An Analytical Approach for Initial Allocation of Discharge Permits with Consideration of the Water Environmental Capacity and Industrial Technical Feasibility

: The tradable discharge permit is an efﬁcient marketing twist to deal with negative externality of water pollution, and initial allocation discharge permits (IADP) is a tough prerequisite for a health tradable discharge permit system. However, the allocation rules focusing on both water environment quality attainment and economic-technical feasibility have not been achieved. Therefore, a three-tier analytical framework including control units, industries, and enterprises has been established for satisfying the requirements of water environment quality attainment and technical feasibility. The framework includes three calculation modules and one justiﬁcation module. A case study of the framework is carried out in the Changzhou City of Taihu Lake, China. Through comparison of the results of industry IADP, the technology-based initial allocation discharge permits scheme in control unit 7 is adopted as the ﬁnal initial allocation discharge permits scheme, and the reduction plan chemical industry is carried out in control unit 8; ﬁnally, the initial allocation discharge permits results after reduction are employed as the ﬁnal initial allocation discharge permits scheme. This three-tier analytical framework could ensure environmental protection, equality, efﬁciency, and make the scheme be more acceptable for enterprise stakeholders. In addition, it could support for optimization of industrial layouts and industrial structure upgrading to some degree.


Introduction
The tradable discharge permit (TDP) has been adopted as an efficient marketing twist to deal with the negative externality of water pollution in many regions caused by the rapid economic and demographic growth [1][2][3]. It was implemented in the management of atmospheric pollution and water pollution in the United States after Dales (1968) first promoted the concept of TDP in 1968 [4,5]. The initial allocation of discharge permits (IADP) is a tough issue because it is the prerequisite of a free tradable permits market [6]. The guarantee of IADP equality and efficiency would determine the acceptability of all stakeholders including enterprise and government, and finally achieve the optimum assembly of cost efficiency, environmental protection and economic growth.
The approach of IADP includes two main categories: auction and grandfathering [7]. Auction is a significant means for IADP, and is usually used in permits distribution of uniformly mixed pollutant such as carbon dioxide; it has advantages of innovation incentives and building competitive markets [1]. However, for the local wastewater pollutants, the polluters tend to obey this approach reluctantly due to the spare expense burden.
Grandfathering is a method in which enterprises receive free permits based on their historical discharges, and free allocation of permits and possible benefits of trading make of WECIADP and TIADP (technology-based IADP) is compared, and the corresponding scheme is adopted. In addition, during the process of WECIADP calculation, the influence coefficient and response coefficient which indicate the region dominant industry is adopted, which make the results more favorable to the local economic development.
This paper is organized as follows: Section 2 emphasizes the methodology, including the introduction of the three-tier analytical framework, and the detailed analysis process. Section 3 illustrates the study region and data sources. Section 4 offers a discussion of a range of analysis results and applications of the case study. The last section evaluates the value and deficiency of the analytical framework.

Analytical Framework
A three-tier analytical framework including control units-industries-enterprises is established for IADP ( Figure 1). It includes three calculation modules and one justification module. WEC and the total permissible industrial discharge (PID) calculation is in the level of control units, and enterprises TIADP is in the level of enterprises. In the level of industry, it contains two calculation process, WECIADP and TIADP. In the judgment module, the results of industry WECIADP and TIADP are compared, then the appropriate scheme of IADP is adopted.
IADP, the industry IADP is introduced as the middle tier. Identification of enterprises IADP under the same industry types make it more technical feasible. The industry IADP which comes from the WEC of the control unit (WECIADP) is obtained through the information entropy method, and the industry IADP based on technology (technologybased IADP) is attained through summation of enterprises IADP under the same kind of industry type. Then, the value of WECIADP and TIADP (technology-based IADP) is compared, and the corresponding scheme is adopted. In addition, during the process of WECIADP calculation, the influence coefficient and response coefficient which indicate the region dominant industry is adopted, which make the results more favorable to the local economic development.
This paper is organized as follows: Section 2 emphasizes the methodology, including the introduction of the three-tier analytical framework, and the detailed analysis process. Section 3 illustrates the study region and data sources. Section 4 offers a discussion of a range of analysis results and applications of the case study. The last section evaluates the value and deficiency of the analytical framework.

