CN102999709B - A kind of rural activity district underground water risk stratification assessing zonings method - Google Patents

Abstract

A kind of rural activity district underground water risk stratification assessing zonings method, by setting up a set of index system with judging, sampled point is arranged with the method for spatial discretization, gather underground water water sample, and according to primary election index feature screening index, by the degree of membership of the weight and each sampled point index that calculate each index, comprehensive evaluation is made to local ground watering pollution risk, and subregion and classification are carried out to underground water risk.The present invention is directed to China's rural activity district features of pollution, there is stronger adaptability and generalization, can science, reasonably assessment is made to the underground water risk class in rural activity district and risk distribution.

Description

A kind of rural activity district underground water risk stratification assessing zonings method

Technical field

The invention belongs to groundwater contamination and risk assessment field, specifically for the new feature of current rural activity district Groundwater Contamination Risk, propose a kind of reasonably rural activity district underground water risk stratification and assessing zonings method.

Background technology

In the water-using structure that China is current, underground water occupies 20% of national total supply, 70% of potable water output, 40% of the field irrigation water yield, 38% of industrial water consumption, this water-using structure can not change in a short time, and therefore underground water is most important to China's sustainable development that is economic and society from now on.Meanwhile, China is a large agricultural country, and agricultural chemicals annual production reaches 500,000 t, occupies the second in the world.Use amount is at about 230,000 t, and average amount of application is 2.33kg/hm ².And the agricultural chemicals used approximately only has 20% to be attached on crop, the agricultural chemicals of 80% finally enters soil, and under extreme weather or irrigation situation, agricultural chemicals very easily permeates the ground water, pollutes it.In recent years, underground water Pesticides in rural areas in our country was detected often.If cotton growing area, Tianmen county, Hubei Province of China is because of a large amount of applying pesticides, cause local potable water to be subject to severe contamination, in underground water, 1605 content reach 1.25mg/L, exceed national standard 375 times, DDT content reaches 0.44mg/L, exceeds standard 1.25 times, and indivedual villages and small towns well water can not be drunk completely; Organophosphorus pesticide is detected in the underground water of Dezhou City some areas, and the phenomenon that ubiquity exceeds standard, the underground water in Kaxgar Prefecture Chan Mian great county Shache county and Yingjisha two county 70% receives the pollution in various degree such as DDVP, benzene hexachloride, D.D.T. (dichloro-diphenyl-trichloroethane), Azodrin, flolimat.Therefore, rural activity district agricultural chemicals has become a kind of problem become increasingly conspicuous to the pollution of underground water, and as the underground water of the most of regional drinking water source of China once be polluted, the health of the mankind will be subject to grave danger, it is administered in addition will also be an extremely difficult process.

How scientificlly and effectively the rural activity district Groundwater Contamination Risk of a large amount of agricultural chemicals is used to assess to these, for agricultural chemicals administrative authority provides strong foundation, thus final reduce in agricultural production process the generation of the pollution of groundwater environment and spread, taking precautions against agricultural chemicals to the ecosystem and the issuable risk of health, is a problem in science urgently to be resolved hurrily.Rural activity district agricultural chemicals is to the pollution problem of underground water, one is less to relating to of agricultural chemicals in current quality of groundwater standard, conventional pesticide is only D.D.T. (dichloro-diphenyl-trichloroethane) and benzene hexachloride, because China forbids this two kinds of agricultural chemicals already, different novel agrochemicals is brought into use in different regions, therefore, to again be examined closely evaluation index; Two is choosing evaluation method, and objective comprehensive evaluation method is also make to risk the basis accurately estimated; 3rd, rural activity district groundwater contamination situation is in the displaying visually in space, and realize classification and the subregion of rural activity district Groundwater Contamination Risk, this proposes requirements at the higher level to the risk assessment of rural activity district groundwater contamination.

Summary of the invention

The object of the present invention is to provide one comparatively reasonably rural activity district underground water risk stratification assessing zonings method, rural activity district risk subregion and classification are evaluated underground water potential risk district.

For achieving the above object, rural activity district provided by the invention underground water risk stratification assessing zonings method, by setting up a set of index system with judging, sampled point is arranged with the method for spatial discretization, gather underground water water sample, and according to primary election index feature screening index, by the degree of membership of the weight and each sampled point index that calculate each index, comprehensive evaluation is made to local ground watering pollution risk, and subregion and classification are carried out to underground water risk.

