多环芳烃（PAHs）是指两个或两个以上的苯环以稠环或非稠环形式连接在一起形成的一类有机化合物。由于PAHs具有“致癌、致畸、致突变性”受到国内外学者广泛关注，美国环保署（US EPA）于1982年公布了16种优先控制PAHs污染物。随着经济的发展和人口的迅速增长，以及能源利用效率低等多重因素，我国PAHs排放量呈逐年递增趋势，对生态系统尤其是水生生物产生了越来越严重的威胁。为了控制污染的发展，保护和改善环境质量，世界各国逐步推出水生生物基准（ALC）来保护本国的环境，以期达到保护本国水生生物的目的。目前，我国正在陆续开展水生生物基准研究工作。但是，我国水生生物基准研究起步较晚，并且存在本土物种生态毒理学数据缺乏的问题，这将在较大程度上影响我国水生生物基准研究的进程，也将对生态风险评估等环境管理工作产生一定的影响。本文以太湖及辽河流域为例，探究我国重点流域中PAHs分布组成及迁移特征，运用初步生态风险评估方法确定我国典型PAHs水生生物基准阈值目标污染物名单。为了解决本土物种生物毒性数据匮乏的问题，进行“三门八科”九种本土水生生物生态毒理学试验，运用K-S检验和数量级差异分析方法探究非本土水生生物毒性数据进行我国本土水生生物基准研究的可行性，并针对种间关系估算模型（ICE）的三种评价方法进行比对分析，以期获得最适于我国本土水生生物基准阈值研究的ICE模型预测数据评价方法，同时为我国系统地开展水生生物基准研究工作提供科学方法和技术支持。本论文的主要研究结论如下：（1）本文通过对太湖流域2010-2013年表层沉积物中PAHs暴露浓度分析结果显示，太湖流域污染区域逐渐由北部梅梁湖区缓慢向湖心及东部湖区转移。4年中PAHs由4环占主导地位（49.8%-67.5%）逐渐转变成由5-6环等高环占主导地位（38.6%-61.4%）。从PAHs单体组成上分析，太湖流域表层沉积物中PHE、FLU所占比例较高（15.1%-24.7%、13.8%-19.5%），其次为FL和PYR。另外，初步生态风险评估表明，22个采样点位中PHE、FLU、BaP均超过了生态风险评价低值（ERL），对当地水生态系统构成潜在生态威胁。（2）本文通过对辽河流域2010-2013年表层沉积物中PAHs暴露浓度分析结果显示，辽河流域PAHs含量逐年下降，辽河水系水质优于浑太水系。4年PAHs检出最高浓度均在浑太水系沈阳段（22#、23#）。4年中，低环PAHs中PHE含量最高，占2-3环18.9%-31.7；中环PAHs中FLU所占比例最大，达到15.4%-28.6%；高环PAHs中IP、DA及BghiP仅在个别点位被检出，主要组成为BaP、BaA、BkF及BbF，其中BaP含量最高，占高环含量9.6%-24.5%。PHE、FLU及BaP在多个点位的SQG-Q值超过0.1，达到中度负面效应，对水生生态系统构成潜在威胁。（3）本文分别对典型PAHs污染物（PHE、FLU、BaP）进行“三门八科”9种本土水生生物（锦鲤和麦穗鱼，泥鳅，泽蛙蝌蚪，大型溞，青虾，摇蚊幼虫，水丝蚓，水螅属）的急、慢性毒理学试验；同时对我国本土水生生物急性基准阈值（CMC）和慢性基准阈值（CCC）进行了推导，PHE的CMC为0.033 mg/L，CCC为0.012 mg/L；FLU的CMC为0.570 mg/L，CCC为0.174 mg/L；BaP的CMC为0.426 μg/L，CCC为0.134 μg/L。（4）本文针对数据相对丰富的PHE、FLU两种污染物，通过本土与非本土生物毒性数据拟合SSD曲线进行物种敏感性差异比较，运用K-S test检验方法探究非本土水生生物毒性数据在我国本土基准阈值推导过程中的可行性。结果显示：本土与非本土物种敏感性分布存在一定的差异，这表明直接使用非本土水生生物毒性数据来推导我国本土水生生物基准阈值是存在不确定性和风险的。（5）本文通过本土实测水生生物毒性数据与ICE模型三种评价方法所得预测数据进行比对分析，结果显示：PHE、FLU能被评价方法①、②很好地预测，BaP仅能被方法②预测，另外方法②相比方法①预测数据更多，因此在本土ICE发展的初级阶段，优先选用模型参数要求较为宽松的评价方法②，从而获取较为丰富的、包含多个物种分类的生物毒性数据，使水生生物基准阈值研究更具代表性和全面性。

Polycyclic aromatic hydrocarbons (PAHs) are a group of organic compounds which are composed of multiple aromatic rings that bound though both fused ring or non-fused ring. Because of their well known carcinogenesis, teratogenesisand mutagenesis, sixteen PAHs were listed as priority polltuants by US EPA in 1982. In China, the input of PAHs into environment have been increasing as the increase of population and development of economy, which has become a threaten to aquatic lives and ecosystem. To protect aquatic ecosystem and improve environmental quality, aquatic life criteira (ALC) have been develped by many developed countries.In China, the researches on ALC in China lag behind those developed countries. Though related researches have been carried out in China in recent years, it was limited the establishment of ALC in China because of lack of toxicological data for native species and enviromental management (i.e., environmental risk assessment (ERA)). This paper took the Liaohe River (representing the watershed in traditional heavy industrial area) and the Taihu Lake (representing the watershed in fast devleoping area) as examples to study the distribution and tranporting of PAHs and to list the priority PAHs pollutants through ERA method. To complement the toxicological data for native species in China, toxicity tests using nine native species from "Three phyla and Eight familes" were conducted. In order to find the most suitable evaluation standard of Interspecies Correlation Estimation (ICE) models for the native aquatic organisms ALC, the measured data were compared with the predicted data from three evaluation standard of