Analytical Framework
A three-tier analytical framework including control units-industries-enterprises is established for IADP ( Figure 1). It includes three calculation modules and one justification module. WEC and the total permissible industrial discharge (PID) calculation is in the level of control units, and enterprises TIADP is in the level of enterprises. In the level of industry, it contains two calculation process, WECIADP and TIADP. In the judgment module, the results of industry WECIADP and TIADP are compared, then the appropriate scheme of IADP is adopted.

WEC and PID Calculation of Control Units
The WEC is prerequisite for IADP determination and is not allowed to be exceeded by total emission amount, so as to maintain local water environment quality. As domestic living and agriculture are the basic activities for human living, and the service industry has a low discharge amount, PID is set as total mass limitation of industrial discharge in a control unit, which is equal to WEC minus domestic discharge, agriculture discharge, Sustainability 2021, 13, 13743 4 of 16 service industry discharge and margin of safety. Consequently, the water environment condition could be better protected, and the water quality targets could be better achieved. Meanwhile, the same industry may have different IADP just because of their different spatial location and relative environment endowment, with all other things being equal. Thus, the IADP may be an effective forcing mechanism for industry layout optimization, and realize efficient utilization of WEC fully.
A one-dimensional water quality model is employed to calculate the water environment capacity of each control unit. To realize spatial expression of water environmental endowment and be possible for the further IADP of industry and enterprise, the concept of control units is introduced to realize division of spatial units in an appropriate spatial scale. Aquatic ecological function regionalization at the watershed scale was first put forward in America in 1980 and provided support for the management of American total maximum daily loads (TMDL). To achieve basin water quality target management in China [15], Meng then put forward a pollutant total amount control technique in terms of control units [16]. Lei explained the detailed process of control unit divisions [17]. The divisions of control units, considering both the river basin characteristics and administrative districts, increases the availability of water quality management. The study area is divided into 13 control units according to the natural-social condition, in which control units 1, 2, 3 and 4 belong to Liyang, control units 5 and 6 belong to Jintan, and the Urban District includes control units 7, 8, 9, 10, 11, 12 and 13. Considering the local river characteristic with medium river width and relative stable river velocity during the whole year in Changzhou, a one-dimensional water quality model is employed to calculate the water environment capacity of each control unit in Changzhou [18,19]: where WEC a is the water environment capacity of control unit a, Q 0 is the main river design flow, q is the tributary design flow, C 0 is the pollutants background concentration in control units, Cs is the pollutants objective concentration in control units, x is the length of the river in control units, K is the pollutant degradation coefficient, and u is the river average flow velocity of control units. The PID in each control unit is calculated according to the proportion of current discharge amount of each pollution source, including agriculture source, industry source, service industry source, domestic source and margin of safety, which is about 10% of WEC according to the research result of Fu et al. [20]. The corresponding value in every control unit is calculated through gridded GIS technology, except the industry source; the detailed calculation process is shown in the paper [21].
in which, PID a means the PID in control unit a, WEC a is the WEC in control unit a, and W a means the weight of industry source in control unit a.

Technology-Based IADP Calculation of Enterprises
For a certain industry, due to the limitation of economic and technical feasibility, cleaner production standard and industry emission standard, the same industrial type always owns the relatively consistent production process and pollutants emission intensity to some degree [22]. Due to the information absence of the best available technology, the average emission intensity in a certain industrial type can be regarded as the normal discharge level in a certain area. The best available technology is replaced by average discharge intensity as a benchmark in this study. The preliminary enterprises IADP amount is calculated through multiplying average discharge intensity of the whole study region and enterprise output value under the same industrial type. The enterprise which owns lower discharge intensity would be allocated relatively more discharge permits. Therefore, it may encourage enterprises to take more clean production measures.
where, ADI b indicates the average discharge amount of industry b of the whole study region, DA b indicates the discharge amount of industry b of the whole study region, IOV b indicates the industry output value of industry b of the whole study region, ETI ADP abc indicates the IADP of enterprise c of industry b in the control unit a, and EOV abc indicates the enterprise output value of enterprise c of industry b in control unit a.