Described evaluation method, wherein, index system is divided into three layers, and top layer is destination layer, and middle layer is rule layer, and bottom is indicator layer; According to pollutant classification, the pollutant class of rule layer separately includes organic pollutants, inorganic pollutant and heavy metal; Organic contaminant comprises the Insecticides (tech) & Herbicides (tech) in rural activity district, acaricide and germifuge; Inorganic pollutant comprises the inorganic ion of conventional polluted underground water, potential of hydrogen, total hardness, turbidity, chemical oxygen demand (COD) and biochemical oxygen demand; Heavy metal comprises copper, zinc, iron, lead, nickel, cadmium, chromium, arsenic and mercury.

Described evaluation method, wherein, the method for spatial discretization arranges sampled point, is be evenly arranged on evaluation region by sampled point, and adopt plum blossom to layout method, sampled point quantity is not less than 15; When gathering underground water water sample, underground water should be taken from.

Described evaluation method, wherein, primary election index feature be by Comparative indices between correlativity and water quality standard more each other.

Described evaluation method, wherein, the calculating of weight is the comprehensive importance being carried out measurement index by the product of integrated risk exponential sum unordered index two indices; Computing formula is:

W＝[w(1),w(2),…,w(n)] _1×n

Wherein, W is the synthetic weights vector of n index, and ρ is the integrated risk index of each index, for the unordered index of each index.

Described evaluation method, wherein, degree of membership be desired value and risk class grade scale are substituted into selected by membership function refer to calculate, subordinated-degree matrix T and membership function t=t (k, j) are as follows respectively:

$T={\left[\begin{array}{cccc}{t}_{11}& t_{12}& \·\·\·& t_{1n}\\ t_{21}& t_{21}& \·\·\·& t_{2n}\\ \·& \·& & \·\\ \·& \·& \·\·\·& \·\\ \·& \·& & \·\\ t_{51}& t_{52}& \·\·\·& t_{5n}\end{array}\right]}_{5\×n},$ Then t _ijcomputing method be:

For forward index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=5)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=5)\end{array}\right.$

For reverse index:

Described evaluation method, wherein, integrated risk index determines its importance with the frequency of grade residing for pollutant and risk class; Computing formula is:

$\mathrm{\ρ}\left(j\right)=\frac{\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}(j=1,\·\·\·,n,k=1,\·\·\·,5)$

In above formula, α (j, k) is for a jth index is below or above the number of k grade, and u (k) is score value corresponding to k grade, if underground water risk falls into 5 types, then u (1) to u (5) is respectively 1,2,3,4,5.

Described evaluation method, wherein, unordered index is the weight size representing rural activity district underground water risk indicator with entropy, and first do standardization to raw data, the standardized method of data is as follows:

For forward index:

$y(i,j)=\frac{x(i,j)-f_2\left(j\right)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,m,j=1,\·\·\·,n)$

In formula, f ₁(j)=max{s (i, j) | i=1 ..., m}, f ₂(j)=min{s (i, j) | i=1 ..., m}

For reverse index:

$y(i,j)=\frac{f_1\left(j\right)-x(i,j)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,u,j=1,\·\·\·,v)$

Computing information entropy: $h\left(j\right)=-k\underset{j=1}{\overset{m}{\mathrm{\Σ}}}s(i,j)\ln s(i,j)$

In above formula, $s(i,j)=\frac{y(i,j)}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}y(i,j)}(i=1,\·\·\·,m,j=1,\·\·\·,n),$ When x (i, j)=0, f (i, j)=0.00001; K=1/ln (m).

Calculating entropy is weighed:

Described evaluation method, wherein, synthetic weights vector carries out matrix multiplication operation with the transposition of subordinated-degree matrix, computing formula is: Q=WT', wherein, and Q=[q (1), q (2), q (3), q (4), q (5)], by above-mentioned Matrix Multiplication with V={v (j), j=1,2,3,4,5}, v ₁, v ₂, v ₃, v ₄, v ₅value be respectively 100,80,60,40,20, an integrate score value between 20 to 100 can be obtained, so above-mentioned multiple attribute synthetical evaluation is become the result of a single index.

