A method for estimating organic carbon in regional deep soil profiles
By simulating the negative exponential function parameters in the topsoil organic carbon and establishing prediction equations, the problems of high estimation cost and difficulty in sampling in deep soil organic carbon are solved, and the accurate estimation of deep soil organic carbon in regional areas is achieved.
Patent Information
- Application Number
- CN202210574690.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-24
- Publication Date
- 2025-03-18
- Estimated Expiration
- 2042-05-24
AI Technical Summary
The prior art has high costs and difficult sampling work to implement when estimating the organic carbon storage in deep soil, resulting in the inability to expand to the regional scale.
By simulating the negative exponential function parameters of the topsoil organic carbon at a depth of 0-40 cm, screening the predictors, and establishing the prediction equations of the negative exponential function parameters, combining the negative exponential function, the deep soil organic carbon at the regional scale is estimated.
This method can accurately estimate the organic carbon content of deep soil profiles, reduce the cost and difficulty of sampling work, and verify its applicability in other areas.
Smart Images
Figure CN115078687B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of soil organic carbon estimation methods, and in particular relates to a method for estimating organic carbon in a regional deep soil profile. Background Art
[0002] Soil carbon storage accounts for about 66%-75% of the global terrestrial ecosystem carbon storage, of which 55% of organic carbon is stored in deep soil (>1m). In previous studies, deep organic carbon has often been neglected. However, estimating the organic carbon storage of soil profiles, especially deep soil organic carbon storage, is of great significance to fully explore the emission reduction and carbon sink benefits of soil carbon pools. Therefore, deep soil should be taken into account in soil organic carbon estimation. Accurately estimating the organic carbon storage of the entire soil profile is crucial to understanding regional soil organic carbon storage and carbon sink potential, and further proposing scientific and reasonable regional soil carbon sink management measures.
[0003] In the current deep soil organic carbon estimation research, the deep soil organic carbon estimation mainly adopts the depth distribution function, such as negative exponential function, logarithmic function, power function, etc. This method requires the collection of soil samples of at least 3-4 soil layers at a soil depth of 0-1m to measure the soil organic carbon content to obtain the depth distribution function parameters. The regional deep soil sampling is labor-intensive, time-consuming, labor-intensive, and expensive. Therefore, this method limits the research on deep soil organic carbon at the regional scale.
[0004] From this, it can be seen that in the current estimation of deep soil organic carbon, there are cost limitations and difficulties in implementing deep sampling, which makes it impossible to expand the estimation of deep soil organic carbon to the regional scale. Summary of the invention
[0005] In view of the problems existing in the above-mentioned prior art, the present invention proposes a method for estimating organic carbon in regional deep soil profiles, which simulates negative exponential function parameters based on the topsoil organic carbon at a depth of 0-40 cm to solve the problems of high cost and difficulty in implementing sampling work that limit the estimation of organic carbon in regional deep soil.
[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:
[0007] A method for estimating organic carbon in regional deep soil profiles, comprising:
[0008] Step 1: Select an area for soil sampling, measure soil organic carbon and soil properties, fit the measured soil organic carbon data with the existing depth function that simulates the vertical distribution of soil organic carbon, and select the optimal depth distribution function that can simulate the vertical change of soil organic carbon, i.e., the negative exponential function, by comparing the fitting results;
[0009] Step 2: Based on the relationship between the surface soil properties measured by the soil samples at a depth of 0-40 cm in each profile and the negative exponential function parameters in step 1, the prediction factors of the negative exponential function parameters are screened;
[0010] Step 3: Based on the regression relationship between the prediction factors in step 2 and the parameters of the negative exponential function, an equation for predicting the parameters of the negative exponential function is established, and the equation is combined with the negative exponential function in step 1 to estimate the deep soil organic carbon at the regional scale.
[0011] Preferably, the negative exponential function of step 1 is as follows:
[0012] C x =C e +(C0-C e )×exp(-kx) (1)
[0013] Among them, C x is the soil organic carbon content at the depth of x cm soil layer (g kg -1 ), x is the soil depth (cm), C e is the soil organic carbon content when the change of deep soil organic carbon tends to be stable (g kg -1 ), k is the change rate of organic carbon in soil profile (cm -1 ), C0 is the soil organic carbon content in the x0 soil layer (g kg -1 ), x0 is defined as the top 0 cm soil layer.