IADP Calculation of Industries
The module of industry IADP calculation contains two processes. One is WECIADP calculation based on the results of WEC and PID calculation module. Each control unit has different industrial types; the information entropy method is adopted to calculate the weight of different industrial types in the twelve control units separately. There are six criteria chosen in the weight assignment of industrial types, including industry output value, influence coefficient, response coefficient, pollutant production, pollutant discharge, and generation and discharge coefficient. Those criterions were chosen by the principles of equality and efficiency. The influence coefficient and response coefficient are introduced to recognize the dominant industry under the regional current economic development status and assign high weight value to the dominant industry; thus, providing a forcing mechanism for industry structure upgrading.
in which, WECI ADP ab means the WECIADP of industry b in control unit a, IW ab means the industry weight of industry b in control unit a, and PID a means the PID in control unit a. The detailed process of information entropy method in this study is below: where h ai indicates the entropy of index i in control unit a, P abi indicates the probability of index i of industry b in control unit a, m indicates the index m including industry output value, influence coefficient, response coefficient, pollutant production, pollutant discharge, or generation and discharge coefficient.
where β ai indicates entropy weight of index i in control unit a.
where IW ab indicates weight of industry b in control unit a. The other process in the module of industry IADP calculation is TIADP calculation of industry.
ITIADP is attained through summation of each enterprise IADP under the same kind of industry type in a certain control unit.
where ITI ADP ab indicates the IADP of industry b in control unit a.

IADP Scheme Justification
In the justification module, the value of IWECIADP and ITIADP are compared, then the fit scheme of IADP is adopted according to the corresponding results.
If the value of IWECIADP is larger than ITIADP, it means the preliminary enterprises IADP scheme could guarantee the PID would not be exceed by IADP, and finally maintain water environment quality. Thus, the preliminary enterprises TIADP and ITI-ADP are adopted as the final IADP scheme. The spare PID could be allocated for new industry enterprise.
If the value of IWECIADP is lower than ITIADP, it means the WEC may be exceeded by IADP once the preliminary enterprises IADP scheme is accepted. Therefore, the reduction plan of a certain industry must be put forward. To ensure the adopted IADP amount is no more than IWECIADP, each relative enterprise of the certain industry has to improve the level of clean production and sewage treatment; thus, to reduce emission intensity per value, the enterprise IADP is recalculated according to WEC-based IADP and the enterprises output value.
IETI ADP abc = EOV abc × I ADI ab (13) where, IOV ab indicates the industry output value of industry b in control unit a, I ADI ab indicates the theoretical average discharge intensity of industry b in control unit a, IETI ADP abc indicates the ideal IADP of enterprise c of industry b in control unit a, EOV abc indicates the enterprise output value of enterprise c of industry b in the control unit a. I ITI ADP ab indicates the ideal IADP of industry b in control unit a. As a result, the ideal enterprise TIADP and ITIADP is adopted as the final IADP scheme.