Described evaluation method, wherein, the evaluation of classification and subregion, adopt ArcGIS to carry out subregion and classification to underground water risk: by corresponding with integrate score for each sample point coordinate, be loaded in ArcGIS, space interpolation is carried out to the discrete data of each sampled point, relatively IDW interpolation, Kriging interpolation, NaturalNeighbor interpolation and Spline interpolation four kinds of interpolation methods, select the interpolation method meeting spatial spreading evaluation result most, finally, adopt the method can obtain whole area of space continuous print water quality score, according to each water quality scoring rank interval, water quality result is spatially classified, obtain region continuous print underground water risk degree and zoning figure.

Rural activity district of the present invention underground water risk stratification assessing zonings method, index system is comprehensive, evaluation method meets China's reality, can promote in China's rural activity district underground water risk assessment

Accompanying drawing explanation

Fig. 1 is the block schematic illustration of the assessment indicator system of certain rural activity district underground water risk in embodiment.

Fig. 2 is the schematic flow sheet that the present invention sets up assessment indicator system.

Fig. 3 is certain rural activity district underground water risk stratification block plan in embodiment.

Embodiment

New rural activity district underground water risk stratification assessing zonings method provided by the invention, by tentatively setting up a set of comprehensive, representative assessment indicator system, method based on spatial discretization arranges sampled point, rational method is adopted to gather underground water water sample, and according to primary election index feature screening index, by calculating the rational weight of each index and the rational degree of membership of each sampled point index, comprehensive evaluation is made to local ground watering pollution risk, and subregion and classification are carried out to underground water risk.

The structure of comprehensive, representative assessment indicator system that the present invention sets up and the method for selecting index are: in the present invention, index system is divided into three layers, top layer is destination layer and local ground watering risk, middle layer is the classification that rule layer and Risk Evaluation Factors divide, and bottom is indicator layer and the quantifiable index relevant to Groundwater Contamination Risk.According to pollutant classification, rule layer comprises three, is respectively organic contaminant, inorganic pollutant and heavy metal.The agricultural chemicals that organic contaminant comprises rural activity district conventional comprises some Insecticides (tech) & Herbicides (tech)s, acaricide and germifuge; Inorganic pollutant comprises the inorganic ion of conventional polluted underground water, potential of hydrogen, total hardness, turbidity, chemical oxygen demand (COD) and biochemical oxygen demand etc.; Heavy metal comprises copper, zinc, iron, lead, nickel, cadmium, chromium, arsenic, mercury etc.First, set up the index system of three-decker, then from these indexs, tentatively choose pollutant index according to regional features of pollution, the principle of screening is representative, stability and can availability.

The points distributing method of spatial discretization of the present invention is evenly arranged on evaluation region by sampled point, adopts plum blossom to layout method, sampled point quantity is not less than 15, if there are civilian well or monitor well in locality, utilize existing well as far as possible, and Optimizing on this basis.

The present invention gathers the method for underground water water sample, is that water sample should be taken from underground water, needs the water in existing well to extract out, avoids gathered water sample to be the water that rainfall in the past or runoff directly enter in well, and sampling after stable level, clarification.

Index screening method of the present invention, be by Comparative indices between correlativity and compare their water quality standard, the higher (R of the correlativity if there is several index ²be greater than 0.9), and its water quality standard is close, then can replace these indexs by one of them index.Related coefficient calculates in MatlabR2009a software, and its computing formula is:

$r_{\mathrm{ij}}=\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}[x(i,j)-h\left(j\right)][x(i,k)-h\left(k\right)]}{\sqrt{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}[x(i,j)-h\left(j\right){]}^2{[x(i,j)-h\left(k\right)]}^2}}(i=1,\·\·\·,m,j=1,\·\·\·,n,k=1,\·\·\·,n)$

In formula, h (j)=average{x (i, j) | i=1 ..., m}, h (k)=average{x (i, k) | i=1 ..., m}, i are i-th sampled point, j and k is respectively jth and kth item index.

Weighing computation method of the present invention is the comprehensive importance being carried out measurement index by the product of the exponential sum that comprehensively exceeds standard unordered index two indices.The larger impact on risk of the index that comprehensively exceeds standard is larger, and meanwhile, unordered index is larger, shows that this index is the most remarkable to the variable effect of entire system, and the method combined both therefore adopting calculates comprehensive weight.Computing method are:

W＝[w(1),w(2),…,w(n)] _1×n

Wherein, W is the synthetic weights vector of n index, the integrated risk index that ρ (j) is j index, for the unordered index of j index.

