A method for investigating regional soil heavy metal net input flux

Abstract

The application provides a regional soil heavy metal net input flux investigation method, comprising: obtaining sample data sets of soil heavy metals and soil bulk density of a region to be measured, a land use type map and source emission intensity characteristics; determining an influence effect of the land use type of the region to be measured on soil heavy metals and a soil heavy metal residual data set according to the soil heavy metal data set and the land use type map; constructing a robust absolute principal component score / robust geographic weighted regression model combined with the land use information based on the residual data set and the influence effect, and performing heavy metal source analysis according to the robust absolute principal component score / robust geographic weighted regression model to obtain heavy metal net contribution concentrations of each emission source; and determining the current net input flux of the emission source according to the heavy metal net contribution concentrations of each emission source, the soil bulk density and the source emission intensity characteristics. The application solves the high cost problem of regional soil heavy metal net input flux investigation.

Description

Technical Field

[0001] This invention relates to the field of soil pollutant investigation technology, and in particular to a method for investigating the net input flux of heavy metals in regional soils. Background Technology

[0002] Soil is both a source and a sink of pollutants. Within the soil medium, pollutants undergo a dynamic input / output process. Simply investigating current soil pollutant levels and source emission characteristics is insufficient to describe the accumulation characteristics of soil pollutants. Understanding the net input flux of pollutants is crucial for effective early warning of soil pollution.

[0003] Traditionally, input / output survey methods are often used to determine the net input flux of soil pollutants. This method involves identifying the main input and output pathways of a target pollutant, determining the flux of each input or output pathway, and calculating the net input flux based on GIS spatial analysis. However, this traditional survey method has several drawbacks: (i) soil pollutants typically have multiple input / output pathways that need to be investigated, such as atmospheric deposition, irrigation, fertilizer and pesticide application, plant uptake, infiltration, and volatilization; (ii) parameters related to pollutant input / output pathways (e.g., the diffusion coefficient of pollutants in the atmosphere and the permeability coefficient in soil) are often spatially variable, thus requiring the investigation of numerous related environmental factors; and (iii) the investigation of some parameters related to input / output pathways (e.g., source composition spectroscopy) often encounters spatial inaccessibility or lack of monitoring permits. Therefore, traditional surveys of net input fluxes of soil pollutants are often difficult and costly.

[0004] Receptor models are commonly used source apportionment models. These models can qualitatively determine the source type and quantitatively determine the source contribution based on pollutant sample data from some receptor sites. In actual field environments, the pollutant content in collected soil samples is the content after the input / output balance has been achieved. Therefore, source apportionment receptor models based on these soil sample data can be used to determine the net contribution of each emission source.

[0005] Unlike water and air, soil parent material typically contains pollutants such as heavy metals. Therefore, the "natural background" sources of certain pollutants in the soil medium are naturally occurring. With the intensification of human activities, their negative impact on the environment has gradually attracted public attention. Regulating the intensity of human activities and controlling the emission of anthropogenic pollution sources has been recognized as one of the most effective environmental protection measures. Therefore, in environmental management practice, it is necessary to prioritize the exogenous input of pollutants. Source apportionment receptor models can provide an alternative method for quantifying the net exogenous input flux of soil pollutants. Furthermore, since source apportionment receptor models do not require emission conditions, meteorological and topographic data, nor do they need to track the complex migration processes of pollutants, they are expected to avoid the difficulties encountered by traditional input / output survey methods, thus saving significant manpower and resources when investigating the net exogenous input flux of pollutants. Receptor models originated from the source apportionment of air pollutants. Traditional receptor models are non-spatial models and are sensitive to outliers. Due to the relatively stable spatial distribution characteristics of air pollutants, receptor models often achieve satisfactory application results in the source apportionment of air pollutants. However, soil pollutants typically exhibit greater spatial heterogeneity compared to air pollutants, which may reduce the source resolution accuracy of traditional receptor models (such as absolute principal component score / multiple linear regression (APCS / MLR)). Furthermore, human activities often generate more outliers, which could further impact the source resolution accuracy of traditional receptor models. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, the purpose of this invention is to provide a method for investigating the net input flux of heavy metals in regional soils. This invention solves the problem of high cost in the prior art for investigating the net input flux of pollutants in regional soils.