Data Sources and Study Area
Changzhou City, which is a prefecture-level city in China (Figure 2), is used as the study area. It contains an urban district and two county areas (Liyang and Jintan) and can be divided into 13 control units according to the river basin characteristics. The Urban District contains control units 7, 8, 9, 10, 11, 12 and 13; Liyang contains control units 1, 2, 3, and 4; and Jintan contains control units 5 and 6. The locations of control units are shown in Figure 2. The population is 4.71 million people, and the gross domestic product is 0.92 billion yuan. The local dominated industry type is the chemical industry and manufacturing industries, such as the textile industry. It is a region in which the contradiction of the economy and the environment is quite striking; the high population and economic density of Changzhou induces widespread contamination of the local environment. Although the high treatment ratio of water pollution, the discharge amount from the secondary industry still damages the local water environment. Based on the preliminary analysis, control unit 7 has larger water environment capacity and relative less industrial discharge, in contrast to control unit 8. The industry types with the highest discharge amount in Changzhou City are also located in the two control units, including the textile industry in control unit 7 and the chemical industry in control unit 8. They can be used to present the whole calculation process of the three-tier model, so they are selected for typical analysis in this study. The data is mainly collected from the 2018 Changzhou Statistical Yearbook (pollution census data, water resources bulletin) from the year 2018 [23]. Types of pollutants that are discharged into the water body are highly variable depending on economic activities and human living, including Chemical Oxygen Demand (COD), Biochemical oxygen demand (BOD), NH4 + -N, phosphorus, and heavy metals. There is no need to analyze all pollutants in the study. According to the water quality status of Changzhou, COD is adopted as the pollutant index. In this study, COD was determined using dichromate as the oxidant in acidic solution. Based on the preliminary analysis, control unit 7 has larger water environment capacity and relative less industrial discharge, in contrast to control unit 8. The industry types with the highest discharge amount in Changzhou City are also located in the two control units, including the textile industry in control unit 7 and the chemical industry in control unit 8. They can be used to present the whole calculation process of the three-tier model, so they are selected for typical analysis in this study.

Results of Three-Tier Model
The data is mainly collected from the 2018 Changzhou Statistical Yearbook (pollution census data, water resources bulletin) from the year 2018 [23]. Types of pollutants that are discharged into the water body are highly variable depending on economic activities and human living, including Chemical Oxygen Demand (COD), Biochemical oxygen demand (BOD), NH 4 + -N, phosphorus, and heavy metals. There is no need to analyze all pollutants in the study. According to the water quality status of Changzhou, COD is adopted as the pollutant index. In this study, COD was determined using dichromate as the oxidant in acidic solution.

Results of Three-Tier Model
Through the calculation of the three-tier analytical framework with the case study of Changzhou City, China, the IADP amount of control unit, industries and enterprises is achieved (Figure 3). In this study, the study objects are focused on the three tiers of control units-industry-enterprise. Therefore, the regions with industry-dominant structure should be selected as representative study areas. Meanwhile, the two different regions with rich WEC and poor WEC should be both considered to analyze the impact of WEC amount on IADP. As a result, to make clear the analysis process of the model framework, the control unit 7 and 8 are chosen as researched regions for typical industry initial allocation of discharge permits. There are 13 and 18 kinds of industries in control unit 7 and control unit 8, respectively. The purpose of this study is to show the process of the three-tier analytical approach for IADP; thus, there is no need to analyze all the industry types. Meanwhile, in order to achieve the control of main pollution sources, the textile industry and the chemical industry, which are the main pollution industries, are adopted as typical industries for IADP research. unit 7 and control unit 8, respectively. The purpose of this study is to show the process of the three-tier analytical approach for IADP; thus, there is no need to analyze all the industry types. Meanwhile, in order to achieve the control of main pollution sources, the textile industry and the chemical industry, which are the main pollution industries, are adopted as typical industries for IADP research.
In the control unit tier of analytical framework, the PID amount of control unit 7 and 8 is 4231.03 t and 394.63 t; in the enterprises tier, the average discharge intensity of textile industry and chemical industry is 1.78 kg per 10,000 yuan and 1.24 kg per 10,000 yuan, respectively. Then, the enterprises IADP based on each industry is achieved. In the industries tier, the WECIADP and TIADP of the textile industry in control unit 7 and the chemical industry in control unit 8 are calculated separately. As the TIADP of the textile industry in control unit 7 is less than its WECIADP, the TIADP is acceptable; then, the IADP amount of each tier is arranged according to the technology-based scheme. As the TIADP of the chemical industry of control unit 8 is larger than its WECIADP, the reduction scheme should be put forward for chemical industry in control unit 8.   (2), the permissible industrial discharge of each control unit is calculated according to the water environment capacity, historical pollutant discharge amount, and emission reducing potential of each pollutant source ( Table 1). The result indicates that the PID value is quite different, along with the change of WEC and local industrial structure. It can be assigned into four categories according to the sort of WEC amount and proportion of industrial value in each control unit: rich WEC In the control unit tier of analytical framework, the PID amount of control unit 7 and 8 is 4231.03 t and 394.63 t; in the enterprises tier, the average discharge intensity of textile industry and chemical industry is 1.78 kg per 10,000 yuan and 1.24 kg per 10,000 yuan, respectively. Then, the enterprises IADP based on each industry is achieved. In the industries tier, the WECIADP and TIADP of the textile industry in control unit 7 and the chemical industry in control unit 8 are calculated separately. As the TIADP of the textile industry in control unit 7 is less than its WECIADP, the TIADP is acceptable; then, the IADP amount of each tier is arranged according to the technology-based scheme. As the TIADP of the chemical industry of control unit 8 is larger than its WECIADP, the reduction scheme should be put forward for chemical industry in control unit 8.