Integrated risk index of the present invention, be consider according to the worst situation of evaluation result, the contribution of more serious pollutant to risk that exceed standard is larger, therefore can determine its importance by the frequency of grade residing for pollutant and risk class.Adopt the weight of following formula parameter:

$\mathrm{\ρ}\left(j\right)=\frac{\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}(j=1,\·\·\·,n,k=1,\·\·\·,5)$

In above formula, α (j, k) is for a jth index is lower than the number of (or higher than) k grade, and u (k) is score value corresponding to k grade, in the present invention, underground water risk falls into 5 types, and from u (1) to u, (5) are respectively 1,2,3,4,5.Owing to only making assessment to pollution risk here, health risk is not considered, therefore, do not do the harmfulness contrast between index.

Unordered index of the present invention is the weight size representing rural activity district underground water risk indicator with entropy, and this is because the physics meaning of entropy is that system is more unordered, entropy is larger, and system is more unordered, and the information comprised is also larger, therefore the size of effectiveness can be represented with entropy, or the size of weight.First need to do standardization to raw data, the standardized method of data is as follows:

For forward (being worth more Risks larger) index:

$y(i,j)=\frac{x(i,j)-f_2\left(j\right)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,m,j=1,\·\·\·,n)$

In formula, f ₁(j)=max{s (i, j) | i=1 ..., m}, f ₂(j)=min{s (i, j) | i=1 ..., m}

For reverse (being worth less risk larger) index:

$y(i,j)=\frac{f_1\left(j\right)-x(i,j)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,u,j=1,\·\·\·,v)$

Computing information entropy: $h\left(j\right)=-k\underset{j=1}{\overset{m}{\mathrm{\Σ}}}s(i,j)\ln s(i,j)$

In above formula, $s(i,j)=\frac{y(i,j)}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}y(i,j)}(i=1,\·\·\·,m,j=1,\·\·\·,n),$ When x (i, j)=0, f (i, j)=0.00001; K=1/ln (m).

Calculating entropy is weighed:

The calculating of rational degree of membership of the present invention, with tradition adopt fuzzy mathematics method calculate degree of membership unlike, it is considered herein that desired value not only can not belong to category-A but also belong to category-B according to the theory of fuzzy mathematics, it strictly can only be positioned at a grade interval, or belongs to category-A, belong to category-B, can not the two have concurrently, based on this, when certain index is positioned at a Risk interval, its degree of membership should be 1, and other interval is 0.Assuming that subordinated-degree matrix is

$T={\left[\begin{array}{cccc}{t}_{11}& t_{12}& \·\·\·& t_{1n}\\ t_{21}& t_{21}& \·\·\·& t_{2n}\\ \·& \·& & \·\\ \·& \·& \·\·\·& \·\\ \·& \·& & \·\\ t_{51}& t_{52}& \·\·\·& t_{5n}\end{array}\right]}_{5\×n},$ Then t _k,jcomputing method be:

For forward index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=5)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=5)\end{array}\right.$

For reverse index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=1)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=1)\end{array}\right.$

The calculating of comprehensive evaluation value of the present invention, be carry out matrix multiplication operation to the transposition of W and T, computing formula is: Q=WT', wherein, Q=[q (1), q (2), q (3), q (4), q (5)], by above-mentioned Matrix Multiplication with V={v (j), j=1,2,3,4,5}, v ₁, v ₂, v ₃, v ₄, v ₅value be respectively 100,80,60,40,20, an integrate score value between 20 to 100 can be obtained, so above-mentioned multiple attribute synthetical evaluation is become the result of a single index.

Classification of the present invention and assessing zonings, adopt ArcGIS to carry out subregion and classification to underground water risk, first, by corresponding with integrate score for each sample point coordinate, be loaded in ArcGIS, space interpolation is carried out to the discrete data of each sampled point, relatively IDW interpolation, Kriging interpolation, NaturalNeighbor interpolation and Spline interpolation four kinds of interpolation methods, optimize the interpolation method meeting spatial spreading evaluation result most, finally, adopt the method can obtain whole area of space continuous print water quality score, according to each water quality scoring rank interval, water quality result is spatially classified, obtain region continuous print underground water risk degree and zoning figure.

