is a preset value, specifically 0.5.

[0009] Furthermore, in step S4, a mapping relationship is established between the theoretical inflection point logarithmic wavenumber and the effective elastic thickness. Specifically, a cubic polynomial is used to fit the functional relationship between the theoretical inflection point logarithmic wavenumber and the logarithm of the effective elastic thickness.

[0010] Furthermore, in step S4, a series of theoretical coherence spectra corresponding to effective elastic thickness values ​​are pre-calculated, using a geometric series with values ​​ranging from 1 km to 256 km.

[0011] Furthermore, the input parameters for the elastic plate deflection model in step S4 include: crustal thickness, crustal density, mantle density, ratio of deep load to total load, and phase angle.

[0012] Furthermore, the two-dimensional continuous wavelet transform described in step S1 employs complex wavelet basis functions.

[0013] Furthermore, the full-grid effective elastic thickness distribution array generated in step S6 is output as a contour map for visualization analysis of the spatial distribution of lithospheric mechanical strength.

[0014] The beneficial effects of this invention are: In an exemplary embodiment of the present invention, the monotonic mapping relationship between the preset inflection point wavenumber and the effective elastic thickness in the coherence curve is utilized to simplify the nonlinear fitting problem, which requires tens to hundreds of iterations in traditional methods, into one-dimensional interpolation. Furthermore, the inversion relies solely on the coherence inflection point wavenumber as the inversion feature, avoiding the multi-parameter coupling problem in traditional methods, resulting in stable and reproducible inversion results. Compared to traditional methods that require tens to hundreds of iterations for each window, this exemplary embodiment only requires one inflection point extraction and one interpolation to complete the inversion, significantly improving computational efficiency.

[0015] Meanwhile, by using cubic polynomial fitting, iterative identification, and automatic truncation of the oscillating portion of the high-wavenumber coherence curve, the inversion results become insensitive to noise in the measured data. This is because cubic polynomial fitting can extract the overall trend of the coherence curve, while iterative truncation gradually strips away the noise-dominated high-wavenumber segment without modifying the original coherence values ​​throughout the process. Thus, it achieves robust noise suppression while preserving the effective signal, exhibiting strong noise robustness.

[0016] Furthermore, since the inversion relies on only one characteristic quantity, the inflection point wavenumber, it avoids solution instability caused by multi-parameter coupling, and the inversion results have better repeatability and robustness. Attached Figure Description

[0017] Figure 1 A flowchart of a lithosphere effective elastic thickness inversion method based on intelligent truncation and inflection point matching, provided as an exemplary embodiment of the present invention; Figure 2 A comparison chart of coherence curves for synthesized data at Te=80 km, provided as an exemplary embodiment of the present invention; Figure 3 A comparison chart of coherence curves for synthesized data testing at Te=20 km, provided as an exemplary embodiment of the present invention; Figure 4 A schematic diagram of the actual terrain of the North American continent provided for an exemplary embodiment of the present invention; Figure 5 Provided as an exemplary embodiment of the present invention Figure 4 The corresponding Bouguer gravity anomaly diagram; Figure 6 The adoption provided for an exemplary embodiment of the present invention Figure 1 Te distribution plot obtained by method inversion; Figure 7 The Te distribution map is provided as an exemplary embodiment of the present invention and is obtained by inversion using a conventional method. Detailed Implementation

