The CUSUM out-of-control table to monitor changes in fish stock status using many indicators
The CUSUM out-of-control table to monitor changes in fish stock status using many indicators
复制标题
使用多种指标监测鱼类种群状况变化的 CUSUM 失控表
DOI：
10.1051/alr/2009021
复制
发表时间：
2009
影响因子：
1.1
通讯作者：
P. Petitgas
中科院分区：
文献类型：
--
作者：
P. Petitgas
使用一套指标评估鱼类种群的一种方法是红绿灯法。在这种方法中，不同指标的时间序列被映射到一个共同的色标上，以突出显示指标交叉引用限值时发生的警报。然而，到目前为止，这一程序缺乏一个统计框架。在这里，我们提出了一个合适的统计框架的累积总和（Cumulative sum，CIMUM）监测计划。CRAMUM是一种统计过程控制方法，根据定义的性能标准检测与参考平均值的偏差。利用CNOUM监测方案，当指示器越过对应于错误警报和非警报的定义概率的定义的受控限制时，可以触发警报信号（即，精确度和功率）。构建了一个CRACUM失控偏差表，用作诊断表。在该表中，不同指标的偏差是量化的，并以类似的差异单位列出，这有助于对其进行综合评估。CNOUM失控表还显示偏差如何随时间累积，从而提供库存历史视图。该程序适用于北海鳕鱼种群，以说明如何使用一套来自研究调查数据的指标，可以实现渔业独立的综合评估。所使用的指标与空间分布、丰度、长度结构、成熟长度和表观死亡率有关。自2001年以来，该种群被发现超出其参考限度，并且自那时以来状况持续恶化。
One method to assess fish stocks using a suite of indicators is the traffic light approach. In this approach, the time series of the different indicators are mapped on a common colour scale to highlight alerts that occur when indicators cross reference limit values. Until now, however, the procedure has lacked a statistical framework. Here, we propose the cumulative sum (CUSUM) monitoring scheme as a suitable statistical framework. CUSUM is a statistical process control method that detects deviations from a reference mean, according to defined performance criteria. With the CUSUM monitoring scheme, alarm signals can be triggered when indicators cross defined in-control limits that correspond to defined probabilities of false alarm and non-alarm (i.e., precision and power of the CUSUM monitoring scheme). A table of CUSUM out-of-control deviations is constructed to serve as a diagnostics table. In this table, the deviations in the different indicators are quantitative and given in similar units of variance, which facilitates their integrated assessment. The CUSUM out-of-control table also shows how deviations accumulate over time and thus provides a view of the stock history. The procedure was applied to the North Sea cod stock to illustrate how a fishery-independent integrated assessment can be achieved using a suite of indicators derived from research survey data. The indicators used were related to the spatial distribution, abundance, length structure, length at maturity and apparent mortality. The stock was found to be outside its reference limits from 2001 and has shown a continued degradation in status since this time.