In other words, technical scheme of the present invention is the rural activity district underground water risk domain Evaluation on distribution method based on discrete sampling point, it is characterized in that by tentatively setting up a set of comprehensive, representative assessment indicator system, based on method collection, the detection underground water water sample of spatial discretization, and according to index feature screening index again, by calculating the rational weight of each index and the rational degree of membership of each sampled point index, comprehensive evaluation is made to local ground watering pollution risk, and subregion and classification are carried out to underground water risk.

1, build index system and index is screened: in the present invention, index system is divided into three layers, top layer is destination layer and local ground watering risk, middle layer is the classification that rule layer and Risk Evaluation Factors divide, and bottom is indicator layer and the quantifiable index relevant to Groundwater Contamination Risk.According to pollutant classification, rule layer comprises three, is respectively organic contaminant, inorganic pollutant and heavy metal.The agricultural chemicals that organic contaminant comprises rural activity district conventional comprises some Insecticides (tech) & Herbicides (tech)s, acaricide and germifuge; Inorganic pollutant comprises the inorganic ion of conventional polluted underground water, potential of hydrogen, total hardness, turbidity, chemical oxygen demand (COD) and biochemical oxygen demand etc.; Heavy metal comprises copper, zinc, iron, lead, nickel, cadmium, chromium, arsenic, mercury etc.First, set up the index system of three-decker, then from these indexs, tentatively choose pollutant index according to regional features of pollution, the principle of screening is representative, stability and can availability.

2, the layout of sampled point and the collection of underground water water sample: sampled point adopts the plum blossom method of layouting to be evenly arranged on evaluation region, sampled point quantity is not less than 15, if there are civilian well or monitor well in locality, utilize existing well as far as possible, and Optimizing on this basis.Water sample should be taken from underground water, needs the water in existing well to extract out, avoids gathered water sample to be the water that rainfall in the past or runoff directly enter in well, and sampling after stable level, clarification.

3, index screening method: compare their water quality standard, the higher (R of the correlativity if there is several index by the correlativity between Comparative indices ²>0.9), and its water quality standard is close, then can replace these indexs by one of them index.

4, weighing computation method: the comprehensive importance being carried out measurement index by the product of the exponential sum that comprehensively exceeds standard unordered index two indices.The larger impact on risk of the index that comprehensively exceeds standard is larger, and meanwhile, unordered index is larger, shows that this index is the most remarkable to the variable effect of entire system, and the method combined both therefore adopting calculates comprehensive weight.

(1) calculating of integrated risk index: consider according to the worst situation of evaluation result, the contribution of more serious pollutant to risk that exceed standard is larger, therefore can determine its importance by the frequency of grade residing for pollutant and risk class.Adopt the weight of following formula parameter:

$\mathrm{\ρ}\left(j\right)=\frac{\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}(j=1,\·\·\·,n,k=1,\·\·\·,5)$

In above formula, α (j, k) is for a jth index is lower than the number of (or higher than) k grade, and u (k) is score value corresponding to k grade, in the present invention, underground water risk falls into 5 types, and from u (1) to u, (5) are respectively 1,2,3,4,5.Owing to only making assessment to pollution risk here, health risk is not considered, therefore, do not do the harmfulness contrast between index.

(2) calculating of unordered index: entropy is a physics concept, and entropy more Iarge-scale system is more unordered, and the information comprised is also larger, therefore can represent the size of effectiveness with entropy, or the size of weight.First need to do standardization to raw data, the standardized method of data is as follows:

For forward (being worth more Risks larger) index:

$y(i,j)=\frac{x(i,j)-f_2\left(j\right)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,m,j=1,\·\·\·,n)$

In formula, f ₁(j)=max{s (i, j) | i=1 ..., m}, f ₂(j)=min{s (i, j) | i=1 ..., m}

For reverse (being worth less risk larger) index:

$y(i,j)=\frac{f_1\left(j\right)-x(i,j)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,u,j=1,\·\·\·,v)$

Computing information entropy: $h\left(j\right)=-k\underset{j=1}{\overset{m}{\mathrm{\Σ}}}s(i,j)\ln s(i,j)$

In above formula, $s(i,j)=\frac{y(i,j)}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}y(i,j)}(i=1,\·\·\·,m,j=1,\·\·\·,n),$ When x (i, j)=0, f (i, j)=0.00001; K=1/ln (m).