[0018] The technical solution 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, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0019] See Figure 1 , Figure 1 The flowchart illustrates an exemplary embodiment of the present invention, showing a method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching, applied to a processor, and including the following steps: Step S1: Read the gridded terrain data and Bouguer gravity data stored on the hard disk or in memory, and use two-dimensional continuous wavelet transform to calculate the observation coherence of each grid point to obtain the observation curve of coherence as a function of wavenumber. Step S2: For the observation curves of the corresponding grid points obtained in Step S1, the non-physical rising part of the high wavenumber segment in the observation curves is automatically identified and removed by using cubic polynomial fitting and iterative truncation in a semi-logarithmic coordinate system to obtain the truncated monotonically decreasing coherence spectrum. Step S3: Perform cubic polynomial fitting again on the truncated monotonically decreasing coherence spectrum obtained in step S2 in a semi-logarithmic coordinate system to solve for the logarithmic wavenumber corresponding to the coherence value when it is preset, and record it as the logarithmic wavenumber of the observation inflection point. Step S4: Based on the elastic plate deflection model, pre-calculate the theoretical coherence spectrum corresponding to a series of effective elastic thickness values, extract the theoretical inflection point log wavenumber corresponding to each effective elastic thickness value using the same method as in step S3, and establish the mapping relationship between the theoretical inflection point log wavenumber and the effective elastic thickness. Step S5: Substitute the observed inflection point logarithmic wavenumber obtained in step S3 into the theoretical inflection point logarithmic wavenumber and into the mapping relationship established in step S4. The effective elastic thickness of the lithosphere at the target grid point is obtained by one-dimensional interpolation. Step S6: Traverse all grid points within the study area and repeat steps S2 to S5 to generate a full-grid effective elastic thickness distribution array, thus obtaining the inversion result of the effective elastic thickness of the lithosphere.

[0020] Specifically, in this exemplary embodiment, the traditional full-spectrum nonlinear iterative fitting problem is transformed into a one-dimensional interpolation problem based on the inflection point wavenumber.

[0021] In the specific implementation, the computer system's processor first reads the gridded terrain data and Bouguer gravity data stored on the hard disk or in memory, calls the two-dimensional continuous wavelet transform function to perform frequency domain transformation on the terrain and gravity data of each grid point, obtains complex wavelet coefficients, and calculates the curve of observation coherence as a function of wavenumber based on these coefficients (step S1). For the non-physical oscillations caused by noise in the high wavenumber range of this curve, the processor performs cubic polynomial fitting in a semi-logarithmic coordinate system and automatically identifies cutoff points through iteration, discarding the oscillating parts to obtain an effective coherence sequence that retains only a monotonically decreasing trend (step S2). Next, the processor performs cubic polynomial fitting again on this effective sequence to solve for the logarithmic wavenumber corresponding to the preset coherence value as the observation inflection point (step S3).

[0022] On the theoretical model side, the processor pre-calculates a series of theoretical coherence spectra corresponding to theoretical Te values ​​based on input parameters such as crustal thickness and density using an elastic plate flexure model. It then establishes a mapping function between the theoretical inflection point logarithmic wavenumber and Te using the same truncation and inflection point extraction methods (step S4). Finally, the observed inflection points are substituted into this mapping function, and the Te value of the target point is quickly obtained through one-dimensional interpolation (step S5). This process is repeated to traverse all grid points, ultimately outputting a full-grid Te distribution array (step S6).

[0023] In summary, by utilizing the monotonic mapping relationship between the preset inflection point wavenumber and the effective elastic thickness in the coherence curve, the nonlinear fitting problem requiring tens to hundreds of iterations in traditional methods is simplified to one-dimensional interpolation. Furthermore, the inversion relies solely on the coherence inflection point wavenumber as the inversion feature, avoiding the multi-parameter coupling problem in traditional methods, resulting in stable and reproducible inversion results. Compared to traditional methods requiring tens to hundreds of iterations for each window, this exemplary embodiment only requires one inflection point extraction and one interpolation to complete the inversion, significantly improving computational efficiency.

[0024] By using cubic polynomial fitting, iterative identification, and automatic truncation of the oscillating portion of the high-wavenumber coherence curve, the inversion results become insensitive to noise in the measured data. This is because cubic polynomial fitting can extract the overall trend of the coherence curve, while iterative truncation gradually removes the noise-dominated high-wavenumber segment without modifying the original coherence values ​​throughout the process. This achieves robust noise suppression while preserving the effective signal, resulting in strong noise resistance.

