【作者】 张永亮；

【导师】 毕军； 张炳；

【作者基本信息】 南京大学 ， 环境科学， 2012， 博士

【摘要】 水污染物排污交易是当前受到普遍关注的环境经济政策。由于减排成本较高的排污者可以通过交易机制向减排成本较低的排污者购买排污权,排污交易被视为可在成本有效前提下实现既定环境质量目标的有效手段。然而,除了成本因素之外,排污交易政策的设计还要考虑其所要规制的污染物性质。对于COD这类非均匀混合、并且具备迁移转化性质的污染物,其对环境质量的影响不仅取决于污染物排放水平,还取决于排放的空间位置和迁移转化过程。当在流域中开展排污交易时,本应在某一区域排放的污染物会转移到另一区域排放,一个区域的水环境质量得到了改善,却可能会造成另一区域水环境质量的恶化,这决定了水污染物排污交易的相对复杂性。因此,有必要对流域水污染物排污交易的水环境质量影响,以及如何通过政策设计避免负面影响展开研究。江苏省太湖流域是我国较早开展COD排污交易研究与实践的地区之一。本研究基于江苏省太湖流域COD排污交易政策的研究成果与实践经验,从排污者的微观行为角度出发,研究排污交易对流域水环境质量的影响。同时,针对不同政策设计条件下(主要包括命令控制政策、自由交易和交易比率约束条件)交易市场对流域水环境质量影响的差异进行研究,旨在提出能够保证经济有效性与资源配置公平性,同时达到较好环境效果的政策方案,为太湖流域COD排污交易政策的优化和完善提供决策支持。结合国内外相关学者研究的成果、水污染物排污交易的相关案例,以及我国水污染物排污交易的实践情况,本研究筛选并确定了双边交易作为排污交易市场的交易模式。系统分析了排污企业在双边交易市场下的污染物削减、购买及销售排污权等决策行为,构建了基于主体决策的、离散成本条件下的双边交易市场理论模型。选择太湖流域的武进港小流域作为研究区,以武进港流域内的52家主要排污企业作为研究对象,通过实地调研、文献调查等方法收集整理了排污企业的地理信息、排污数据及污染物处理成本函数等基础数据。结合双边交易市场理论模型和研究区数据,综合运用NETLOGO主体建模软件,GIS和计算机编程技术,构建了太湖流域水污染物排污交易仿真系统,对自由交易条件下的排污交易市场进行了仿真。同时,结合WASP水质模型,模拟了自由交易市场对流域水环境质量的影响。交易比率是可以调节排污交易市场水环境质量影响的主要政策要素。结合国内外相关学者的研究经验和我国水环境管理的实际情况,本研究构建了基于分区的武进港流域水污染物排污交易比率系统,并对交易比率约束条件下的交易市场及其水环境质量影响进行了仿真和模拟。最后,对命令控制政策、自由交易和交易比率约束条件下交易市场的仿真模拟结果进行了比较和分析。结果显示,命令控制政策对控制流域工业污染、改善上下游水质起到了显著效果,然而却增加了企业的污染物削减成本和流域污染物处理总成本。自由交易市场较为充分地体现了交易机制的成本效益优势,但由于流域下游的排污企业可以轻易地从上游购买排污权并在下游排放,导致下游COD排放量明显增加,入湖断面出现了水质恶化的情况。根据水体的污染物迁移转化特性对不同区域间的排污权交易设定交易比率,同样可以保证交易机制的成本优势,并且在水质影响的最坏情形下,各河段的水质仍然保持了较好的水平,各断面均未出现水质超标的情况。流域水污染物排污交易的政策设计需要在成本效益、公平性和实现水质目标的不确定性上寻求权衡(trade-off).基于研究结果,本研究指出：交易比率约束下的交易机制是目前太湖流域水污染物排污交易较为适宜的权衡方案。结合当前太湖流域水污染物排污交易政策的设计与实践情况,有必要在下一步的政策优化和完善过程中引入基于分区的交易比率这一政策要素。更多还原