[0007] To achieve the above objectives, the present invention provides the following solution:

[0008] A method for investigating net input fluxes of heavy metals in regional soils, comprising:

[0009] Obtain sample datasets of heavy metals and bulk density in the soil of the area to be tested, as well as land use type maps and source emission intensity characteristics;

[0010] Based on the soil heavy metal dataset and land use type map, determine the impact of land use type on soil heavy metals in the area to be tested and the soil heavy metal residual dataset.

[0011] Based on the residual dataset and the impact effect, a robust absolute principal component score / robust geographical weighted regression model combining categorized land use information is constructed, and heavy metal sources are analyzed according to the robust absolute principal component score / robust geographical weighted regression model to obtain the net contribution concentration of heavy metals from each emission source.

[0012] The current net input flux of each emission source is determined based on the net heavy metal contribution concentration, soil bulk density, and source emission intensity characteristics.

[0013] Preferably, the soil bulk density was determined at each soil sampling point using the ring cutter method, and Cu, Fe, Pb, Zn, Mn, Cd, and Cr in the soil heavy metal sample dataset were analyzed using atomic absorption spectrophotometry, and As in the soil heavy metal sample dataset was analyzed using atomic fluorescence spectrometry; land use type maps were obtained using GIS methods.

[0014] Preferably, the source emission intensity characteristics of each heavy metal emission source are investigated, wherein the source emission intensity characteristics include emission intensity and emission time.

[0015] Preferably, the expression for the effect of land use type on soil heavy metals is:

[0016]

[0017] Where lu(u) is the land use type at location u, lu0 is the land use type corresponding to the lowest average heavy metal content, and m[·] is the average heavy metal concentration corresponding to the land use type in parentheses.

[0018] Preferably, the expression for the soil heavy metal residual dataset is:

[0019] r(u) = z(u) - l(u),

[0020] Where r(u) represents the residual data of heavy metals in the soil at location u, and z(u) represents the sample data of heavy metals in the soil at location u.

[0021] Preferably, the expression for the RAPCS / RGWR-CLU source resolution model is:

[0022]

[0023] Where, r j (u) represents the residual data of heavy metal j in the soil at location u, RAPCS is the difference between the robust factor analysis score corresponding to the anthropogenic zero sample data and the robust factor analysis score corresponding to the heavy metal residual data, and p is the number of heavy metal emission sources identified by the robust factor analysis. k (u) represents the k-th RAPCS at position u, b 0j (u) is the local regression intercept corresponding to the residual data of heavy metal j at location u, b kj (u) is the local regression slope at position u corresponding to the k-th RAPCS of heavy metal j.

[0024] Preferably, the expression for the net heavy metal contribution concentration of the emission source is:

[0025]

[0026] in, Let l be the net contribution concentration of heavy metals from the k-th emission source at location u. j (u) represents the effect of land use type at location u on soil heavy metal j.

[0027] Preferably, the net input flux expression of the emission source is:

[0028]

[0029] in, Let denoted by ρ(u) be the net input flux of heavy metal j from n emission sources at location u, d be the soil sampling depth, and ρ(u) be the soil bulk density at location u. Let t be the concentration contribution of heavy metal j from the k-th emission source at location u. k Let α be the emission time of the k-th emission source. k This is the ratio between the emission flux of the k-th emission source at the time of the soil survey and the average emission flux over the entire emission period.