[0025] Meanwhile, since the inversion relies on only one characteristic quantity, the inflection wavenumber, it avoids solution instability caused by multi-parameter coupling, and the inversion results have better repeatability and robustness.

[0026] The following content will explain the specific implementation method of each step.

[0027] More preferably, in an exemplary embodiment, in step S1, the terrain data and Bouguer gravity data are first processed using a two-dimensional Fourier transform or wavelet transform. Then, for each spatial point (x, y) (i.e., subsequent grid points), the coherence is calculated according to formula (1). γ 2 ( k ), and obtained the observation curve of coherence as a function of wavenumber.

[0028] (1) in, γ 2 ( k ) represents coherence (range 0 to 1). k For wave number, G ( k )and H ( k These are the Fourier transforms (or wavelet coefficient transforms) of Bouguer gravity data and terrain data, respectively. The symbol represents the complex conjugate, and < > represents the average over wavenumber bands or azimuth angles. The correlation here... γ 2 ( k That is, observational correlation. .

[0029] More preferably, in an exemplary embodiment, step S2 specifically includes: Select data points whose coherence in the grid points described in step S1 is within the range of a preset lower threshold to a preset upper threshold, and construct an initial analysis sequence; In a semi-logarithmic coordinate system, a cubic polynomial is fitted to the current sequence, and the coefficients of the fitted polynomial are determined by the least squares method. The minimum point of the fitted curve is then calculated. If the minimum point is located within the current sequence, the data from that minimum point to a higher wavenumber range is truncated and deleted. Repeat the above steps of fitting, calculating the minimum point, and truncating and deleting until the minimum point is located at the highest wavenumber end of the current sequence to obtain the truncated monotonically decreasing coherence spectrum.

[0030] Specifically, in this exemplary embodiment, the specific implementation method of intelligent truncation in step S2 is defined as follows: The processor first selects data points from the observed coherence curves whose coherence values ​​fall within the range of a preset lower threshold to a preset upper threshold (preferably 0.05 to 0.95), and constructs an initial analysis sequence. , This range selection excludes extreme values ​​where coherence is close to 0 or 1, thus avoiding the influence of boundary effects on the fit.

[0031] Subsequently, the processor operates in a semi-logarithmic coordinate system (the horizontal axis represents the natural logarithm of the wavenumber, and the vertical axis represents the coherence value, i.e., ...). Under the following conditions, a cubic polynomial fit is performed on the current sequence. The cubic polynomial fit is in the form of: The least squares method is used to determine the four coefficients a, b, c, and d of the polynomial.

[0032] Based on the fitted polynomial, the processor calculates the minimum point of the fitted curve (e.g., the point where the first derivative is zero). If the wavenumber corresponding to the minimum point is located in the middle of the current sequence (rather than at the endpoint), the high wavenumber segment to the right of the minimum point is determined to be a noise-dominated, non-physical rise segment, and the processor truncates and deletes that segment of data.

[0033] The processor then uses the remaining sequence as the new current sequence and repeats the above-described loop of fitting, finding the minimum point, judging, and truncation until the minimum point appears at the highest wavenumber end (i.e., the right endpoint) of the current sequence. At this point, the remaining sequence has shown a strictly monotonically decreasing characteristic and is determined to be a valid coherent sequence.

[0034] In summary, the iterative truncation method in step S2 eliminates the need for manually setting the truncation wavenumber threshold, achieving fully automated noise identification and removal. Cubic polynomial fitting captures the overall trend of the coherence curve and is insensitive to local random oscillations, thus gradually stripping away noise-dominated high-frequency bands while preserving effective low-frequency signals. The entire process does not modify the original coherence values, only performing truncation and deletion, avoiding signal distortion that might be introduced by manual filtering.

[0035] More preferably, in an exemplary embodiment, in step S3, the truncated monotonically decreasing coherence spectrum is fitted again with a cubic polynomial in a semi-logarithmic coordinate system to obtain the fitting polynomial coefficients; the observed coherence is then... Substituting 0.5 into the polynomial, we solve for the corresponding logarithmic wavenumber, denoted as the logarithmic wavenumber at the observation inflection point. .

