水生生物对三唑磷的物种敏感度分布研究

针对水环境中日益严重的有机磷农药污染问题，选择广泛使用的三唑磷作为研究对象，利用其对水生生态系统中不同营养层次生物物种的半数效应浓度(median effective concentration, EC₅₀)，建立了基于对数-逻辑斯蒂分布的水生生物物种敏感度分布(species sensitivity distribution, SSD)模型，并采用概率图和吻合度检验方法对该模型进行了检验和评价.结果表明,水生生物对三唑磷的 SSD 服从对数-逻辑斯蒂分布，其参数为 α=-0.4788±0.2381，β=0.7546±0.1078.基于该 SSD 模型，获得三唑磷的 5% 危害浓度(hazardous concentration for 5% of the species, HC₅)值为 1.992×10^-3 mg/L，并推导出三唑磷的最大浓度基准值(criteria maximum concentration, CMC)值为 9.96×10^-4 mg/L.对 HC₅、CMC 与单一物种的安全浓度的比较研究指出，基于 SSD 方法制定环境质量标准更为严格，也更接近于真实的生态环境.另外，根据渤海莱州湾海域中三唑磷的监测数据，预测了其对物种的潜在影响比例(potentially affected fraction, PAF)为 0.36%，对该水域生态环境的影响处于较低风险水平.

Species Sensitive Distribution for Aquatic Biota Exposed to Triazophos

The increasing pollution of organophosphorus pesticides (OP) in water have been of concerns. Taking the widely used triazophos as the object, a species sensitivity distribution (SSD) model was developed using a log-logistic distribution based on the median effective concentrations (EC50) of triazophos to aquatic species at various trophic levels, and then the model was tested and evaluated using probability plots and good-of-fit tests. The results showed that the SSD for aquatic biota exposed to triazophos was well fitted by a log-logistic distribution, which was totally determined by the two parameters, alpha = -0.4788 and beta = 0.7546, with standard error 0.2381 and 0.1078 respectively. Based on the SSD model, the hazardous concentration for 5% of the species (HC5) and the criteria maximum concentration (CMC) of triazophos were 1.992 x 10(-3) mg/L and 9.96 x 10(-4) mg/L, respectively. Through comparing the HC5 and CMC with the safe concentration for single-species, it could be found that environmental quality criteria derived from the SSD model was more strict, and closed to the real ecological environment. In addition, according to the reported data, the potentially affected fraction (PAF) of species exposed to triazophos in the Laizhou Bay (Bohai Sea, China) area was 0.36% predicted by the SSD model, and the corresponding risk might not be significant.