[0030] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0031] This invention acquires sample datasets of soil heavy metals and soil bulk density, land use type maps, and source emission intensity characteristics of the area to be tested. Based on the heavy metal datasets and land use type maps, it determines the impact of land use type on soil heavy metals and obtains a residual dataset of soil heavy metals. Based on the residual datasets and the impact effects, it constructs a robust absolute principal component score / robust geographically weighted regression (RAPCS / RGWR-CLU) model incorporating categorized land use information. The RAPCS / RGWR-CLU model is then used to analyze heavy metal sources, obtaining the net contribution concentration of heavy metals from each emission source. Based on the net contribution concentration of heavy metals from each emission source, soil bulk density, and source emission intensity characteristics, the current net input flux of the emission source is determined. This invention solves the high-cost problem of regional soil heavy metal net input flux surveys. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This invention provides a flowchart for investigating net pollutant input flux based on a source apportionment receptor model.

[0034] Figure 2 This is a schematic diagram illustrating the principle of pollutant net input flux survey based on a source apportionment receptor model, provided for an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments 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, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0036] The purpose of this invention is to provide a method for investigating the net input flux of heavy metals in regional soils. This invention solves the problem of high cost in the prior art for investigating the net input flux of heavy metals in regional soils.

[0037] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] like Figure 1 As shown, this invention provides a method for investigating the net input flux of heavy metals in regional soils, comprising:

[0039] Step 100: Obtain sample datasets of heavy metals and bulk density in the soil of the area to be tested, land use type maps, and source emission intensity characteristics;

[0040] Specifically, the study area of ​​this invention is located around a typical copper smelting area. The study area covers approximately 304 square kilometers. The average annual precipitation in this area is 1826.4 mm, and the prevailing wind direction throughout the year is easterly and northeasterly. This study area is a traditional agricultural area. Dryland is mainly planted with maize, and paddy fields are mainly planted with rice. The main soil type in this study area is red soil. In copper smelting, various heavy metals are inevitably emitted into the surrounding environment through atmospheric deposition. The regional environmental survey was conducted in December 2021, and 230 soil samples (0-20 cm) were collected. The soil bulk density was determined at each soil sampling point using the ring cutter method. The pretreatment of the soil samples included natural air drying at room temperature, removal of debris, grinding, and sieving through a 0.149 mm sieve. In this invention, the concentrations of Cu, Fe, Pb, Zn, Mn, Cd, and Cr in the soil samples were determined using atomic absorption spectrophotometry, and the concentration of As in the soil samples was determined using atomic fluorescence spectrometry. Quality control was performed using a standard reference material (RSM) (GBW07401 - Standard Material for Soil Composition Analysis). The recoveries of heavy metals in the RSM ranged from 93.5% to 102.5%. The relative standard deviations of the analytical results for all heavy metals were less than 5%.

[0041] Furthermore, source emission intensity characteristics were obtained, including emission intensity and emission time. The Guixi Copper Smelter commenced operation in 1986. Therefore, in this study, the emission times t for the "agricultural input" source and the "atmospheric deposition" source were set to 43 years and 35 years, respectively. The input intensity of agricultural chemicals in farmland within the study area was relatively stable. Therefore, the adjustment coefficient corresponding to the "agricultural input" source (i.e., α) was set to 1, meaning the input flux from the "agricultural input" source was equal to the average input flux over the entire emission period. The atmospheric deposition heavy metal input flux in Guixi City gradually increased with the expansion of copper smelting capacity. Based on historical expansion information of the copper smelting scale in Guixi City, the adjustment coefficient (i.e., α) corresponding to the "atmospheric deposition" source was set to 0.5, meaning the input flux from the "atmospheric deposition" source was twice the average input flux over the entire emission period.

[0042] Furthermore, a spatial distribution map of land use types was obtained using GIS methods. Specifically, the resolution of the land use map for the study area was 30m × 30m. Due to the potential impact of agricultural practices on certain soil heavy metals, the land use types in this study area were further divided into agricultural land and non-agricultural land.