[0014] Preferably, the step 2 is specifically: selecting C0 as a prediction factor to predict the negative exponential function parameter C e , C0 is the organic carbon content of the topsoil at a depth of 0-20 cm; the change rate of organic carbon in the topsoil at a depth of 0-40 cm, k0, is selected as the prediction factor to predict the negative exponential function parameter k;
[0015] The change rate of soil organic carbon in the topsoil k0 (cm -1 ) is calculated as:
[0016] k0=-(ln(C1)-ln(C0)) / (x1-x0) (2)
[0017] Where: x1 is the upper boundary of the sampling depth of the next layer of the actual surface layer; C1 is the soil organic carbon content in the x1 soil layer (g kg -1 ).
[0018] Preferably, the step 3 is specifically:
[0019] Step 3.1: Establish a prediction model using C0 as a predictor e The regression equation is:
[0020] C e=a×C0+b (3)
[0021] In the formula: a, b is C e The constant of the linear regression equation with C0;
[0022] Step 3.2: Establish a regression equation to predict k using k0 as a predictor:
[0023] k=c×k0+d (4)
[0024] Where: c, d are constants of the linear regression equation of k and k0;
[0025] Finally, we can obtain C by combining equations (3) and (4) e Substituting ρ and k into equation (1), the soil organic carbon content in the deep profile can be estimated.
[0026] Preferably, the estimation method can accurately estimate the soil organic carbon content in soil layers with a depth of 0-500 cm.
[0027] Compared with the prior art, the present invention has the following beneficial effects:
[0028] The regional-scale deep soil organic carbon estimation method described in the present invention fully considers the main factors affecting the vertical distribution of soil organic carbon, screens prediction factors through correlation analysis between 0-40cm soil properties and negative exponential function parameters, and establishes a prediction equation for the negative exponential function parameters. Using the predicted negative exponential function parameters in combination with the negative exponential function, the organic carbon content of the deep soil profile can be accurately estimated. In addition, the applicability of this method in estimating soil organic carbon in other regions has been verified. The present invention takes into account the main factors affecting the vertical distribution of soil organic carbon, and the analysis method is reasonable, which can solve the problem of estimating deep soil organic carbon in restricted areas such as the difficulty in implementing deep sampling work. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.
[0030] In the attached picture:
[0031] Figure 1 is a flow chart of the method of the present invention;
[0032] Figure 2 This is an overview of the study area of the present invention;
[0033] Figure 3 This is a comparison diagram of the depth distribution function of the present invention;
[0034] Figure 4 is a relationship diagram between the negative exponential function parameters and the prediction factors of the present invention;
[0035] Figure 5 It is the vertical distribution diagram of organic carbon in the soil profile of 0-500 cm of the present invention;
[0036] Figure 6 This is a verification diagram of the deep soil organic carbon estimation method of the present invention. DETAILED DESCRIPTION
[0037] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0038] Example:
[0039] Step 1: Review existing studies and collect the depth distribution functions currently used for soil organic carbon estimation. A total of 7 different depth distribution functions were collected, namely negative exponential function, exponential decreasing function, power function, logarithmic function, type III exponential function, first-order inverse polynomial function and modified exponential function.
[0040] The soil sampling area of the present invention is the Loess Plateau. Figure 2 As shown in the figure, 70 soil profiles were collected from the north-south transect of the Loess Plateau, 32 soil profiles were collected from specific points, and 267 soil profiles were collected from the region, and brought into the laboratory to measure soil organic carbon and soil properties. The data from the north-south transect of the Loess Plateau were used to establish a deep soil organic carbon estimation method, and the specific point and regional data were used to verify the simulation accuracy of the method. According to the cross-validation principle, the 70 soil profile data (depth 0-500cm) from the north-south transect of the Loess Plateau were randomly divided into a training set (n=56) and a validation set (n=14). The training set was used to construct a deep soil organic carbon estimation method, and the validation set was used to evaluate the accuracy of the deep soil organic carbon estimation method.
