A method for nondestructive estimation of root system characteristic parameters and biomass of a single plant based on ground penetrating radar

Abstract

The application discloses a kind of single plant root system characteristic parameters based on ground penetrating radar and its biomass nondestructive estimation method, comprising the following steps: 1) using ground penetrating radar to carry out nondestructive measurement to standing wood root system to obtain root system scanning original data;2) based on root system sampling data, root system positioning and depth correction are carried out;3) selecting the time interval parameter irrelevant to radar wave signal amplitude intensity estimates root system diameter;4) using the corresponding relationship between root system diameter and root system spatial position, using single plant root system growth characteristics to reconstruct three-dimensional root system configuration;5) root system is regarded as circular truncated cone model, and the geometric characteristics of root system are calculated;6) root density is determined, and root system biomass is estimated in combination with root system volume characteristic parameter.The root system geometric characteristic and root system biomass estimation method provided by the application not only effectively avoids the influence of radar wave signal amplitude intensity on estimation effect, and the method also lays a foundation for periodic nondestructive measurement of root system.

Description

Technical Field

[0001] This invention relates to a non-destructive method for estimating root characteristic parameters and root biomass based on ground-penetrating radar root scanning data of a single plant. Background Technology

[0002] Forest underground biomass is a core component of carbon allocation in forest ecosystems. Root systems, as a major component of forest underground biomass, not only provide nutrients and water to the above-ground parts of plants but also play a crucial role in maintaining the diversity and stability of forest ecosystems. For a long time, to accurately obtain plant root biomass and map its distribution, invasive survey methods such as digging, uprooting, soil clods, and soil core sampling have been commonly used. While these methods can yield reliable quantitative measurements, they are undoubtedly labor-intensive, time-consuming, and non-repeatable. Ground Penetrating Radar (GPR) is a non-destructive geophysical detection technology that uses high-frequency electromagnetic waves to locate the burial position of underground roots. Its principle is based on the difference in dielectric properties between plant roots and soil, which causes changes in the amplitude, phase, and frequency of GPR electromagnetic waves, allowing for the interpretation of the characteristics of buried roots. Thus, using the echo signals from GPR root detection, root biomass and its characteristic parameters can be estimated, enabling convenient, rapid, and periodic measurements of plant roots.

[0003] Patent CN104570133A discloses a method for detecting tree roots using ground-penetrating radar (GPR). This method determines root depth and estimates root biomass by analyzing the strength and waveform characteristics of the echo signal. Patent CN110133643A discloses a method and device for detecting plant roots. It interprets the biological characteristics of plant roots based on information such as radar wave frequency and the intensity of the electromagnetic field of the echo signal. Furthermore, it can estimate root characteristic parameters using echo signals such as amplitude area and amplitude width. While these methods can quickly and easily estimate root parameters, their accuracy heavily depends on the amplitude intensity of the radar wave signal. Moreover, the detection quality of GPR is easily affected by noise interference from the soil environment. Therefore, methods for estimating root biomass and root characteristic parameters based on radar wave signal amplitude intensity require further improvement. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a non-destructive estimation method for root characteristic parameters and their biomass based on ground-penetrating radar root scanning data, thus avoiding the influence of radar wave signal amplitude intensity and soil noise on the accuracy of target parameter estimation.

[0005] The technical solution adopted in this invention is as follows: First, the plant root system is scanned using a square grid method based on ground-penetrating radar technology. The spatial location of the buried roots is located from the original scan results, and depth correction is performed. Second, the diameter of the root system is estimated using waveform parameters that are independent of the amplitude and intensity of the radar signal. Then, combining the spatial distribution and diameter information of the root system, a three-dimensional root system configuration is reconstructed based on the growth characteristics of individual plant roots. Finally, the root system is treated as a frustum-shaped model, thereby estimating the characteristic parameters and biomass of the root system. The detailed steps of the method are as follows: 1) Root scanning scheme: Centered on the root collar of the tree to be tested, two scanning points are set up in both horizontal and vertical directions. A square grid of measuring lines, 2-4 meters long, is used, with an interval of 0.1-0.2 meters between adjacent parallel measuring lines. The ground-penetrating radar is operated at a preset antenna frequency, moving uniformly forward along the measuring lines to scan the root system, thus obtaining raw root detection data. In actual detection, the length and interval of the measuring lines can be determined based on the tree's age and the approximate distribution of the root system.