IADP of Each Industry
(1) WECIADP of each industry Calculation module of control units: According to the module of WEC and PID calculation in the framework model and Formula (2), the permissible industrial discharge of each control unit is calculated according to the water environment capacity, historical pollutant discharge amount, and emission reducing potential of each pollutant source ( Table 1). The result indicates that the PID value is quite different, along with the change of WEC and local industrial structure. It can be assigned into four categories according to the sort of WEC amount and proportion of industrial value in each control unit: rich WEC and industry-dominant structure, poor WEC and industry-dominant structure, rich WEC and agriculture-dominant structure, poor WEC and agriculture-dominant structure. For example, control unit 8 is an industry-oriented economic structure, while it has high WEC; therefore, both the PID proportion and value are high. However, the economic structure in control unit 7 is also industry-oriented, due to the lower WEC, and the PID value is low. In

Calculation module of industries:
In order to achieve the IADP of each industry in the top-down way (based on WEC), the weight of each industry in a certain control unit is calculated through the information entropy method, according to Formulas (7)-(10) ( Table 2), It is a comprehensive assessment of industry output value, industry importance, pollution production, and waste discharge by six indicators. Through calculation by combining the PID results in Table 1, the WECIADP of the textile industry in control unit 7 is 271.22 t and the WECIADP of the chemical industry in control unit 8 is 1877.91 t, according to Formula (5). (2) Technology-based IADP of each industry Calculation module of enterprises: In order to achieve the IADP of each industry in the bottom-up way (based on technology), the enterprises IADP of the same industry types in one control unit is calculated through multiplying industry average performance of the whole study region by enterprises output value according to Formula (11); there are 243 textile enterprises in control unit 7 and 143 chemical enterprises in control unit 8.
Calculation module of industries: The TIADP can be achieved by summation of enterprises IADP according to Formulas (3) and (4); the TIADP of the textile industry in control unit 7 is 223.38 t and the TIADP of the chemical industry in control unit 8 is 2395.42 t ( Table 2).

IADP Scheme Justification
The first principle of the model framework in this study is to guarantee that the WEC of every control unit would not be exceeded by pollutants discharge. Hence for industry, the WECIADP is becoming the absolute upper limitation, and the actual IADP amount could not be more than the WECIADP.
Justification module: The results of IADP based on WEC and technology are shown in Table 3; it can be found that in control unit 7, the WECIADP value of the textile industry is higher than TIADP; thus, the technology-based initial allocation approach is acceptable.
The IADP amount of the textile industry is 223.38 t, and the IADP of each textile enterprise is shown in Table 3. According to the justification module, the TIADP scheme is adopted as the final IADP scheme. In control unit 8, the WECIADP value of textile industry is lower than TIADP, thus the reduction plan of chemical industry must be put forward. To ensure the adopted IADP amount is no more than the WECIADP, each chemical enterprise has to improve the level of clean production and sewage treatment; thus, to reduce emission intensity per value, the enterprise IADP is recalculated according to the WECIADP and enterprises output value according to Formulas (12)- (14) in the justification module. The results are shown in Table 3. The recalculated results of IADP based on WEC is adopted as the final IADP scheme. The detailed reduction plan for regions, industries and enterprises can be ensured. The same industry ought to own the relative equal discharge level ideally. The reduction plan only according to the proportion of output value would reflect equality by improving the level of industry emissions without discrimination of any enterprise under the same industry, and would be acceptable for stakeholders. There are 243 enterprises of the textile industry in control unit 7 and 143 enterprises of the chemical industry in control unit 8. The purpose of this study is to show the process of the three-tier analytical approach for IADP; there is no need to analyze all the enterprises. Meanwhile, in order to achieve the control of main pollution sources, the two enterprises which are the main pollution sources are adopted as typical enterprises for IADP research in control unit 7 and 8, separately (Table 3).