[0043] Step 200: Determine the impact of land use type on soil heavy metals and the soil heavy metal residual dataset in the area to be tested based on the soil heavy metal dataset and land use type map;

[0044] The average concentrations of As, Cu, Pb, and Zn in topsoil (0-20 cm) were significantly higher in agricultural land than in non-agricultural land (p<0.05) (Table 1). These heavy metals may be closely related to agricultural inputs, such as the application of pesticides and fertilizers. With the development of remote sensing technology, land use maps have become increasingly easy to obtain. Therefore, land use effects should be incorporated into source apportionment models. However, previous studies on soil pollutant source apportionment rarely incorporated this auxiliary information to improve the accuracy of source apportionment. Table 1 shows the average concentrations of heavy metals in soils of different land use types and the results of variance analysis. Table 1 is shown below:

[0045] Table 1. Average concentrations of heavy metals in soils of different land use types and results of variance analysis.

[0046]

[0047]

[0048] Note: Land use type significantly (p<0.05) affected the concentrations of As, Cu, Pb, and Zn.

[0049] Step 300: Based on the residual dataset and the impact effect, construct a robust absolute principal component score / robust geographical weighted regression (RAPCS / RGWR-CLU) model that combines categorized land use information, and perform heavy metal source analysis according to the RAPCS / RGWR-CLU model to obtain the net contribution concentration of heavy metals from each emission source;

[0050] Table 2 shows the results of three factor analyses: robust factor analysis incorporating the heavy metal residual dataset (RFA-residual), robust factor analysis using the original heavy metal dataset (RFA-raw), and conventional factor analysis incorporating the original heavy metal dataset (TFA-raw). Each factor analysis model extracted three principal factors. In this study, the cumulative variance explained by the three factors (F1, F2, and F3) was 85.54% for RFA-residual, 81.54% for RFA-raw, and 78.07% for TFA-raw. Table 2 is shown below:

[0051] Table 2. Rotated (maximum variance rotation) factor loading matrix obtained from the factor analysis model (n = 230).

[0052]

[0053]

[0054] Note: RFA-residual combines robust factor analysis of the heavy metal residual dataset; RFA-raw combines robust factor analysis of the raw heavy metal dataset; TFA-raw combines traditional factor analysis of the raw heavy metal dataset.

[0055] For each factor analysis model, F1 was strongly correlated with Fe, Mn, Cr, and Pb in the topsoil (Table 2). Fe and Mn are abundant in the Earth's crust and are often used as indicator elements for "natural background" sources. Therefore, F1 may represent a "natural background" source of heavy metals in the soil. For each factor analysis model, F2 was strongly correlated with As, Cd, and Cu in the topsoil (Table 2). The survey showed that copper smelting is the largest source of heavy metals in this study area. During copper smelting, heavy metals such as copper and cadmium are inevitably emitted into the surrounding soil through atmospheric deposition. Therefore, F2 may represent an "atmospheric deposition" source of heavy metals in the soil. For each factor analysis model, F3 in the topsoil was strongly correlated with Zn and weakly correlated with Cu (Table 2). In agricultural practices, large amounts of pesticides and fertilizers are often used to maintain high crop yields. These pesticides and fertilizers typically contain high levels of zinc and copper. Therefore, F3 may represent an "agricultural input" source.

[0056] Specifically, this embodiment discloses an existing method for investigating net pollutant input fluxes:

[0057] Traditional calculations of net input fluxes of soil pollutants are typically based on the principle of mass conservation. This method requires accounting for each major input / output pathway. The traditional framework for calculating net input fluxes of heavy metals in regional soils is as follows:

[0058] (1) Total input flux of heavy metals in soil:

[0059] Identify the main pathways of heavy metal input in the soil of the study area;

[0060] Using GIS spatial analysis methods and relevant auxiliary information, the input flux generated at location u by each input pathway (i) is calculated. The sum of .

[0061] (2) Total heavy metal fluxes from soil:

[0062] Identify the main pathways of heavy metal export from soil in the study area;

[0063] Using GIS spatial analysis methods and relevant auxiliary information, the input flux generated at location u by each input pathway (j) was calculated. The sum of .