[0036] It should be noted that, in this exemplary embodiment, a coherence value of 0.5 is chosen as the inflection point because this value is located in the range where coherence decreases most steeply, is most sensitive to changes in the effective elastic thickness of the lithosphere, and is least affected by high wavenumber noise. It is the optimal threshold supported by both theory and experiment.

[0037] More preferably, in an exemplary embodiment, in step S4, the known crustal thickness is first determined... z c Earth's crust density ρ c Mantle density ρ m The ratio of deep load to total load, F; phase angle. δ Using an elastic plate deflection model, the theoretical coherence spectra corresponding to a series of effective elastic thickness values ​​are pre-calculated. The theoretical coherence spectra corresponding to a series of geometric Te values ​​(1–200 km, taken in a geometric progression, such as 1, 2, 4, 8…256 km) are then calculated. The formula for calculating the theoretical coherence spectrum is as follows: (2) In the formula, The terrain response, representing the surface load, The terrain response to underground loads, This represents the gravitational response to surface loads. This represents the gravity response to underground loads. A flexural filter representing surface load. A flexural filter representing underground load. Indicates density ratio, This represents the ratio of deep load to surface load. Indicates the density difference between the crust and mantle. Indicates stiffness, E For style modulus (take 10) 11 Nm), σ is Poisson's ratio (value 0.25), G is the gravitational constant (value 6.67 × 10⁻⁶). -6 mGal.m 2 / kg), g is the acceleration due to gravity (taken as 9.81m / s²). 2 ), k represents the wave number, z c Indicates the thickness of the Earth's crust. ρ c The value represents the density of the Earth's crust, and F represents the ratio of deep load to total load. ρ m T represents the density of the mantle. e Indicates the effective elastic thickness.

[0038] Then, the theoretical inflection point logarithmic wavenumber corresponding to each effective elastic thickness value Te is extracted using the same method as in step S3. It should be noted that this is a theoretical coherence curve, which is unaffected by noise. Therefore, only cubic polynomial fitting is needed, and truncation is not required; that is, only step S3 is needed, and step S2 is not required.

[0039] Finally, the theoretical inflection point logarithmic wavenumber was established through cubic polynomial fitting. Logarithm of effective elastic thickness Functional relationship between them: .

[0040] It should be noted that: (1) The mapping relationship is established only once before the inversion begins. Subsequent inversions of all grid points directly call this mapping function for interpolation, without having to repeatedly calculate the theoretical coherence spectrum. Compared with the traditional method of performing dozens of theoretical spectrum calculations and fittings at each grid point, this scheme saves a significant amount of computation time.

[0041] (2) The advantage of using a geometric series is that in the small Te range (e.g., 1-20 km), the change in lithospheric intensity has a significant impact on the inflection point wavenumber, requiring denser sampling points to ensure fitting accuracy; in the large Te range (e.g., 100-200 km), the rate of change of the inflection point wavenumber with Te slows down, allowing for sparser sampling points. The geometric series precisely meets this requirement, automatically providing higher sampling density in low-value areas and automatically reducing sampling density in high-value areas, thereby achieving efficient distribution of sample points. In other words, the small Te range is sensitive to the inflection point wavenumber.

[0042] More preferably, in an exemplary embodiment, in step S5, the observed inflection point logarithmic wavenumber obtained in step S3 is substituted into the theoretical inflection point logarithmic wavenumber and then into the mapping relationship established in step S4, and the effective elastic thickness of the lithosphere at the target grid point is obtained by one-dimensional interpolation.