[0006] 2) Further, the root scan data undergoes preprocessing, including signal position adjustment, time window trimming, background removal, mean filtering, and gain adjustment. Subsequently, a hyperbola is used to fit the root scan imaging results, with the apex of the hyperbola considered as the location of the buried root, thus converting the radar wave reflection signal into the three-dimensional spatial location of the root system. The purpose of the root scan image preprocessing is to focus the radar wave reflection signal onto the root reflection area, thereby achieving root localization. Therefore, although the amplitude intensity of the radar wave signal changes during processing, it does not affect the root localization.

[0007] 3) Further, depth correction is performed on the root system location results: A ground-penetrating radar with the same technical parameters as in step 1) is used to scan the reflector at the known burial depth, thereby obtaining the radar wave propagation time and distance relationship, and thus correcting the root system burial depth. This indicates the correction result for root system localization, where This represents the planar coordinates of the ground-penetrating radar locating the vertex of the root hyperbola within the square grid measurement area. The correction depth for the vertex of the root hyperbola. i This indicates the sequence number of the vertices of the detected root hyperbola.

[0008] 4) Further, the diameter of the root system is estimated: Since the time it takes for the radar wave to radially pass through the root system cross section is not related to the amplitude intensity of the radar wave signal, but only to the size of the root system diameter (the larger the root diameter, the longer the time interval between the radar wave entering and exiting the root system), the root system diameter is estimated based on the time interval parameter of the radar wave vertically and radially passing through the root system cross section, thereby obtaining a one-to-one correspondence between the spatial position of the root system and the root system diameter.

[0009] 5) Further, three-dimensional reconstruction of root system configuration: Considering that the root system of a single plant grows roughly along the previous root direction and is affected by factors such as positive gravity and soil porosity, a root system configuration reconstruction strategy is formulated to achieve three-dimensional reconstruction of the root system configuration.

[0010] 6) Further, the characteristic parameters of the root system configuration are estimated: The entire root system is considered as a combination of multiple small root segments. For each root segment element, it is regarded as a frustum model. The characteristic parameters of the root system are estimated using the formulas for the surface area and volume of a frustum. For the characteristic parameters of the complete root system, the samples of each root segment can be superimposed to obtain the characteristic parameters of the entire root system configuration.

[0011] 7) Further, the root biomass is estimated based on the root volume: the root density is obtained by measuring the relationship between the root dry weight and the root volume, and the root biomass is estimated based on the root density and the root volume calculated by the frustum model.

[0012] Based on the seven steps described above, the root system characteristic parameters and root biomass are estimated from the perspective of root system modeling, thereby avoiding the influence of radar wave signal amplitude intensity on root biomass estimation. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Furthermore, the following drawings are intended only to illustrate and explain the present invention and do not limit the scope of the invention.

[0014] Figure 1 This is a flowchart illustrating a method for non-destructive estimation of plant root characteristic parameters and their biomass according to a specific embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of a ground-penetrating radar root system square grid scanning scheme according to a specific embodiment of the present invention; Figure 3a It is the original root system image obtained by ground penetrating radar scanning along a measurement line of the square grid method at a frequency of 900 MHz; Figure 3b It uses Matgpr software to... Figure 3aThe result after performing signal position adjustment and time window trimming operations; Figure 3c It uses ReflexW software to... Figure 3b The result after background removal, mean filtering, and gain adjustment. Figure 3d Yes Figure 3c The hyperbola of the root reflection signal is fitted, and the position of the hyperbola vertex is marked and extracted to obtain the spatial location of the root system. Figure 4 This is a ground-penetrating radar root depth correction scheme according to a specific embodiment of the present invention; Figure 5 This is a schematic diagram of two time points when radar waves radially cross the root system; Figure 6 This is a root system frustum model according to a specific embodiment of the present invention; Detailed Implementation To more clearly and completely present the objectives, solutions, and advantages of this invention, the specific embodiments of this invention will be described below in conjunction with the accompanying drawings.

