Discriminating method for large-area scale sediment source

Abstract

The invention discloses a large-area scale sediment source discrimination method, and relates to the technical field of sediment source discrimination. The method comprises the following specific steps: dividing a large-area scale watershed into different sub-areas; sediment samples in the subareas are collected, and a sediment sediment sample solution is prepared; drawing a calibration curve based on the fingerprint identification factor library; sequentially measuring the prepared sediment sediment sample solution under a corresponding calibration curve to obtain the content of metal elements in the sediment sample, and determining the sediment fingerprint element characteristics of the sub-regions; and solving the warehousing sediment contribution rate of each sub-region by using a composite fingerprint identification technology. According to the invention, based on the principle of the least square method, the sediment contribution rate of each sediment producing area is accurately solved by using a multivariate mixed model and a composite fingerprint identification technology; reasonable dispatching of the reservoir area silt can be achieved by accurately estimating the amount of the reservoir-entering silt and the source of the reservoir-entering silt.

Description

Technical Field

[0001] The invention relates to the technical field of sediment source identification, and in particular to a method for identifying sediment sources on a large regional scale. Background Art

[0002] The sediment problem of the Three Gorges Reservoir has always been one of the focal issues of concern to all sectors of society. Finding out the amount of sediment entering the Three Gorges Reservoir and the source of the sediment entering the reservoir and accurately estimating it is an important means to effectively solve the sediment in the reservoir area and rationally dispatch the sediment in the reservoir. Due to the special topographic and geomorphic conditions and rainfall conditions in the Three Gorges Reservoir area, soil erosion and the sediment problem in the reservoir area caused by it are very serious. However, due to the small number of hydrological stations with sediment monitoring data in the reservoir area, it is difficult to estimate the amount of sediment entering the Three Gorges area and determine the source of sediment. In the prior art, the research methods on the source of sediment are mainly divided into two categories: traditional methods for studying the source of sediment and methods for studying the source of sediment using fingerprint recognition technology. Although various traditional methods for studying the source of sediment can be used to obtain general information and characteristics of the source of sediment production, each method has more or less certain shortcomings and limitations. Since the 1970s, the use of fingerprint recognition technology to quantitatively study the source of sediment has begun to be adopted by some scholars at home and abroad. Although the analysis methods are constantly updated, the experimental tools are constantly improved, and the research area is constantly expanded, systematic research methods are rare. The river sediment transport part often uses the numerical value of hydrological data, or uses experiments or water-sediment formulas to calculate the total sediment transport, resulting in errors in the calculation results. Therefore, for technical personnel in this field, how to accurately identify the source of sediment based on field survey data is an urgent problem to be solved. Summary of the invention

[0003] The purpose of the present invention is to provide a method for distinguishing the source of sediment on a large regional scale to solve the problems raised in the background technology.

[0004] To achieve the above object, the present invention provides the following solution: a method for distinguishing the source of sediment on a large regional scale, the specific steps of which include the following:

[0005] Divide the large regional scale watershed into different sub-regions;

[0006] Collecting sediment samples in the sub-area to prepare a sediment sample solution;

[0007] Draw a calibration curve based on the fingerprint identification factor library;

[0008] The prepared sediment sample solutions are sequentially measured under corresponding calibration curves to obtain the metal element content in the sediment sample and determine the sediment fingerprint element characteristics of the sub-area;

[0009] The composite fingerprint recognition technology is used to solve the sediment contribution rate of each sub-area.

[0010] Preferably, the steps of preparing the silt sediment sample solution are:

[0011] The collected sediment samples were naturally air-dried, and dead branches and leaves, crop roots and coarse gravel were manually picked out before grinding, and all of them were passed through a 100-mesh sieve;

[0012] The sieved sediment samples were weighed with a fixed mass and subjected to microwave digestion in a fixed order;

[0013] After digestion, dilute to volume with 1% nitric acid solution, mix well and wait for testing.