[0043] In this process, after the processor obtains the logarithmic wavenumber of the observed inflection point, it needs to substitute it into the mapping relationship established in step S4 to solve for the corresponding Te value. Since the pre-calculated Te sample points are discrete, the observed inflection point usually cannot be exactly equal to the theoretical inflection point of a certain sample point. Therefore, it is necessary to obtain the accurate Te value through interpolation. Linear interpolation is the simplest method: the processor first finds two sample points adjacent to the observed inflection point in the theoretical mapping data, assumes that the inflection point and Te have a linear relationship in the local interval, and calculates the corresponding Te value proportionally.

[0044] More preferably, in an exemplary embodiment, in step S6, the full-grid effective elastic thickness distribution array generated in step S6 is output as a contour map for visualization analysis of the spatial distribution of lithospheric mechanical strength.

[0045] Specifically, after traversing all grid points within the study area and generating a full-grid Te distribution array, the processor outputs this array as a contour map or color block map for the visualization and analysis of the spatial distribution of lithospheric mechanical intensity. The contour map generation process is as follows: the processor sets several contour levels (e.g., Te = 20, 40, 60, 80, 100 km) based on the value range of the Te array, then uses a contour tracing algorithm to calculate the spatial location of each contour line in the grid, and finally draws it as lines. The contour map can intuitively display the spatial distribution characteristics of Te, helping geological interpreters quickly identify high-Te craton cores, low-Te orogenic belts, and the gradient zones between them.

[0046] To verify the effectiveness and reliability of the above exemplary embodiments, the following content will demonstrate synthetic data testing and practical applications: (1) Synthetic data testing Table 1 shows the results of the synthetic data test (where the internal load accounts for 0.33% of the total load, and Gaussian white noise with a standard deviation of 0.15 is added). The results indicate that when the true Te is high (≥ 40 km), the traditional full-spectrum fitting results are stable between 26–30 km, failing to effectively distinguish the true values ​​at 40 km, 60 km, 80 km, and even 120 km, exhibiting significant underestimation and high-value saturation. In contrast, the method of this invention maintains good resolution across the entire range, clearly distinguishing lithosphere intensities, and is particularly stable in the high-value range of 40–120 km.

[0047] Table 1. Comparison of Te inversion results between the method of this invention and the traditional method in synthetic data testing.

[0048]

[0049] Figure 2 The diagram shows the theoretical coherence, the coherence after adding noise, and the coherence curves predicted by traditional inversion and the inversion method of this patent when the actual Te is 80 km. It can be seen that due to the influence of high wavenumber noise, traditional inversion is difficult to fit the coherence curve. When the actual Te is 10 km and 20 km, the inversion results of this invention are 8.5 km and 18.2 km, respectively, which are close to those of the traditional method (9.4 km and 20.4 km), both showing good performance. This is in... Figure 3 The graphs showing the theoretical coherence, the coherence after adding noise, and the coherence curves predicted by the traditional inversion and the inversion of this patent when the actual Te is 20km demonstrate that the coherence curves of both inversion methods are well fitted.

[0050] In summary, the method of the present invention significantly improves the ability to resolve high Te regions while maintaining the accuracy of low Te estimation, and overcomes the high-value saturation defect of traditional methods.

[0051] (2) Practical application effect To verify the effectiveness of this invention in actual data, the actual terrain of North America was used ( Figure 4 ) and Bouguer gravity data ( Figure 5 The test was conducted using a total of 116,281 grid points. The effective elastic thickness Te distribution of the lithosphere was inverted using both the method of this invention and the traditional full-spectrum least squares fitting method. The method of this invention took 18 minutes, while the traditional full-spectrum least squares fitting method took 47 minutes.

[0052] The inversion results of the method of this invention show that ( Figure 6 The Canadian Shield exhibits high Te values ​​(>100km), indicating that the ancient craton region possesses extremely strong lithospheric mechanical strength; the Cordillera orogenic belt exhibits low Te values ​​(approximately 20km) and has a steep Te gradient transition zone with the Shield.