[0064] The formula for the net input flux of heavy metals in the soil at location u is:

[0065]

[0066] The main input pathways of heavy metals are atmospheric deposition and agricultural input (i.e., irrigation water, fertilizers, and pesticides), while the main output pathways of heavy metals from soil are crop removal and surface runoff. In this study area, the impact of copper smelting activities on the surrounding environment is primarily through atmospheric deposition. This invention uses an atmospheric diffusion model (AERMOD) to simulate the atmospheric deposition input flux of heavy metals. This model is based on existing atmospheric diffusion statistical theory. The model has three modules: an atmospheric diffusion module, a topographic data preprocessing module, and a meteorological data preprocessing unit. The simulation process requires surface meteorological data (i.e., wind frequency, wind direction, and humidity), topographic / elevation data, and source emission data (i.e., chimney height and inner diameter, flue gas velocity, flow rate, temperature, heat capacity, flue gas density, and flue gas molecular weight). Other input / output pathways are closely related to land use type. Therefore, land use type maps are valuable auxiliary information in surveys of net heavy metal input fluxes.

[0067] Specifically, this embodiment compares the existing model with the model in this embodiment:

[0068] This invention compares the source resolution accuracy of RAPCS / RWR-CLU, RAPCS / RGWR, and APCS / MLR. First, the relative root mean square error (RRMSE) and global Moran's I are used as evaluation metrics. The formula for calculating RRMSE is as follows:

[0069]

[0070] Where n is the number of soil samples (230 in this study), m i c is the heavy metal concentration of the i-th soil sample. i This represents the total contribution concentration of the main pollution sources identified by the receptor model in the i-th soil sample. A lower RRMSE means that the pollutant concentrations resolved by the receptor model are closer to the actual concentrations at these sample locations. Secondly, global Moran's I is used to evaluate the fit of the regression model to these receptor models. This metric is commonly used to quantify the spatial autocorrelation of the regression residuals. The closer global Moran's I is to zero, the weaker the spatial autocorrelation of the residual data at the sample location, and the better the regression fit. Finally, negative source contribution rates are also used as an important indicator of source apportionment quality. A higher negative contribution rate corresponds to poorer source apportionment capability.

[0071] This invention also compares the accuracy of net input flux determined by the RAPCS / RGWR-CLU-based survey method with existing input / output survey methods. First, the corresponding RRMSE was determined based on 230 sample sites. Second, the human, material, and time inputs for the two methods were compared.

[0072] In this invention, IBM SPSS (version 26.0) was used for multiple linear regression (MLR) and factor analysis; "GWmodel" and "robustbase" (http: / / cran.r-project.org / ) in the R statistical software package (version 4.3.2) were used for robust geographic weighted regression and robust factor analysis, respectively; EIAPro 2018 was used to execute the AERMOD model; and ArcGIS (version 10.2) was used for geocomputation and spatial analysis.

[0073] The validation metrics for the three source apportionment receptor models (RAPCS / RGWR-CLU, RAPCS / RGVR, and APCS / MLR) are shown in Table 3. RAPCS / RGWR-CLU produced the lowest RRMSE. Furthermore, the global Moran's I generated by RAPCS / RGWR-CLU was closer to zero than that generated by RAPCS / RGVR and APCS / MLR (Table 3). In addition, RAPCS / RGWR-CLU had fewer negative source contributions compared to RAPCS / RGWR and APCS / MLR (Table 4). Therefore, RAPCS / RGWR-CLU is the most efficient source apportionment receptor model. The main reasons may be as follows: (i) the traditional APCS / MLR is a non-spatial model that does not consider the spatial heterogeneity among soil factors, making it difficult to effectively apportion source contributions within local subregions; (ii) the traditional APCS / MLR is susceptible to outlier samples, which are often widely present in anthropogenically polluted areas; and (iii) although RAPCS / RGWR overcomes the above problems, land use type data closely related to agricultural emission sources have not yet been included in the source apportionment receptor model. Table 3 shows the validation metrics for the source apportionment receptor model (n=230), and Table 4 shows the average contribution rate of heavy metal sources obtained from the source apportionment receptor model (n=230). Tables 3 and 4 are shown below:

[0074] Table 3. Validation metrics for the source apportionment receptor model (n=230)

[0075]

[0076]

[0077] Note: RRMSE is the root mean square error obtained from the source contribution concentration and sample test concentration obtained from the source apportionment receptor model; Moran's I is the global Moran's I exponent of the regression residuals in the source apportionment receptor model.

[0078] Table 4. Average contribution rate of heavy metal sources obtained from the source apportionment receptor model (n=230)

[0079]

[0080] Step 400: Determine the current net input flux of the emission sources based on the net heavy metal contribution concentration, soil bulk density, and source emission intensity characteristics of each emission source.

[0081] Specifically, in the net input flux determined by the existing input / output survey method, As is 81.94 mg m -2 a -1 Cd was 4.01 mg m -2 a -1 Cr is 6.05 mg m -2 a -1 Cu is 429.20 mg m -2 a -1 Pb was 41.68 mg m -2 a -1 Zn content was 131.94 mg / m³. -2 a -1 (Table 5). In the RRMS determined with reference to the net input flux determined by the traditional input / output survey method, As is 0.05, Cd is 0.03, Cr is 0.08, Cu is 0.06, Pb is 0.05, and Zn is 0.05 (Table 5). Therefore, for each investigated heavy metal, the net input flux obtained based on RAPCS / RGWR-CLU is close to the net input flux obtained by the traditional input / output survey method. Table 5 shows the average net input flux (mg / m³) of exogenous input sources obtained by the two methods. -2 a -1 ) and its validation metrics (n=230). Table 5 is shown below:

[0082] Table 5. Average net input flux (mg / m³) of exogenous input sources obtained by the two methods -2 a -1 ) and its verification indicators (n=230).

[0083]

[0084] Note: Average net input flux from exogenous sources, average net input flux of heavy metals from the combined effects of atmospheric deposition and agricultural input sources; RRMSE, root mean square error obtained with reference to the net input flux of heavy metals obtained by the traditional input / output survey method.

[0085] Despite employing drastically different technical approaches, both methods yielded similar results. However, the source apportionment receptor model-based approach was significantly less expensive in terms of human and material resources than traditional input / output survey methods. The main reasons are: (i) existing input / output survey methods require investigating multiple input / output pathways of pollutants (e.g., atmospheric deposition, irrigation, fertilizer and pesticide application, plant uptake and infiltration) in real-world field environments; (ii) for each input / output pathway at the regional scale, relevant parameters (e.g., the atmospheric diffusion coefficient of pollutants) are often spatially variable rather than constant; and (iii) due to the lack of access to or monitoring permits, relevant parameters for some input / output pathways (e.g., source composition spectroscopy) may be difficult to obtain. In contrast, the source apportionment receptor model only requires the investigation and analysis of relevant soil samples, which is typically necessary for soil environmental surveys. Therefore, RAPCS / RGWR-CLU is a cost-effective solution for determining net pollutant input fluxes.

[0086] Specifically, receptor models originate from source apportionment of air pollutants. Existing source apportionment receptor models, such as APCS / MLR, are typically built in variable space and are sensitive to outliers. Previous researchers proposed RAPCS / RGWR based on APCS / MLR to mitigate the spatial heterogeneity of soil properties and the impact of outliers on source apportionment accuracy. If categorical information is closely related to heavy metal emissions, this categorical information should be incorporated into the source apportionment model. For each heavy metal, the sample data is divided into two parts based on categorical information (land use type in this invention):

[0087] The expression for the effect of land use type on soil heavy metals is as follows:

[0088]

[0089] Where lu(u) is the land use type at location u, lu0 is the land use type corresponding to the lowest average heavy metal concentration, and m[·] is the average heavy metal concentration corresponding to the land use type in parentheses.