[0014] Preferably, the step of drawing the calibration curve is:

[0015] Based on the fingerprint identification factor library, standard solutions of different concentrations are configured;

[0016] After the standard solution is prepared, the emission intensity is measured in the order of the concentration of the mixed solution from low to high using a plasma electron coupling spectrometer;

[0017] After the measurement is completed, the calibration curve is drawn with the measured emission intensity value as the ordinate and the concentration value of the mixed solution as the abscissa.

[0018] Preferably, the calculation formula for the metal element content is:

[0019]

[0020] Among them, ω represents the content of metal elements in the sediment sample, ρ1 represents the measured concentration of metal elements in the sediment sample solution, ρ0 represents the measured concentration of metal elements in the blank solution, V0 represents the fixed volume of the sample after digestion, and m1 represents the weighed amount of the sediment sample.

[0021] Preferably, the specific steps of solving the sediment contribution rate of each sub-area by using the composite fingerprint recognition technology are as follows:

[0022] Using a multivariate discriminant analysis method, the best fingerprint factor combination is found in the sediment fingerprint element characteristics of the sub-area;

[0023] Based on the optimal fingerprint factor combination, the contribution rate of sediment entering the reservoir in each sub-area is solved through a multivariate mixed model.

[0024] Preferably, the expression of the multivariate mixed model is:

[0025]

[0026] Among them, C i represents the concentration of fingerprint identification factor i in sediment deposited at the output section of the watershed, P s represents the sediment contribution ratio of sediment source s, C si represents the average concentration of fingerprint identification factor i in sediment source s, m represents the number of sediment sources, and n represents the number of fingerprint identification factors.

[0027] Preferably, the sum of the sediment contribution rates of each sub-area is 1, that is:

[0028]

[0029] Preferably, the fingerprint identification factor library includes chromium, copper, lead, zinc, barium, titanium, manganese, aluminum, iron, calcium and magnesium.

[0030] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: based on the principle of least squares method, the sediment contribution rate of each sand-producing area is accurately solved by utilizing multivariate mixed model and composite fingerprint recognition technology; reasonable scheduling of sediment in the reservoir area can be achieved by accurately estimating the amount of sediment entering the reservoir and the source of the sediment entering the reservoir. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0032] Figure 1 is a flow chart of the method of the present invention;

[0033] Figure 2 This is the area division map of the Three Gorges Reservoir area of ​​the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] The purpose of this invention is to provide a method for distinguishing the source of sediment in a large area. Figure 1 As shown, the specific steps include the following:

[0036] S1. Divide the large regional scale watershed into different sub-regions;

[0037] S2, collecting sediment samples in the sub-areas and preparing sediment sample solutions;

[0038] S3, drawing a calibration curve based on the fingerprint recognition factor library;

[0039] S4, sequentially measuring the prepared sediment sample solutions under the corresponding calibration curves to obtain the metal element content in the sediment samples and determine the sediment fingerprint element characteristics of the sub-regions;

[0040] S5. Use composite fingerprint recognition technology to solve the sediment contribution rate of each sub-area.

[0041] In this embodiment, the Three Gorges Reservoir is taken as an example, and the method of the present invention is used to conduct a comprehensive analysis of the sediment sources in the Three Gorges Reservoir area.

[0042] Sediment sampling:

[0043] From July to August 2017, 92 sediment samples were collected based on the current status of the reservoir topographic map and watershed map. The sediment samples included floodplain sediments and slope deposits, including 6 samples from the upstream section of the Wubu River estuary, 7 samples from the Wubu River Basin, 11 samples from the Yulin River Basin, 12 samples from the Long River Basin, 4 samples from the downstream section of the Long River estuary, 5 samples from the upstream section of the Modao Creek estuary, 13 samples from the Modao Creek Basin, 5 samples from the Pengxi River Basin, 16 samples from the Daning River Basin, 8 samples from the Xiangxi River Basin, and 5 samples from the downstream section of the Xiangxi River estuary. The samples involved 4 districts and 7 counties in Chongqing, including Banan District, Yubei District, Changshou District, Fengdu County, Shizhu County, Wanzhou District, Yunyang County, Fengjie County, Wuxi County, Wushan County, and Xingshan County in Yichang.