[0053] In contrast, the results of traditional inversion methods ( Figure 7 The high-value saturation phenomenon appears after Te is about 40km. The Te values ​​in the Canadian Shield and the Cordillera orogenic belt are both about 40km, which makes it impossible to distinguish the intensity difference between the ancient craton and the young orogenic belt, which is completely inconsistent with geological understanding.

[0054] In summary, the method of this invention demonstrates good applicability and noise resistance in ancient cratons, young orogenic belts, and complex tectonic regions affected by noise, verifying its effectiveness and reliability as a rapid inversion method for the effective elastic thickness of the lithosphere.

[0055] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching, applied to a processor, characterized in that: Includes the following steps: Step S1: Read the gridded terrain data and Bouguer gravity data stored on the hard disk or in memory, and use two-dimensional continuous wavelet transform to calculate the observation coherence of each grid point to obtain the observation curve of coherence as a function of wave number. Step S2: For the observation curves of the corresponding grid points obtained in Step S1, the non-physical rising part of the high wavenumber segment in the observation curves is automatically identified and removed by using cubic polynomial fitting and iterative truncation in a semi-logarithmic coordinate system to obtain the truncated monotonically decreasing coherence spectrum. Step S3: Perform cubic polynomial fitting again on the truncated monotonically decreasing coherence spectrum obtained in step S2 in a semi-logarithmic coordinate system to solve for the logarithmic wavenumber corresponding to the coherence value when it is preset, and record it as the logarithmic wavenumber of the observation inflection point. Step S4: Based on the elastic plate deflection model, pre-calculate the theoretical coherence spectrum corresponding to a series of effective elastic thickness values, extract the theoretical inflection point log wavenumber corresponding to each effective elastic thickness value using the same method as in step S3, and establish the mapping relationship between the theoretical inflection point log wavenumber and the effective elastic thickness. Step S5: Substitute the observed inflection point logarithmic wavenumber obtained in step S3 into the theoretical inflection point logarithmic wavenumber and into the mapping relationship established in step S4. The effective elastic thickness of the lithosphere at the target grid point is obtained by one-dimensional interpolation. Step S6: Traverse all grid points within the study area and repeat steps S2 to S5 to generate a full-grid effective elastic thickness distribution array, thus obtaining the inversion result of the effective elastic thickness of the lithosphere.

2. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: Step S2 specifically includes: Select data points whose coherence in the grid points described in step S1 is within the range of a preset lower threshold to a preset upper threshold, and construct an initial analysis sequence; In a semi-logarithmic coordinate system, a cubic polynomial is fitted to the current sequence, and the coefficients of the fitted polynomial are determined by the least squares method. The minimum point of the fitted curve is then calculated. If the minimum point is located within the current sequence, the data from that minimum point to a higher wavenumber range is truncated and deleted. Repeat the above steps of fitting, calculating the minimum point, and truncating and deleting until the minimum point is located at the highest wavenumber end of the current sequence to obtain the truncated monotonically decreasing coherence spectrum.

3. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: The coherence value mentioned in step S3 is a preset value, specifically 0.

5.

4. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: In step S4, the mapping relationship between the theoretical inflection point logarithmic wavenumber and the effective elastic thickness is established. Specifically, a cubic polynomial is used to fit the functional relationship between the theoretical inflection point logarithmic wavenumber and the effective elastic thickness logarithm.

5. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: In step S4, a series of theoretical coherence spectra corresponding to effective elastic thickness values ​​are pre-calculated, using a geometric series with values ​​ranging from 1 km to 256 km.

6. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: The input parameters for the elastic plate deflection model in step S4 include: crustal thickness, crustal density, mantle density, ratio of deep load to total load, and phase angle.

7. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: The two-dimensional continuous wavelet transform described in step S1 uses complex wavelet basis functions.

8. The method for inverting the effective elastic thickness of the lithosphere based on intelligent truncation and inflection point matching according to claim 1, characterized in that: The full-grid effective elastic thickness distribution array generated in step S6 is output as a contour map for visualization analysis of the spatial distribution of lithospheric mechanical strength.

Images