[0090] The expression for the soil heavy metal residual dataset is:

[0091] r(u) = z(u) - l(u),

[0092] Where r(u) represents the residual data of heavy metals in the soil at location u, and z(u) represents the sample data of heavy metals in the soil at location u.

[0093] Based on the soil heavy metal residual dataset, RAPCS / RGWR was further used to analyze the source contributions. In RAPCS / RGWR, robust geographically weighted regression was performed using the residual data as the dependent variable and RAPCS as the explanatory variable.

[0094]

[0095] Where, r j (u) represents the residual data of heavy metal j in the soil at location u, RAPCS is the difference between the robust factor analysis score corresponding to the anthropogenic zero sample data and the robust factor analysis score corresponding to the heavy metal residual data, and p is the number of heavy metal emission sources identified by the robust factor analysis. k (u) represents the k-th RAPCS at position u, b 0j (u) is the local regression intercept corresponding to the residual data of heavy metal j at location u, b kj (u) is the local regression slope at position u corresponding to the k-th RAPCS of heavy metal j.

[0096] The expression for the net contribution concentration of heavy metals from the emission source is:

[0097]

[0098] in, Let l be the net contribution concentration of heavy metals from the k-th emission source at location u. j (u) represents the effect of land use type at location u on soil heavy metal j.

[0099] The net input flux expression for the emission source is:

[0100]

[0101] in, Let denoted by ρ(u) be the net input flux of heavy metal j from n emission sources at location u, d be the soil sampling depth, and ρ(u) be the soil bulk density at location u. Let t be the concentration contribution of heavy metal j from the k-th emission source at location u. k Let α be the emission time of the k-th emission source. k This is the ratio between the emission flux of the k-th emission source at the time of the soil survey and the average emission flux over the entire emission period.

[0102] Specifically, pollutants in soil typically have multiple input and output pathways. Net pollutant input flux is crucial for accurately predicting future pollution development. Effective pollution early warning is only possible when understanding pollutant levels and net input flux within a target area. Existing technologies for investigating each input and output pathway of soil pollutants involve determining multiple relevant parameters, often consuming significant human and material resources. Because traditional input / output surveys require determining many relevant parameters, the cumulative transmission effect of errors inevitably exists in the final survey results. This invention proposes a method based on a source apportionment receptor model to determine net pollutant input flux, avoiding the need to investigate relevant parameters in various input and output pathways. However, traditional source apportionment receptor models originate from pollution source apportionment in atmospheric media, and these models are non-robust and non-spatial. The strong spatial heterogeneity and outliers of soil factors may negatively impact the source apportionment accuracy of traditional receptor models. Furthermore, pollutant emission characteristics at the regional scale are often closely related to specific sub-regions (such as land use types).

[0103] It is noteworthy that RAPCS / RGWR-CLU can analyze both "exogenous input" and "natural background" sources, while traditional input / output survey methods rarely focus on "natural background" sources. Furthermore, the net input flux obtained through source apportionment receptor models is the average net input flux over the entire emission cycle. In real soil environments, net input flux often varies with adjustments for human activity intensity. Therefore, the emission history and intensity of pollution sources are crucial for accurately simulating current and future net input fluxes. However, the net input flux obtained through traditional input / output survey methods is the result at the time of the survey. Source emission characteristics may differ across different study cases. Therefore, trends in human activity intensity and the quantitative relationship between human activity intensity and pollution emissions are essential for accurately simulating future changes in net input flux.

[0104] Soil typically exhibits a strong adsorption capacity for heavy metals. Leaching of heavy metals in soil is generally relatively weak. Therefore, heavy metals tend to accumulate in the topsoil layer. This study focuses on the topsoil (0-20 cm). If the leaching of the target pollutant is strong, the sampling depth of the target receptor should be deeper. Otherwise, it will lead to certain survey errors.