Calculating entropy is weighed:

(3) calculating of comprehensive weight: computing method are:

W＝[w(1),w(2),…,w(n)] _1×n

Wherein, W is the synthetic weights vector of n index, and ρ is the integrated risk index of each index, for the unordered index of each index.

5, the calculating of rational degree of membership, with tradition adopt fuzzy mathematics method calculate degree of membership unlike, it is considered herein that desired value not only can not belong to category-A but also belong to category-B according to the theory of fuzzy mathematics, it strictly can only be positioned at a grade interval, or belongs to category-A, belong to category-B, can not the two have concurrently, based on this, when certain index is positioned at a Risk interval, should be 1 in its degree of membership of this Risk interval, other interval is 0.Assuming that subordinated-degree matrix is

$T={\left[\begin{array}{cccc}{t}_{11}& t_{12}& \·\·\·& t_{1n}\\ t_{21}& t_{21}& \·\·\·& t_{2n}\\ \·& \·& & \·\\ \·& \·& \·\·\·& \·\\ \·& \·& & \·\\ t_{51}& t_{52}& \·\·\·& t_{5n}\end{array}\right]}_{5\×n},$

Then t _j,kcomputing method be:

For forward index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=5)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=5)\end{array}\right.$

For reverse index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=1)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=1)\end{array}\right.$

6, the calculating of comprehensive evaluation value: carry out matrix multiplication operation to the transposition of W and T, computing formula is: Q=WT', wherein, Q=[q (1), q (2), q (3), q (4), q (5)], by above-mentioned Matrix Multiplication with V={v (j), j=1,2,3,4,5}, v ₁, v ₂, v ₃, v ₄, v ₅value be respectively 100,80,60,40,20, an integrate score value between 20 to 100 can be obtained, so above-mentioned multiple attribute synthetical evaluation is become the result of a single index.

7, the classification of underground water risk and assessing zonings: first, by corresponding with integrate score for each sample point coordinate, be loaded in ArcGIS, space interpolation is carried out to the discrete data of each sampled point, relatively IDW interpolation, Kriging interpolation, NaturalNeighbor interpolation and Spline interpolation four kinds of interpolation methods, optimize the interpolation method meeting spatial spreading evaluation result most, finally, adopt the method can obtain whole area of space continuous print water quality score, according to each water quality scoring rank interval, water quality result is spatially classified, obtain region continuous print underground water risk degree and zoning figure.

Below in conjunction with Fig. 1 and Fig. 2, embodiments of the present invention are illustrated.

1, assessment indicator system is set up.This rural activity district long-term cropping type is paddy rice, and according to Zoned application agricultural chemicals feature, organic contaminant is selected to poison barnyard grass with poison, and inorganic pollutant chooses DOC, COD, BOD ₅, IC, nitrite, ammonia nitrogen, nitre nitrogen, chloride, sulfate, PH, dissolved oxygen DO, 12 indexs such as conductivity, heavy metal chooses 8 indexs such as chromium, manganese, iron, nickel, copper, zinc, cadmium, lead, 21 indexs altogether.Detect the desired value of each index, and consult its water quality standard, can with reference to the standard of the close pollutant of character for the standard do not had in GB.

2, do correlation analysis to each index, the related coefficient square between index is all less than 0.9, shows to have independence between each index, and therefore, these 21 indexs are all as this underground water risk assessment.

3, according to the following formula, integrated risk index is calculated.

$\mathrm{\ρ}\left(j\right)=\frac{\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\underset{k=1}{\overset{5}{\mathrm{\Σ}}}\mathrm{\α}(j,k)\×u\left(k\right)}(j=1,\·\·\·,n,k=1,\·\·\·,5)$

In above formula, α (j, k) is for a jth index is lower than the number of (or higher than) k grade, and u (k) is score value corresponding to k grade, in the present invention, underground water risk falls into 5 types, and from u (1) to u, (5) are respectively 1,2,3,4,5.

4, according to following formula, standardization is done to raw data X.