[0016] A flowchart illustrating a non-destructive estimation method for single plant root system characteristic parameters and biomass based on ground-penetrating radar is shown below. Figure 1 As shown, it includes the following steps: Step S101: Experimental Design and Data Acquisition. A 2.4*2.4 m square grid of measurement lines was established centered on the root collar of the tree to be tested, with 12 lines horizontally and 12 lines vertically. The distance between adjacent parallel measurement lines was 20 cm. Figure 2 As shown. The ground-penetrating radar antenna frequency was set to 900 MHz. The ground-penetrating radar was moved with the defined measurement lines as a reference to perform root system scanning, thereby obtaining the raw root system scan data. Since the root neck obstructed the two measurement lines passing through the root neck in the horizontal and vertical directions, ground-penetrating radar scanning along these two measurement lines was abandoned.

[0017] Step S102: Radar scan image preprocessing and root system localization depth correction. Based on the original root system scan image obtained in step S101, as shown below... Figure 3a As shown, to highlight the reflective effect of the root system and locate the root system, software such as Matgpr and ReflexW were used to preprocess the root scan data, including adjusting the signal position and pruning time window (e.g., ...). Figure 3b As shown), based on this, background removal, mean filtering, and gain adjustment are performed (such as...). Figure 3c As shown in the image, after preprocessing, the hyperbolic image of the root system is clearer. Subsequently, a hyperbolic model is used to fit the root system's reflected radar wave image, as shown in the image. Figure 3dAs shown by the white curve in the middle, the vertices of the hyperbola are extracted and their corresponding three-dimensional coordinates are regarded as the spatial location of the root system. Figure 3d As shown by the white dot in the middle.

[0018] Subsequently, root depth correction was performed. Three ground stakes, each 40 cm long and 1 cm in diameter, were horizontally inserted into the vertical soil profile, spaced 40 cm apart, with burial depths of 10 cm, 20 cm, and 30 cm respectively. A 900 MHz ground-penetrating radar was used to scan the three reflectors, obtaining the radar wave propagation time and distance relationship, thereby correcting the depth of the root hyperbola apex obtained in the previous steps. The implementation plan for the root burial depth correction experiment is as follows: Figure 4 As shown, the corrected root system spatial location is used It means that among them This represents the planar coordinates of the vertices of the hyperbola representing the root reflection signal. Let be the depth of the vertex of the hyperbola reflecting the root system signal. i This indicates the sequence number of the vertices of the hyperbola containing the root reflection signal.

[0019] Step S103: Root diameter estimation. The time point at which radar waves radially penetrate the root system is defined from the ground-penetrating radar A-scan image that passes perpendicularly through the apex of the hyperbola. and the timing of root emergence ,like Figure 5 As shown. The time interval parameter (i.e., the time it takes for radar waves to radially cross the root system cross section) is used. The diameter is estimated, thus avoiding the influence of radar wave signal strength on the estimation of root characteristic parameters.

[0020] Step S104: Three-dimensional reconstruction of root system configuration. Combining the spatial location of the root system obtained in step S102 and the root diameter obtained in step S103, a root system configuration reconstruction scheme is formulated. The reconstruction of the root system configuration references the root growth characteristics of a single plant, namely, that new roots grow roughly along the direction of previous roots, and that the growth of new roots is constrained by positive gravity and soil porosity.

