淀山湖鱼类调查样本量优化设计

Optimization of sample size for lake fish resources survey

渔业资源的独立调查是获得高质量数据的重要手段,可以为渔业资源评估与管理提供科学依据,而样本量(站位数量)直接影响调查数据的质量。以淀山湖鱼类资源调查为例,利用重采样方法,研究样本量对调查结果精确度和准确性的影响。结果表明:在减少样本量的情况下同样可获得较高精确度和准确性的结果。不同指数的相对估计误差(REE)随着样本量的增加而降低,当样本量大于等于8时大部分指数的REE均趋于稳定;REE与指数的空间变异性呈正相关关系,REE存在指数间差异和季节差异。为保持相同精确度,不同指数在同一季节所需样本量不同,分析鱼类多样性指数所需样本量最少,其次是鱼类群体资源量指数,分析单个物种资源量指数所需样本量最多;同一指数不同季节所需样本量也不相同,分析资源量丰度指数冬季所需样本量最多,分析鱼类多样性指数夏季所需样本量最多。适当调整样本量可以在保证调查结果精确度和准确性的情况下降低调查对资源量较小的鱼类的负面影响。

Fishery-independent surveys are essential for collecting high quality data to support fisheries stock assessment and management. Sample size（ number of sites） plays a significant role in determining the quality of survey data. A fishery survey in Dianshan Lake was conducted as a case study to analyze the effect of sample size on accuracy and precision for estimates of fish abundance or fish species diversity using resampling techniques. Results showed sample size can be reduced while still achieving relatively high precision and accuracy for most indices measuring abundance and biodiversity. The relative estimated error（ REE） for all the indices increased when the sample size decreased in four sampling seasons. For most indices in the four sampling seasons,the REE was stable or slightly increased when the sample size decreased from 24 to 8. The REE had a significant and positive relationship with the CVs of the corresponding indices for the four sampling seasons and had different levels of values for each index in the four seasons. For the sake of the same precision for different indices in one season,the sample size needed for analyzing fish diversity indices was smaller than the one required while analyzing species abundances indices. To maintain the same precision for one index in different seasons,the sample size needed for analyzing abundance indices and fish diversity indices were the largest in winter and summer respectively. The study suggests that the optimal sample size identified in a lake fishery survey design may be different with respect to different goals. Sample size could be reduced to some extent in the subsequent surveys to reduce the negative effect of the survey on fish species with low abundance in the study area.