[0041] The soil organic carbon data of the training set were fitted with seven existing depth functions that simulated the vertical distribution of soil organic carbon. By comparing the fitting results, the optimal depth distribution function that can simulate the vertical change of soil organic carbon was selected.
[0042] like Figure 3 As shown in a and c, compared with other depth distribution functions, the fitting degree of the negative exponential function to the soil profile (R 2 ) is the highest and the prediction accuracy (RMSE) is the lowest, indicating that the optimal depth distribution function is a negative exponential function.
[0043] The negative exponential function is as follows:
[0044] C x =C e +(C0-C e)×exp(-kx) (1)
[0045] Among them, C x is the soil organic carbon content at the depth of x cm soil layer (g kg -1 ), x is the soil depth, C e is the soil organic carbon content when the change of deep soil organic carbon tends to be stable (g kg -1 ), k is the change rate of organic carbon in soil profile (cm -1 ), C0 is the soil organic carbon content in the x0 soil layer (g kg -1 ), x0 is defined as the upper boundary of the 0 cm soil layer, i.e., the actual sampling depth is 0-20 cm (for example: 0-20 cm).
[0046] Step 2: Based on the correlation analysis between the topsoil properties measured by the soil samples at a depth of 0-40 cm in each profile and the negative exponential function parameters in step 1, screen the predictive factors of the negative exponential function parameters. Specifically, because the topsoil organic carbon (C0) at a depth of 0-20 cm can affect C0 through bioturbation, leaching, and the transport of particulate organic carbon in the voids, e , select C0 as the predictor to predict the negative exponential function parameter C e ;according to Figure 5 According to existing studies, soil organic carbon at a depth of 0-40 cm changes significantly with the soil layer. Subsequently, the change in soil organic carbon gradually decreases and tends to be stable. Therefore, the change rate of topsoil organic carbon at a depth of 0-40 cm, k0, is selected as the prediction factor to predict the negative exponential function parameter k.
[0047] The change rate of soil organic carbon in the topsoil k0 (cm -1 ) is calculated as:
[0048] k0=-(ln(C1)-ln(C0)) / (x1-x0) (2)
[0049] Where: x1 is the upper boundary of the sampling depth of the next layer of the actual surface layer (for example: 20-40cm, defined as 20cm); C1 is the soil organic carbon content in the x1 soil layer depth (gkg -1 ).
[0050] Step 3: Based on the regression analysis of the prediction factors and negative exponential function parameters in step 2, a regression equation for predicting the negative exponential function parameters is established, and this equation is combined with the negative exponential function in step 1 to estimate the deep soil organic carbon at the regional scale. Specifically:
[0051] Step 3.1: As described in step 2: Because the topsoil organic carbon (C0) at a depth of 0-20 cm can be affected by bioturbation, leaching, and the transport of particulate organic carbon in the voids, C e, select C0 as the predictor to predict the negative exponential function parameter C e .according to Figure 4 As shown in a, we can see that C0 and C e There is a significant linear relationship (R 2 =0.61, p<0.001), and the linear equation can accurately simulate C using C0 e , (RMSE = 0.45 g kg -1 ):
[0052] C e =a×C0+b (3)
[0053] In the formula: a, b is C e The constant of the linear regression equation with C0;
[0054] Step 3.2: According to Figure 5 According to previous studies, soil organic carbon at a depth of 0-40 cm changes significantly with the soil layer, and then the change in soil organic carbon gradually decreases and tends to be stable. Therefore, the change rate of topsoil organic carbon at a depth of 0-40 cm, k0, is selected as the prediction factor to predict the parameter k of the negative exponential function. According to Figure 4b, there is a significant linear relationship between k0 and k (R 2 =0.81, p<0.001), and the linear equation can accurately simulate k using k0 (RMSE = 0.02 cm -1 ):
[0055] k=c×k0+d (4)
[0056] Where: c, d are constants of the linear regression equation of k and k0;
[0057] Finally, we can obtain C by combining equations (3) and (4) e Substituting ρ and k into equation (1), the soil organic carbon content in the deep profile can be estimated.