[0021] Step S105: Calculation of root system characteristic parameters and root biomass. The entire root system is considered as a combination of multiple small root segments. For each root segment micro-element, it is treated as a frustum-shaped model, such as... Figure 6 As shown. Using the formulas for the surface area and volume of a frustum of a cone, the characteristic parameters of the root system are estimated. The calculation method for the characteristic parameters of any root segment is as follows: Root segment length l :

[0022] That is, the vertices of the hyperbola of two root system reflection signals in three-dimensional space. and The Euclidean distance between them.

[0023] Root segment lateral area s : s=π·l·(r i +r i+1 (2)

[0024] in: , The vertices of the hyperbola and The corresponding root radius.

[0025] Root segment volume v :

[0026] After obtaining the feature parameters of each root segment sample, the root segments are superimposed to form a complete root system, thereby obtaining the feature parameters of a single root system and all root systems: Total root length L :

[0027] in Indicates the first n The total length of the root system Indicates the first n The first root system m The length of each root segment; Similarly, the total lateral surface area of ​​the root system S :

[0028] Total root volume V :

[0029] Subsequently, root biomass was estimated based on root volume. First, the root density of the studied tree species was measured by selecting a certain amount of root samples and measuring the volume using the water displacement method. The dry weight was measured by the drying method. Thus, the relationship between root density, i.e., root dry weight and volume, can be derived:

[0030] in, Root density, in g / cm³ 3 .

[0031] Based on this, the root biomass is estimated:

[0032] The root system characteristic parameters and root biomass estimation results of a 9-year-old ash tree obtained through the above implementation methods are shown below: Table 1 Comparison of root characteristic parameters and root biomass estimated by ground-penetrating radar non-destructively and manually.

[0033] The root diameter was estimated and the root configuration was reconstructed in three dimensions using a time interval parameter independent of the radar wave signal amplitude intensity. The reconstructed root system was then used as a frustum model to estimate root characteristic parameters and root biomass. The results were compared with those obtained from manual root measurements. The residual percentages for total root length, total lateral area, total volume, and total biomass were 9.28%, 3.14%, 6.98%, and 8.94%, respectively, demonstrating the effectiveness and feasibility of non-destructive estimation of root geometric characteristic parameters and root biomass based on ground-penetrating radar technology.

[0034] The above description is merely an illustrative embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A non-destructive method for estimating the root system characteristics and biomass of a single plant based on ground-penetrating radar, characterized in that, Includes the following steps: S1) Set up a square grid of measurement lines centered on the root collar of the tree to be tested, and move the ground-penetrating radar to scan the root system based on the marked measurement lines to obtain the raw data of the root system scan. S2) The root system scan image is preprocessed and the root system is located, thereby converting the radar wave reflection signal into the three-dimensional spatial position of the root system. S3) Use ground-penetrating radar to scan the pre-buried reflector at a known burial depth, and analyze the radar wave velocity from the reflected signal to correct the depth of the root reflection signal obtained in step S2). S4) Based on the root reflection signal of the ground penetrating radar, select parameters that are independent of the amplitude intensity of the radar wave signal to establish a root diameter estimation model. S5) Based on the one-to-one correspondence between the spatial location of the root reflection signal and the root diameter, the three-dimensional reconstruction of the root system configuration is carried out through interpolation and fitting. S6) The geometric characteristics of the root system are calculated by assuming it to be a frustum model; S7) Root density was measured, and root biomass was estimated based on root density and root volume characteristic parameters; Step S2 includes: S2.1) Preprocessing of root scan images includes adjusting signal position, trimming time window, background removal, mean filtering, gain adjustment, etc. The processing software used is Matgpr and ReflexW. S2.2) The method for root system location is as follows: First, use a hyperbola to fit the image of the root system reflecting radar waves; second, extract the vertices of the fitted hyperbola; finally, regard the three-dimensional spatial location of the hyperbola vertex as the location where the root system is buried. Step S3) includes: S3.1) The frequency of the ground-penetrating radar antenna used is consistent with the root detection frequency. The depth of the pre-buried reflector is known. The reflector is scanned to obtain the relationship between the propagation time and distance of the radar wave, thereby correcting the burial depth of the root system. Specifically, this includes: inserting three ground nails, each 40 cm long and 1 cm in diameter, horizontally into the vertical soil profile, spaced 40 cm apart, with burial depths of 10 cm, 20 cm, and 30 cm respectively. The three reflectors are scanned using a 900 MHz ground-penetrating radar to obtain the relationship between the propagation time and distance of the radar wave. S3.2) Use This indicates the spatial location of the vertex of the root hyperbola after correction, where This represents the planar coordinates of the ground-penetrating radar locating the vertex of the root hyperbola within the square grid measurement area. Let be the depth of the vertex of the root hyperbola. i This indicates the index of the vertex of the root hyperbola that was detected; Step S5) includes: S5.1) The three-dimensional reconstruction scheme of root system configuration is formulated based on the root growth characteristics of a single plant; The growth characteristics of the root system of a single plant described in S5.2 are as follows: the growth of new roots is roughly along the direction of the previous roots, the growth of new roots is constrained by positive gravity, and the growth of new roots is constrained by soil pores. S5.3) The spatial distribution of the root system is reconstructed by connecting the root points. The two vertices of the root hyperbola connected sequentially by a root system are selected. and Between them, fitting interpolation is used to make the connection of the root system smoother and more natural.