Results of Normal Model
As mentioned in the introduction section, the normal approach for IADP is either based on regions or enterprises, without constructing the connection of IADP between regions and enterprises; therefore, the results cannot contain both water quality environment attainment and technology feasibility at the same time.
Through various weight allocation methods, the IADP results among different regions would be achieved. However, the process of how regions IADP affects the local enterprises still needs to be revealed. In the other way, the enterprises IADP could be calculated through their discharge levels or the weight location method, but the total of all enterprises IADP may exceed the available WEC.
As usual, during the weight allocation of IADP, the selection of indicators often take economy amount, discharge intensity and water environment into consideration. In general, the influences of economy development, such as industrial structure and dominant industries, are always neglected during the process of weight allocation of IADP in the previous research [12].

Advancement of Three-Tier Model
Different from the single perspective analysis in the normal IADP methods, which either focus on regions or enterprises, the three-tier model realizes the IADP from multiple perspectives, including control units-industries-enterprises; therefore, it connects the linkages between regions and enterprises, and the IADP results achieves the comprehensive benefits of water quality attainment and technology feasibility.
To determine the industry IADP based on WEC, six indexes are taken into consideration for identification of industry weight, including the influence coefficient and the response coefficient through information entropy. Consequently, the results of industry weight could imply the importance of regional dominant industry, and allow dominant industry to own more IADP. Therefore, the results of IADP would become a possible tool for stimulating industrial structure upgrading. The weight of each industry in a certain control unit is calculated through information entropy method according to Formulas (7)-(10) ( Table 4). It is a comprehensive assessment of economic contribution, industrial importance, and technical feasibility for a certain industry through the six indexes. In general, the influence of industrial importance is always neglected during the process of weight allocation of IADP in the previous research [12]. In order to reveal the advancement, with consideration of the influence of industrial importance on weight calculation of IADP by introducing the indexes of the influence coefficient and response coefficient, two situations are established and compared including the current six-indexes method and the four-indexes method. In the four-indexes method, only the four-indexes-including industry output value, pollution production, waste discharge, ratio of pollution production and water discharge-are considered, while in the six-indexes method, the index of industrial influence coefficient and response coefficient are added to indicate the industry importance. Through the comparison of these two methods, the results show that once the function of industry importance is taken into consideration, the weight of the textile industry in control unit 7 reduces from 0.7636 to 0.6873, although it is still the industry with local highest weight value; meanwhile, the weight value of electric equipment and machinery industry, which is the local dominant industry, rises from 0.0058 to 0.0176. In the control unit 8, although the chemical industry is the industry with the highest value in both of the two situations, the weight value reduces from 0.4953 to 0.4438 due to the relative low industry importance; at the same time, the weight value of the metal smelting and rolling processing industry, which is the local dominant industry, rises from 0.0877 to 0.1060. Through adopting the six-indexes method, the industry with more economic contribution and industry importance would be allocated more IADP. Thus, the IADP results of six-indexes method would support economic development and industrial upgrading to some degree, compared with the previous normal method. Therefore, the six-indexes calculation method is adopted in the weight allocation in this study.   Table 1, the WECIADP of the textile industry in control unit 7 is 271.22 t and the WECIADP of chemical industry in control unit 8 is 1877.91 t, according to Formula (5).
In the background of achieving the goal of carbon neutrality, the initial allocation of emission permits of carbon is quite important in the carbon trade system [24,25], with consequences on the air environment system [26,27]. Different from the IADP of carbon, the water environment capacity has quite significant spatial heterogeneity and plays a key role for IADP calculation. The water environment process is influenced by the natural condition and climate change [28]. The basic water environment endowment revelation based on the proper model is the prerequisite for the accurate calculation of IADP [29]. Integrated water environment management requires significant environmental policy tools [30]. The IADP based on WEC and technology in this study can provide an effective policy tool for integrated water environment management.