[0044] From July to August 2018, in order to increase the number of sampling points, we went to the Three Gorges Reservoir area again to collect 32 sediment samples, including floodplain sediments and slope deposits, including 3 samples from the upstream section of the Wubu River estuary, 3 samples from the Wubu River Basin, 3 samples from the Yulin River Basin, 3 samples from the Long River Basin, 5 samples from the Modao River Basin, 3 samples from the Pengxi River Basin, 8 samples from the Daning River Basin, and 4 samples from the Xiangxi River Basin. The samples involved Changshou District, Fengdu County, Shizhu County, Wanzhou District, Yunyang County, Fengjie County, Wuxi County, Wushan County, and Xingshan County of Yichang City in Chongqing. The collected sediment samples were numbered and naturally air-dried.

[0045] Sample preparation:

[0046] The 124 sediment samples collected were naturally air-dried, and dead branches and leaves, crop roots and coarse gravel were manually picked out. Each sample was prepared according to the relevant provisions of the "Determination of 22 Metal Elements in Solid Waste by Inductively Coupled Plasma Emission Spectrometry". Each sample was accurately weighed 10g, ground, and all passed through a 100-mesh sieve for later use.

[0047] Weigh 0.1 g of each sample after sieving and perform microwave digestion according to the order in Table 1. After digestion, use 1% nitric acid solution to make up the volume, mix well, and wait for testing. At the same time, prepare a blank sample as a control.

[0048]

[0049] Table 1

[0050] Plot the calibration curve:

[0051] Chromium (Cr), copper (Cu), lead (Pb), zinc (Zn), barium (Ba), titanium (Ti), manganese (Mn), aluminum (Al), iron (Fe), calcium (Ca) and magnesium (Mg) were selected as the fingerprint identification factor library. According to the requirements of the concentration of the multi-element mixed solution in Table 2, standard solutions of different concentrations were prepared. The element concentrations in the table are formulated according to the range of element concentrations of the collected sediment samples. After the standard solution is prepared, the emission intensity is measured in the order of the mixed solution concentration from low to high using an IACP6000 plasma electron coupling spectrometer. After the measurement, the measured emission intensity value is used as the ordinate and the concentration value of the mixed solution is used as the abscissa to complete the drawing of the standard curve.

[0052]

[0053]

[0054] Table 2

[0055] Determination:

[0056] The prepared sediment sample solutions are sequentially tested for the concentration of designated elements under the conditions of the corresponding standard curve. If the element concentration of individual samples exceeds the limit of the standard curve, the samples need to be diluted before being tested.

[0057] After the determination is completed, the content of metal elements in the corresponding sediment samples ω (mg / kg) is calculated:

[0058]

[0059] Among them, ω represents the content of metal elements in the sediment sample, ρ1 represents the measured concentration of metal elements in the sediment sample solution, ρ0 represents the measured concentration of metal elements in the blank solution, V0 represents the fixed volume of the sample after digestion, and m1 represents the weighed amount of the sediment sample.

[0060] Furthermore, the specific steps of using composite fingerprint recognition technology to solve the sediment contribution rate of each sub-area are as follows:

[0061] S51. Using the multivariate discriminant analysis method, find the best fingerprint factor combination in the sediment fingerprint element characteristics of the sub-region;

[0062] S52. Based on the optimal fingerprint factor combination, the multivariate mixed model is used to solve the sediment contribution rate of each sub-region.

[0063] The expression of the multivariate mixed model is:

[0064]

[0065] Among them, C i represents the concentration of fingerprint identification factor i in sediment deposited at the output section of the watershed, P s represents the sediment contribution ratio of sediment source s, C si represents the average concentration of fingerprint identification factor i in sediment source s, m represents the number of sediment sources, and n represents the number of fingerprint identification factors.