ICE. Overall, this thesis will provide scientific and technological support for related research of in China. The key results are listed as follows:(1) The results of distribution of PAHs in surface sedimetns in the Taihu Lake from 2010 to 2013 clearly showed that the pollution area had been shifting from the Meilianghu Lake to the center and the east area. During the four years' monitoring, the dominant PAHs have been shifting from four rings PAHs (49.8%-67.5%) to 5-6 rings PAHs (38.6-61.4%). Analysis on the composition of PAHs showed that the surface sediment of the Taihu Lake were dominated by PHE and FLU (15.1%-24.7%、13.8%-19.5%), and followed by FL and PYR. Through ERA, concentrations of PHE and FLU in 22 sampling sites exceeded their Effects Range Low (ERL) values, which indicated potential environmental risks of these two chemicals.(2) The monitoring results of PAHs in surface sediments of the Liaohe River showed a decreasing trend from 2010 to 2013. The water quality of the Liaohe tributary were better than Huntai tributary. The highest level of PAHs in all the four yearss were detected in Huntai tributary (Shenyang section; 22# and 23#). Among PAHs with low aromatic rings, PHE were the dominant (18.8-31.7%); among PAHs with medium aromatic rings, FLU were the most (15.4-28.6%); and among PAHs with high aromatic rings, BaP, BaA, BkF and BbF were dominant, in which BaP were detected at highest concentration (accouting for 9.6-24.5% of high rings PAHs), while IP, DA and BghiP were only detected in several sampling sites. The SQG-Q values of PHE, FLU and BaP in many sampling sites were above 0.1, indicating potential risks of these three chemicals.(3) Based on above results, PHE, FLU and BaP were taken as typical PAHs compounts, and both acute and chronic toxicity tests for these three chemicals were conducted using nine different native species from "Three phyla and Eight familes"(Cyprinus carpio, Pseudorasbora parva, Misgurnus anguillicaudatus, Rana limnocharis, Daphnia magna,Macrobrachium nipponense, Chironomusplumosus, Limnodrilus hoffmeisteri and Hydra). In addition, CMC, CCC for these chemicals were derived based on the toxicity results. For PHE, CMC was 0.033 mg/L and CCC was 0.012 mg/L; for FLU, CMC was 0.570 mg/L and CCC was 0.174 mg/L; and the CMC and CCC for BaP were 0.426 and 0.134 μg/L, respectively.(4) As toxicity data for PHE and FLU are relatively abundant, SSD curves were fitted using both toxicity data of native species and data of non-native species. By means of K-S test and variance analysis for order of magnitude, HC5 of SSD curves were compared. Results showed that HC5 calculated from SSD using native species was different from that from SSD using nonnative species, indicating the uncertainty and unreliability of directly applying toxicity data from non-native species in dering ALC in China.(5) Through comparison between the measured toxicity data from lab-based experiment and the predicted data from ICE models (using three different evaluation method), the ICE models (applyigng evaluation method ① and ②) could well predict ALC of PHE and FLU because the predicted ALC values were within the same order of magnitude, while BaP could only be predicted by method ②. At current early stage, thus, this method should be the prior choice when applying ICE models to aquire toxicity data from more taxonomies, which in turn could increase the representativity and comprehensiveness of ALC.