For forward (being worth more Risks larger) index:

$y(i,j)=\frac{x(i,j)-f_2\left(j\right)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,m,j=1,\·\·\·,n)$

In formula, f ₁(j)=max{s (i, j) | i=1 ..., m}, f ₂(j)=min{s (i, j) | i=1 ..., m}

For reverse (being worth less risk larger) index:

$y(i,j)=\frac{f_1\left(j\right)-x(i,j)}{f_1\left(j\right)-f_2\left(j\right)}(i=1,\·\·\·,u,j=1,\·\·\·,v)$

5, on standardized basis, according to following formula, entropy power is calculated.

Wherein, $h\left(j\right)=-k\underset{j=1}{\overset{m}{\mathrm{\Σ}}}s(i,j)\ln s(i,j)$

In above formula, $s(i,j)=\frac{y(i,j)}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}y(i,j)}(i=1,\·\·\·,m,j=1,\·\·\·,n),$ When x (i, j)=0, f (i, j)=0.00001; K=1/ln (m).

6, according to following formula, comprehensive weight is calculated.

W＝[w(1),w(2),…,w(n)] _1×n

Wherein, W is the synthetic weights vector of n index, and ρ is the integrated risk index of each index, for the unordered index of each index.

7, according to following formula, subordinated-degree matrix is calculated.

For forward index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=5)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=5)\end{array}\right.$

For reverse index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n,k=1)\\ 0& x(i,j)>s(k,j)& (i=1,\·\·\·,m;j=2,\·\·\·,n;k=1)\end{array}\right.$

8, according to following formula, integrate score is calculated.

Q=WT', wherein, Q=[q (1), q (2), q (3), q (4), q (5)], by above-mentioned Matrix Multiplication with V={v (j), j=1,2,3,4,5}, v ₁, v ₂, v ₃, v ₄, v ₅value be respectively 100,80,60,40,20

9, the classification of underground water risk and assessing zonings.First, by corresponding with integrate score for each sample point coordinate, be loaded in ArcGIS, space interpolation is carried out to the discrete data of each sampled point, relatively IDW interpolation, Kriging interpolation, NaturalNeighbor interpolation and Spline interpolation four kinds of interpolation methods, it is best that result shows to adopt Spline method interpolation, namely the interpolation result of the method is adopted, then, according to 0 ~ 20 in ArcGIS, 20 ~ 40, 40 ~ 60, 60 ~ 80, 80 ~ 100, risk integrative score is divided into five classes, wherein distinguish corresponding I class, II class, III class, IV class and V class risk.According to above-mentioned steps, obtain this rural activity district underground water classification of risks classification as shown in Figure 3.The bay status of this rural activity district underground water risk class not only can be obtained from this figure, therefrom can also judge underground water potential risk district, as shown in Figure 3, the underground water risk of the southeast, northeast, western part, east is all higher, therefore, can conclude that these areas are for pollution source, and direction of groundwater flow is from the southeast northwestwards, so the northwestward should be the region of future period groundwater contamination aggravation, be potential risks districts.Therefore, the assessment of the underground water classification of rural activity district and underground water potential risk can be realized by the present invention.

Claims (5)

1. a rural activity district underground water risk stratification assessing zonings method, by setting up a set of index system with judging, sampled point is arranged with the method for spatial discretization, gather underground water water sample, and according to primary election index feature screening index, by the degree of membership of the weight and each sampled point index that calculate each index, comprehensive evaluation is made to local ground watering pollution risk, and subregion and classification are carried out to underground water risk;

Wherein, the calculating of weight is the comprehensive importance being carried out measurement index by the product of integrated risk exponential sum unordered index two indices; Computing formula is:

W＝[w(1)，w(2)，…，w(n)] _1×n；

Wherein, W is the synthetic weights vector of n index, the integrated risk index that ρ (j) is j index, for the unordered index of j index;

Integrated risk index determines its importance with the frequency of grade residing for pollutant and risk class; Computing formula is:

$\mathrm{\ρ}\left(j\right)=\frac{\underset{k=1}{\overset{5}{\Σ}}\mathrm{\α}(j,k)\×u\left(k\right)}{\underset{j=1}{\overset{n}{\Σ}}\underset{k=1}{\overset{5}{\Σ}}\mathrm{\α}(j,k)\×u\left(k\right)},(j=1,...,n,k=1,...,5);$

In above formula, α (j, k) is for a jth index is below or above the number of k grade, and u (k) is score value corresponding to k grade, if underground water risk falls into 5 types, then u (1) to u (5) is respectively 1,2,3,4,5;