Discussion
Different from most previous studies [31,32], which focused on IADP either from the aspect of regions or from the aspect of industries, with less consideration of the comprehensive influences of water environment capacity and industrial structure on IADP, in this paper we established a three-tier analytical framework including control units, industries, and enterprises for satisfying the requirements of water environment quality attainment and economic-technical feasibility. A one-dimensional water quality model and information entropy method are chosen to help establish the framework. The influence coefficient and response coefficient which indicate the local dominant industry are employed in the weight allocation of industry comparison with the previous method. It achieves the successful application of analytical framework in study area. The allocation rules of IADP realize both water environment quality attainment and economic-technical feasibility. The amount of IADP of each enterprise can guarantee the entire local water environment quality attainment, while also being available for each enterprise when considering the average pollutants discharge level.
In the previous research, some scholars realized the IADP under the scale of regions or rivers [33]. The discharge permits of sub-regions were calculated without considering the permits of industry or enterprises of internal sub-regions, while other scholars realized the IADP under the scale of enterprises with the same industry type [2]. The integrated approach with combination of regions IADP and enterprises IADP has still not been achieved based on the previous research. Mesbah et al. developed a real time method for IADP of dischargers in rivers without considering the technology feasibility [34]. The previous IADP methods could not guarantee water quality attainment and technology feasibility at the same time. The relationship among regions, industries and enterprises has not been correctly recognized and established.
The deficiencies have been resolved through the three-tier framework established in this paper. Taking Changzhou City as an example, through the comparison of the previous research methods and the three-tier approach of this study, the significance of the three-tier approach in this study is revealed. Based on the existing IADP methods under the scale of watershed [35], it can be only found that PID of control unit 7 and 8 is 394.63 t and 4231.03 t. The enterprises IADP of the internal control units is unclear. Based on the existing IADP methods under the scale of enterprises [36], the TIADP of enterprises could be calculated (shown in Table 3). The existing methods could not provide detailed policy making and permit allocation from regions to enterprises for government managers. Through the three-tier framework, rather than calculating the PID of regions and TIADP, the WECIADP and TIADP of each industry could be calculated at the same time. Through the justification module, the TIADP of the textile industry in control unit 7 is adopted as the final acceptable amount, which was 223.38 t. The WECIADP of the chemical industry in control unit 8 is adopted as the final acceptable amount, which was 1877.91 t. The reduction scheme should be implicated for the chemical industry in control unit 8. The final enterprises IADP amount which could guarantee both of water quality attainment and technology feasibility is shown in Table 3.
During the calculation of PID, the amount of other pollutant sources and the safety of margin is allocated according to the weight at the same time, such as farming, livestock, secondary industry and service industry [37]. Thus, the IADP amount of industries and enterprise can maintain the local water quality attainment and ensure the original intention of IADP for water quality protection. Each control unit has a certain PID. In the other way, the average discharge intensity and enterprises output value are adopted for enterprises IADP calculation according to the industrial types under the same control units. Thus, the results would be more technically feasible. In the calculation module of industries, through comparison of WECIADP and technology IADP, the suitable scheme is determined according to the rule of "environment first, then feasibility".
In addition, to determine the industry IADP based on WEC, six indicators are taken into consideration for identification of industry weight, including influence coefficient and response coefficient through information entropy. Consequently, the results of industry weight could imply the importance of the regional dominant industry, and allow the dominant industry to own more IADP. More IADP would allow the dominant industry to expand their economic activities and enhance the industrial ratio in the whole local industry. Therefore, the results of IADP would become a possible tool for stimulating industrial structure upgrading.
To realize the wide application of the three-tier analytical framework in other cities, the basic industrial statistical data and the water environment data should be collected.
The proper water quality model should be selected to calculate the local water environment capacity according to the local river basin characteristic. Furthermore, in order to achieve high accuracy and real time water management, high spatial-temporal resolution data should be acquired for monthly and even daily management.
In China, there is still a lack of the direct data of best available technology; therefore, it is replaced by the average discharge intensity under the same industrial type to calculate TIADP value. Inevitably, it will cause uncertainty of the IADP calculation. Due to the diversity of industrial agglomeration, an industry in one region may have a more clean production level and a lower average discharge intensity, and therefore be assigned a low value of TIADP, while in another region, the same industry could be assigned a high value of TIADP due to its rough high-emission production. The lack of a unified national standard would cause unfairness during the product trade in China. Although our method can achieve the goal of water environment protection and technical feasibility, the principal of fairness should be further considered by introducing the national average discharge intensity or accelerating the standard establishment of best available technology in future research.