[0066] The function needs to consider the following two constraints:

[0067] 1. The sediment contribution rate of all sand-producing sources cannot be negative, that is: 0≤P s ≤1;

[0068] 2. The sum of the sediment contribution rates of the sub-areas is 1, that is:

[0069] By satisfying the above-mentioned restrictions and minimizing the function value, the sediment contribution rate of each sand-producing source can be solved.

[0070] Based on the geomorphic map, land use map, slope map and vegetation coverage map of the Three Gorges Reservoir area, the Three Gorges area is divided into the following areas: Figure 2 Each of the seven regions shown has a typical small watershed with measured data. Based on the characteristics of sediment fingerprint elements collected from the representative small watershed in each region, the composite fingerprint recognition technology is used to solve the contribution rate of incoming sediment in each region.

[0071] The Kruskall-Wallis H nonparametric test was used to preliminarily screen 11 fingerprint factors in the sediment samples of the upstream section of the Wubu River estuary, Wubu River, Yulin River, and Long River, and the factors with no significant differences in the source sediment samples were excluded (P>0.05). Among them, the P values ​​corresponding to 6 fingerprint factors were less than or equal to 0.05, which met the conditions for fingerprint factor screening, indicating that the concentrations of these 6 fingerprint factors in sediment samples from different sources were significantly different. The 6 fingerprint factors that met the conditions included Ca, Fe, Mg, Cr, Pb, and Ti; among the 5 fingerprint factors that did not meet the conditions, Al had the smallest difference, and the corresponding statistic H value was 2.688, and the P value was 0.442. The screening results are shown in Table 3.

[0072]

[0073]

[0074] Table 3

[0075] Kruskall-Wallis H parameter-free test was also used to preliminarily screen 11 fingerprint factors in the sediment samples of the upstream section of the Modao River estuary, Modao River, Pengxi River, Daning River, and Xiangxi River, and the factors with no significant difference in the source sediment samples were excluded (P>0.05). Among them, the P values ​​corresponding to 9 fingerprint factors were less than or equal to 0.05, which met the conditions for fingerprint factor screening, indicating that the concentrations of these 9 fingerprint factors in sediment samples from different sources were significantly different. The 9 fingerprint factors that met the conditions included Ca, Fe, Mg, Cr, Cu, Pb, Zn, Mn, and Ti, among which the concentrations of Ca, Fe, and Mg were the most significant, and their P values ​​were all 0.000. Among the 2 fingerprint factors that did not meet the conditions, Al had the smallest difference, and the corresponding statistic H value was 3.924 and the P value was 0.416. The screening results are shown in Table 4.

[0076]

[0077]

[0078] Table 4

[0079] The stepwise discriminant analysis method was used to identify the optimal fingerprint factor combination that could significantly distinguish different sediment sources through the Kruskall-Wallis test.

[0080] Multivariate stepwise discriminant analysis was performed on the six fingerprint identification factors selected from the upstream section of the Wubu River estuary, Wubu River, Yulin River and Long River. The results showed that when the four fingerprint factors of Pb, Ca, Cr and Mg were selected, the overall Wilks' Lmabda value gradually decreased from 0.164 to 0.012, and the cumulative correct discrimination rates were 73.054%, 82.048%, 97.240% and 100.000% respectively. The cumulative discrimination rate exceeded 90%, which met the analysis requirements. Therefore, Pb, Ca, Cr and Mg are the best combination fingerprint factors for the upstream section of the Wubu River estuary, Wubu River, Yulin River and Long River.

[0081] Using these four fingerprint factors, the correct discrimination rate of samples reached 83.3%, 85.7% in the Wubuhe area, 81.8% in the Yulinhe area, 83.3% in the Longhe area, and the total discrimination accuracy rate reached 83.3%.

[0082] A multivariate stepwise discriminant analysis was performed on the nine fingerprint identification factors selected from the upper reaches of the Modao Creek estuary, Modao Creek, Pengxi River, Daning River and Xiangxi River. The results showed that when the five fingerprint factors of Fe, Mg, Ca, Pb and Cr were selected, the overall Wilks' Lmabda value gradually decreased from 0.293 to 0.028, and the cumulative correct discrimination rates were 43.261%, 52.108%, 59.288%, 84.092% and 100.000% respectively. The cumulative discrimination rate exceeded 90%, which met the analysis requirements. Therefore, the five elements of Fe, Mg, Ca, Pb and Cr are the best combination fingerprint factors for the upper reaches of the Modao Creek estuary, Modao Creek, Pengxi River, Daning River and Xiangxi River.