[0058] According to the above method, the regression constants of the Loess Plateau were obtained, and a, b, c, and d were 0.16, 0.81, 1.27, and 0.01, respectively. Using the validation data of the north-south transect of the Loess Plateau, regional sampling data (depth 0-200 cm), and specific point sampling data (depth 0-500 cm) (such as Figure 2 ) to verify the accuracy of this estimation method in predicting soil organic carbon in the region. Figure 6 ,This method can accurately simulate the soil organic carbon content at a depth of 0-500 cm and in the region.
[0059] Verification of measured soil profile data in other regions around the world:
[0060] Using soil profile organic carbon data collected in the ISRIC dataset ( Figure 2 The results are shown in the upper left box to verify whether the above results can be applied to the estimation of soil organic carbon in other regions. Figure 3 b, d, Compared with other depth distribution functions, the fitting degree of negative exponential function to soil profile (R 2 ) is the highest and the prediction accuracy (RMSE) is the lowest, indicating that the negative exponential function is the optimal depth distribution function that can accurately describe the vertical distribution of soil organic carbon in other regions. Figure 4 c, C0 and C e The linear regression relationship between them is also applicable to other regions. However, the regression coefficients of different regions are significantly different, indicating that the regression constant should be based on the C established by the data of the study area. e The linear relationship with C0 is obtained. Figure 4 d, k0 and k are also applicable to other regions. However, the regression coefficients of different regions are significantly different, indicating that the regression constant should be obtained based on the linear relationship between k0 and k established based on the data of the study area.
[0061] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A method for estimating organic carbon in a regional deep soil profile, characterized by: include: Step 1: Select an area for soil sampling, measure soil organic carbon and soil properties, fit the measured soil organic carbon data with the existing depth function that simulates the vertical distribution of soil organic carbon, and select the optimal depth distribution function that can simulate the vertical change of soil organic carbon, i.e., the negative exponential function, by comparing the fitting results; The negative exponential function of step 1 is as follows: C x =C e +(C0-C e )×exp(-kx) (1) Among them, C x is the soil organic carbon content at the depth of x cm soil layer (g kg -1 ), x is the soil depth (cm), C e is the soil organic carbon content when the change of deep soil organic carbon tends to be stable (g kg -1 ), k is the change rate of organic carbon in soil profile (cm -1 ), C0 is the soil organic carbon content in the x0 soil layer (g kg -1 ), x0 is defined as the top 0 cm soil layer; Step 2: Based on the relationship between the surface soil properties measured by the soil samples at a depth of 0-40 cm in each section and the negative exponential function parameters in step 1, the prediction factors of the negative exponential function parameters are screened; the specific step 2 is: selecting C0 as the prediction factor to predict the negative exponential function parameter C e , C0 is the organic carbon content of the topsoil at a depth of 0-20 cm; the change rate of organic carbon in the topsoil at a depth of 0-40 cm, k0, is selected as the prediction factor to predict the negative exponential function parameter k; The change rate of soil organic carbon in the topsoil k0 (cm -1 ) is calculated as: k0=-(ln(C1)-ln(C0)) / (x1-x0) (2) Where: x1 is the upper boundary of the sampling depth of the next layer of the actual surface layer; C1 is the soil organic carbon content in the x1 soil layer depth (gkg -1 ); Step 3: Based on the regression relationship between the prediction factors in step 2 and the negative exponential function parameters, an equation for predicting the negative exponential function parameters is established, and the equation is combined with the negative exponential function in step 1 to estimate the deep soil organic carbon at the regional scale; The step 3 is specifically as follows: Step 3.1: Establish a prediction model using C0 as a predictor e The regression equation is: C e =a×C0+b (3) In the formula: a, b is C e The constant of the linear regression equation with C0; Step 3.2: Establish a regression equation to predict k using k0 as a predictor: k=c×k0+d (4) Where: c, d are constants of the linear regression equation of k and k0; Finally, we can obtain C by combining equations (3) and (4) e Substituting ρ and k into equation (1), the soil organic carbon content in the deep profile can be estimated.
2. A method for estimating organic carbon in a regional deep soil profile according to any one of claim 1, characterized in that: The estimation method can accurately estimate the soil organic carbon content in the soil layer with a depth of 0-500 cm.