Conclusions
A three-tier analytical framework including control units, industries, and enterprises is established, and the trial application for initial allocation of emission permits is succeed in the study area. The top-down and bottom-up approaches are processed simultaneously. In the process of the top-down approach, the WEC and PID are calculated based on onedimensional water quality and historical discharge status, then the IADP of each industry, based on WEC, is achieved. In the procedure of the bottom-up approach, the IADP of each enterprise is obtained according to the industrial average discharge intensity and enterprises output value, then the IADP of each industry, based on technology, is obtained. Finally, through comparison of the results of WECIADP and TIADP, the appropriate enterprises IADP amount is determined. Some advancements have been made in this analytical framework.
(1) Different from the existing research of regional IADP, which only took WEC as an index, in this analytical framework, the WEC is set as a limiting amount, then the PID value is calculated based on WEC. Furthermore, rather than the common research scale of the administrative region, this study is carried out with a relatively small scale of control units, thus it makes it possible to calculate the IADP of industries and enterprises. Meanwhile, the IADP of the same kinds of industries and enterprises are different due to the WEC difference of control units. It may be able to promote industry layout optimization, while enterprises can seek more IADP by moving enterprise location from control units with poor WEC to control units with rich WEC. Furthermore, ensuring the pollutant discharge does not exceed the water environment capacity is the most important prerequisite of IADP determination. (2) Different industry types have significant differences in many aspects, such as clean production level, discharge intensity, importance of industry and so forth; meanwhile, enterprises under the same industry type have relatively uniform discharge intensity. Therefore, in the preliminary determination of enterprises IADP, rather than considering too many indicators, the average discharge intensity and enterprises output value are adopted only for enterprises IADP calculation according to the industrial types in this study. Thus, the results would be more technologically feasible. (3) The industry IADP is introduced as the middle tier and make the results of industry WECIADP and industry technology IADP able to be compared. The determination of industry IADP consider both their WEC limitation and industry characteristics, while identification of enterprises IADP under the same industry types make it more technically feasible. WECIADP results from calculation by the information entropy method conforms more with the principles of equality and efficiency. The IADP of industries and enterprises would be more practicable, while TIADP results acquired by summation of enterprises IADP is more technologically feasible. Furthermore, the final adopted IADP scheme after comparison of results could not only achieve the overall goals of environmental protection and economic growth, but also makes the results more acceptable for enterprise stakeholders. Once the reduction scheme has been put forward to guarantee the WEC is not exceeded, in this analytical framework, we can focus on the specific excessive industries and achieve the detailed excessive amount. Afterwards, the quantitative solution can be applied for each enterprise according to industry types separately.