2. The method for non-destructive estimation of root system characteristic parameters and biomass of a single plant based on ground-penetrating radar according to claim 1, characterized in that: Step S1) includes: S1.1) In the square grid method, the two types of measurement lines in different directions in the horizontal plane are orthogonal to each other. The length of a single measurement line is about 2~4 m, and the distance between adjacent parallel measurement lines is about 0.1~0.2 m. S1.2) The ground penetrating radar advances at a constant speed along the positive direction of the specified measurement line at a set antenna frequency. Each time a scan is performed, the reflection of the radar wave signal by the root system or other dielectric discontinuous media along the vertical cross section of the measurement line can be recorded in the original scan image.

3. The method for non-destructive estimation of root system characteristic parameters and biomass of a single plant based on ground-penetrating radar according to claim 1, characterized in that: Step S4) includes: The parameter selected in S4.1 that is positively correlated with the root diameter is the time interval during which radar waves pass through the root cross section vertically and radially; S4.2) The root diameter estimation model is only related to the duration of radar wave propagation in the radial direction of the root diameter, and is not related to the amplitude intensity of the radar wave signal.

4. The method for non-destructive estimation of root system characteristic parameters and biomass of a single plant based on ground-penetrating radar according to claim 3, characterized in that: Step S6) includes: S6.1) The entire root system is considered as a combination of multiple small root segments, and each root segment micro-element is considered as a frustum model. The characteristic parameters of the root system are estimated based on the formulas for the surface area and volume of the frustum. S6.2) The characteristic parameters of any root segment infinitesimal element are calculated as follows: Root segment length l : (1) That is, the vertices of the hyperbola of two root system reflection signals in three-dimensional space. and Euclidean distance between them; Root segment lateral area s : (2) in: , The vertices of the hyperbola and The corresponding root radius; Root segment volume v : (3) S6.3) The method for calculating the characteristic parameters of a single root system and all root systems is as follows: Total root length L : (4) in Indicates the first n The total length of the root system Indicates the first n The first root system m The length of each root segment; Similarly, the total lateral surface area of ​​the root system S : (5) Total root volume V : (6)。 5. The method for non-destructive estimation of root system characteristic parameters and biomass of a single plant based on ground-penetrating radar according to claim 4, characterized in that: Step S7) includes: S7.1) Root biomass W The estimation method is as follows: (7) in, Root density; S7.2) Root density is defined as the ratio of root dry weight to root volume: (8) Among them, root dry weight The root volume can be measured by the drying method. It can be measured by the water displacement method.

Images