[0105] The beneficial effects of this invention are as follows:

[0106] This invention provides a method for investigating the net input flux of heavy metals in regional soils, comprising: acquiring sample datasets of soil heavy metals and soil bulk density, land use type maps, and source emission intensity characteristics of the area to be tested; determining the impact effect of land use type on soil heavy metals and the soil heavy metal residual dataset based on the heavy metal dataset and land use type map; constructing a robust absolute principal component score / robust geographically weighted regression (RAPCS / RGWR-CLU) model combining categorized land use information based on the residual dataset and the impact effect, and performing heavy metal source analysis based on the RAPCS / RGWR-CLU model to obtain the net contribution concentration of heavy metals from each emission source; and determining the current net input flux of the emission source based on the net contribution concentration of heavy metals from each emission source, soil bulk density, and source emission intensity characteristics. This invention solves the problem of high cost in investigating the net input flux of heavy metals in regional soils.

[0107] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0108] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for investigating the net input flux of heavy metals in regional soils, characterized in that, include: Obtain sample datasets of heavy metals and bulk density in the soil of the area to be tested, as well as land use type maps and source emission intensity characteristics; Based on the soil heavy metal dataset and land use type map, determine the impact of land use type on soil heavy metals in the area to be tested and the soil heavy metal residual dataset. Based on the residual dataset and the impact effect, a robust absolute principal component score / robust geographical weighted regression model combining categorized land use information is constructed, and heavy metal sources are analyzed according to the robust absolute principal component score / robust geographical weighted regression model to obtain the net contribution concentration of heavy metals from each emission source. Based on the net heavy metal contribution concentration, soil bulk density, and source emission intensity characteristics of each emission source, the current net input flux of the emission source is determined. The investigation obtained the source emission intensity characteristics of each heavy metal emission source, including emission intensity and emission time. The expression for the effect of land use type on soil heavy metals is as follows: ， in, For the land use type at location u, It is the land use type corresponding to the lowest average heavy metal concentration. The value in parentheses corresponds to the average heavy metal concentration for the land use type. The expression for the soil heavy metal residual dataset is: ， in, The data represents the residuals of heavy metals in the soil at location u. Data for heavy metal samples in the soil at location u; The expression for the robust absolute principal component score / robust geographically weighted regression model that combines categorized land use information is as follows: ， in, Here, represents the residual data for heavy metal j in the soil at location u; RAPCS is the difference between the robust factor analysis score corresponding to the anthropogenic zero-sample data and the robust factor analysis score corresponding to the heavy metal residual data; and p represents the number of heavy metal emission sources identified by the robust factor analysis. For the k-th RAPCS at position u, It is the local regression intercept corresponding to the residual data of heavy metal j at location u. It is the local regression slope at position u corresponding to the kth RAPCS of heavy metal j; The expression for the net contribution concentration of heavy metals from the emission source is: ， in, Let be the net contribution concentration of heavy metals from the k-th emission source at location u. The effect of land use type at location u on soil heavy metal j; The net input flux expression for the emission source is: ， in, Let be the net input flux of heavy metal j from n emission sources at location u, and d be the soil sampling depth. Let u be the soil bulk density at location u. Let t be the concentration contribution of heavy metal j from the k-th emission source at location u. k Let α be the emission time of the k-th emission source. k This is the ratio between the emission flux of the k-th emission source at the time of the soil survey and the average emission flux over the entire emission period.

2. The method for investigating the net input flux of heavy metals in regional soils according to claim 1, characterized in that, Soil bulk density was determined at each soil sampling point using the ring cutter method, and the concentrations of Cu, Fe, Pb, Zn, Mn, Cd, and Cr in the soil heavy metal samples were determined using atomic absorption spectrophotometry. The concentration of As in the soil heavy metal samples was determined using atomic fluorescence spectrometry. Land use type maps were obtained using GIS methods.

Images