Unordered index is the weight size representing rural activity district underground water risk indicator by entropy power, and first do standardization to raw data, the standardized method of data is as follows:

For forward index:

$y(i,j)=\frac{x(i,j)-f_2\left(j\right)}{f_1\left(j\right)-f_2\left(j\right)},(i=1,...,m,j=1,...,n);$

In formula, f ₁(j)=max{s (i, j) | i=1 ..., m}, f ₂(j)=min{s (i, j) | i=1 ..., m}, s (i, j) represent the threshold value of a jth index in i grade, $s(i,j)=\frac{y(i,j)}{\underset{i=1}{\overset{m}{\Σ}}y(i,j)},(i=1,...,m,j=1,...,n);$

For reverse index:

$y(i,j)=\frac{f_1\left(j\right)-x(i,j)}{f_1\left(j\right)-f_2\left(j\right)},(i=1,...,u,j=1,...,v);$

Computing information entropy: $h\left(j\right)=-k\underset{j=1}{\overset{m}{\Σ}}s(i,j)\ln s(i,j);$

In above formula, when x (i, j)=0, f (i, j)=0.00001; K=1/ln (m);

Calculating entropy is weighed:

Degree of membership be desired value and risk class grade scale are substituted into selected by membership function calculate, subordinated-degree matrix T and membership function t=t (k, j) are as follows respectively:

then t _ijcomputing method be:

For forward index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,...,m;j=2,...,n,k=5)\\ 0& x(i,j)>s(k,j)& (i=1,...,m;j=2,...,n;k=5)\end{array}\right.;$

For reverse index:

$t(k,j)=\left\{\begin{array}{ccc}1& x(i,j)\≤s(k,j)& (i=1,...,m;j=2,...,n,k=1)\\ 0& x(i,j)>s(k,j)& (i=1,...,m;j=2,...,n;k=1)\end{array}\right.;$

Described comprehensive evaluation is carried out matrix multiplication operation by synthetic weights vector with the transposition of subordinated-degree matrix and is obtained, and computing formula is: Q=WT ', wherein, Q=[q (1), q (2), q (3), q (4), q (5)], by above-mentioned Matrix Multiplication with V={v (j), j=1,2,3,4,5}, v ₁, v ₂, v ₃, v ₄, v ₅value be respectively 100,80,60,40,20, an integrate score value between 20 to 100 can be obtained, so above-mentioned multiple attribute synthetical evaluation is become the result of a single index.

2. rural activity district according to claim 1 underground water risk stratification assessing zonings method, wherein, index system is divided into three layers, and top layer is destination layer, and middle layer is rule layer, and bottom is indicator layer;

According to pollutant classification, the pollutant class of rule layer separately includes organic pollutants, inorganic pollutant and heavy metal;

Organic contaminant comprises Insecticides (tech) & Herbicides (tech) and the germifuge in rural activity district;

Inorganic pollutant comprises the inorganic ion of conventional polluted underground water, potential of hydrogen, total hardness, turbidity, chemical oxygen demand (COD) and biochemical oxygen demand;

Heavy metal comprises copper, zinc, iron, lead, nickel, cadmium, chromium, arsenic and mercury.

3. rural activity district according to claim 1 underground water risk stratification assessing zonings method, wherein, the method for spatial discretization arranges sampled point, is be evenly arranged on evaluation region by sampled point, and adopt plum blossom to layout method, sampled point quantity is not less than 15; When gathering underground water water sample, underground water should be taken from.

4. rural activity district according to claim 1 underground water risk stratification assessing zonings method, wherein, primary election index feature be by Comparative indices between correlativity and water quality standard more each other determine.

5. rural activity district according to claim 1 underground water risk stratification assessing zonings method, wherein, the evaluation of classification and subregion, adopt ArcGIS to carry out subregion and classification to underground water risk: each sample point coordinate is corresponding with integrate score value, be loaded in ArcGIS, space interpolation is carried out to the discrete data of each sampled point, from IDW interpolation, Kriging interpolation, the interpolation method meeting spatial spreading evaluation result is most selected in NaturalNeighbor interpolation and Spline interpolation four kinds of interpolation methods, finally, adopt the method can obtain whole area of space continuous print water quality score, according to each water quality scoring rank interval, water quality result is spatially classified, obtain region continuous print underground water risk degree and zoning figure.