【Abstract】 In theory, a watershed-based emission permit system (EPS) could improve regional water quality in all locations based on total pollution load allocation. Such a "Command and control" approach forces every source to reduce pollution without the consideration of heterogeneities among sources, and causes comparatively higher reduction social cost. A tradable discharge permit (TDP) system has considerable potential for providing an additional avenue to produce environmental benefit, which closely approximates what would be achieved through a "Command and control" approach, with relatively lower costs. However, a TDP system for non-uniformly mixed pollutants such as COD is much more complicated than a uniformly mixed pollutant such as CO2, since the extent and spatial pattern of the damages to the environment depend not only upon the level of emissions, but also upon the locations and the transfer characteristics of the emissions. When a pollutant source pays a source downstream for pollutant reductions through trading, water quality in some locations may become better after trading, and that in certain other locations may become worse after trading. A simply designed TDP system without restrictions for water pollution trading in a watershed may bring uncertainties of water quality improvement, and there is the potential that trades will create spatial "hot spots" immediately downstream of pollutant sources that purchase permits.In2008, the Ministry of Environmental Protection (MEP) and the Ministry of Finance (MOF) started a pilot program of water pollution trading in seven provinces. Tai Lake Basin of Jiangsu province was selected as the first region to initiate the program. The water pollution trading programs must be designed to comprehensively improve the water quality in a specific watershed while pursuing cost efficiency. However, since the water pollution trading is still in its beginning stages in China, the water quality issues after trading have not been addressed in the policy deployments of any pilot programs. The institutional arrangement such as the setting of trading ratios are also not adopted by these programs, related researches are also not found in literatures. By the end of2010, none of trading cases happened in Tai Lake basin, far from the amount of trading observed in active markets. Thus, it is also difficult to assess the water quality impacts of the current policy arrangements of water pollution trading through trading cases. Therefore, empirical studies in specific regions for specific pollutants in China are needed to provide more direct information such as the potential cost savings of a trading system and examine the influence of varying trading rules such as the setting of trading ratios on local environment for decision-makers.This research proposed a zonal based trading-ratio (ZTR) system for COD permits trading to achieve cost efficiency while preventing the occurrence of "hot spots". Such a system allows firms to trade permits freely in accordance with the trading ratios among zones they located. Based on the analysis of agents’decision-making behaviors in a ZTR system, the agent-based model which has been widely applied in simulations of artificial markets and emissions trading markets was employed to construct an artificial trading market to control COD in Wujingang watershed of Tai Lake Basin, China.52major pollutant sources from industrial sector in Wujingang watershed were considered as agents. The results of agent-based model such as the transfers of permits among zones and the changes of discharge distributions in the watershed were integrated into a Water Quality Analysis Simulation Program (WASP) model, to predict the water quality impacts of the trading market to the river system.According to recent practices of water pollution trading in China, this research chose continuous bilateral negotiations as the market structure for the analysis. Trades in such a market are made sequentially, and usually bilaterally, at changing non-equilibrium prices. For each pair of agents with different marginal abatement costs, they negotiate bilaterally to decide the price and desired/offered quantity of permits, and reduce their joint abatement costs. Each agent not only negotiates with one another potential trading partner, but also negotiates with several other agents simultaneously. The behaviors of agents in the trading market which based on their own information such as marginal costs (MC) function, initial permits, and the market principles are analyzed. We also consider each agent has its own pollution reduction capability and associated cost structure, which are a countable and finite number of options on pollution abatement rate and a finite discrete cost function where the abatement rate is the independent variable. All agents are assumed to make decisions for minimizing their own total abatement costs, no agent exercises market power, the perfect monitoring and enforcement are available. Finally, the trading model of the market is consisted by each agent’s following decision-marking strategies.The agent-based model of water pollution trading was implemented in NetLogo, a platform suited for simulating spatial logic driven by the multiagent systems (MAS) and Cellular Automata (CA) approach. The agent-based model starts by reading the state variables of agents and global variable of the artificial world in the platform, and operates based on the agents decision-making processes. The GIS data of the watershed was processed and imported into the GIS extension of Netlogo, agents are numbered and arranged in the artificial market in accordance with their zonal locations. Basic data such as COD productions, COD abatement of agents in2007and initial permits were imported into the platform, and the parameters such as discrete abatement rates and associated costs of agents were estimated based on field investigations and previous data. The water quality impacts of TDP system were simulated by the application of Netlogo and WASP.The water quality impacts of ZTR were also simulated by the Netlogo platform and WASP model. Simulation results of EPS, TDP and ZTR were compared and analyzed. The results shown:EPS system achieves higher abatement efficiency than the initial allocation requires. A trading market without trading ratios also meet the requirement of total load control of the watershed. However, it leads to the transactions of399.82tons of permits from other4zones (mainly from zone1) to zone5. COD has not reduced in zone5, but increased267tons. The ZTR system increases the payment costs of purchasing permits to some agents, and conducts the firms at downstream to adopt the strategy of removing more COD by themselves instead of seeking for transactions with upstream agents. In the case of worst water quality impacts, the ZTR system achieves an abatement efficiency of63.74%, which is significantly higher than the initial permits required and the TDP system, and most amount of COD is abated at downstream comparing to other scenarios. Aiming to meet the requirement of environmental regulations, the policy deployment of TDP is most cost efficient, while the ZTR system is slightly worse than the TDP system, and the EPS is the least cost efficient among the three. The water quality standard will be achieved in all segments of Wujingang river under the scenario of EPS with highest cost. A TDP system without restrictions has lowest cost, though, will lead to the violation of water quality standards at segment5in dry season. The ZTR system creates a tradeoff between cost efficiency and environmental quality.The design of a emissions trading system depends not only on cost-efficiency, but also crucially on the nature of the pollutant being regulated and traded. Since the water quality impacts of the pollution that discharged by polluters are closely related to the positions the polluters located, the water pollution trading system for non-uniformly mixed pollutants such as COD may cause violations of predetermined water quality standards over the length of the river. Based on the research results, this research suggests that the trading ratios are a way for ensuring the equivalency of the potential water quality impact between an allowance generated in one location and used in another location in the watershed. The market restricted by ZTR is more efficient on balancing pollution control costs and water quality achievements than the TDP system. In terms of water quality impacts, trading ratios need to be integrated into the policy deployments of the pilot programs in Tai Lake Basin and China.更多还原