[0083] Using these five fingerprint factors, the correct discrimination rate of sediment samples above the mouth of Pengxi River reached 100.0%, 100.0% in the Pengxi River area, 92.3% in the Modaoxi area, 75.0% in the Daning River area, and 75.0% in the Xiangxi River area, with the total discrimination accuracy reaching 85.1%.

[0084] Based on the principle of least squares method, the multivariate mixed model was used to quantitatively calculate the sediment contribution rate of each sand-producing area.

[0085] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for determining the source of sediment at a large regional scale, characterized in that: The specific steps include the following: Divide the large regional scale watershed into different sub-regions; Collecting sediment samples in the sub-area to prepare a sediment sample solution; Draw a calibration curve based on the fingerprint identification factor library; The prepared sediment sample solutions are sequentially measured under corresponding calibration curves to obtain the metal element content in the sediment sample and determine the sediment fingerprint element characteristics of the sub-area; The composite fingerprint recognition technology is used to solve the sediment contribution rate of each sub-area.

2. A method for determining the source of sediment at a large regional scale according to claim 1, characterized in that: The steps of preparing the sediment sample solution are: The collected sediment samples were naturally air-dried, and dead branches and leaves, crop roots and coarse gravel were manually picked out before grinding, and all of them were passed through a 100-mesh sieve; The sieved sediment samples were weighed with a fixed mass and subjected to microwave digestion in a fixed order; After digestion, dilute to volume with 1% nitric acid solution, mix well and wait for testing.

3. A method for determining the source of sediment at a large regional scale according to claim 1, characterized in that: The steps of drawing the calibration curve are: Based on the fingerprint identification factor library, standard solutions of different concentrations are configured; After the standard solution is prepared, the emission intensity is measured in the order of the concentration of the mixed solution from low to high using a plasma electron coupling spectrometer; After the measurement is completed, the calibration curve is drawn with the measured emission intensity value as the ordinate and the concentration value of the mixed solution as the abscissa.

4. A method for determining the source of sediment at a large regional scale according to claim 1, characterized in that: The calculation formula of the metal element content is: Among them, ω represents the content of metal elements in the sediment sample, ρ1 represents the measured concentration of metal elements in the sediment sample solution, ρ0 represents the measured concentration of metal elements in the blank solution, V0 represents the fixed volume of the sample after digestion, and m1 represents the weighed amount of the sediment sample.

5. The method for determining the source of sediment at a large regional scale according to claim 1, characterized in that: The specific steps of solving the sediment contribution rate of each sub-area using the composite fingerprint recognition technology are as follows: Using a multivariate discriminant analysis method, the best fingerprint factor combination is found in the sediment fingerprint element characteristics of the sub-area; Based on the optimal fingerprint factor combination, the contribution rate of sediment entering the reservoir of each sub-area is solved through a multivariate mixed model.

6. A method for determining the source of sediment at a large regional scale according to claim 5, characterized in that: The expression of the multivariate mixed model is: Among them, C i represents the concentration of fingerprint identification factor i in sediment deposited at the output section of the watershed, P s represents the sediment contribution ratio of sediment source s, C si represents the average concentration of fingerprint identification factor i in sediment source s, m represents the number of sediment sources, and n represents the number of fingerprint identification factors.

7. A method for determining the source of sediment at a large regional scale according to claim 6, characterized in that: The sum of the sediment contribution rates of each sub-area is 1, that is:

8. The method for determining the source of sediment at a large regional scale according to claim 1, characterized in that: The fingerprint identification factor library includes chromium, copper, lead, zinc, barium, titanium, manganese, aluminum, iron, calcium and magnesium.

Images