基于正态分布假设检验的湖泊营养盐浓度分级评价方法研究

目前湖泊营养盐浓度级别评价常采用的平均值法忽视了变量的不确定性.本文据此提出一种基于正态分布总体均值假设检验的级别评价方法,并结合我国地表水环境质量标准(GB3838—2002)对营养盐浓度的分类,提出了简便快捷的图示法.结果发现,该方法能体现营养盐浓度变量的不确定性,融合先验级别假设,将决策风险定量化,可提供较为广阔的决策空间.同时,采用该方法对滇池外海观音山监测断面氨氮浓度的水质级别进行评价,发现先验级别假设、样本容量和决策偏好对评价结果有显著影响;与平均值法的比较发现,假设检验法具有更小的错误概率(例如,样本容量13时,风险0.21),评价结果具有更高的可信度,验证了假设检验法的适用性.

A hypothesis test method with normal distribution assumption for grade assessment of lake nutrient concentrations

The uncertainty of nutrient concentration variable is always ignored while assessing the water quality grade. A grade assessment method based on the hypothesis test for the mean of normal distribution was thereby proposed in this study. We introduced the principles of the hypothesis test. Since the nutrient concentration was divided into six grades in China according to the environmental quality standards for surface water (GB3838-2002), a simple graphic method based on the hypothesis test to assess the concentration grade was developed. It can reflect the decision-making risk quantitatively, and improves broader decision space for decision-maker, with emerging prior hypothesis on the grade of nutrient concentration. Using the hypothesis test method, we assessed the water quality grade of ammonia nitrogen of Guanyinshan monitoring station in Lake Dianchi. The results demonstrated significant impacts of prior hypothesis, sample size and the decision preference of decision-maker on grade assessment results. Comparing with the mean value method, we found that the result of hypothesis method had higher credibility with lower falsely inferring a breach of standard risk and low falsely inferring compliance risk. In general, the hypothesis test method proposed in this article was applicable for nutrient concentration grade assessment.