A standard sample method for controlling microfossil data precision: A proposal for higher data quality and greater opportunities for collaboration
A standard sample method for controlling microfossil data precision: A proposal for higher data quality and greater opportunities for collaboration
复制标题
控制微化石数据精度的标准样本方法：提高数据质量和更多合作机会的建议
DOI：
10.1016/j.quaint.2012.10.056
复制
发表时间：
2013
影响因子：
2.2
通讯作者：
Nakagawa T
中科院分区：
文献类型：
--
作者：
Nakagawa T
使用标准样品是控制数据质量的一种简单而有效的方法。在第四纪科学领域，它是同位素和地球化学分析的常用方法，大大提高了这些分析方法的总体可信度。相比之下，微体化石分析通常没有任何这样的系统来确保客观和常规的数据精度和准确性。在本文中，我们提出了一个相对简单和廉价的方法，采用这样的标准样品的方法进行微体化石分析。收集并均质化来自岩心二次取样的拒收材料，并用于标准样品。对每七个沉积物样品中的一个标准样品进行处理和分析。标准数据与长期平均值的偏离定量地表明了误差。如果误差与时间相关（即误差在分析所花费的时间内具有长期或短期趋势），则从标准数据收集的附加信息可用于校正原始数据以提高数据精度。微体化石分析中的标准抽样方法不仅在监测和提高数据质量方面非常有效，而且还增加了合作机会，因为它大大减少了参与分析的所有各方之间的系统性数据偏移。
Using standard samples is a simple yet very effective method to control data quality. In the field of Quaternary science, it is a commonly applied approach for isotope and geochemical analyses, greatly increasing the general level of confidence in these analytical methods. Microfossil analysis, by contrast, does not normally have any such system to secure data precision and accuracy objectively and routinely. In this paper, we propose a relatively easy and inexpensive way to adopt such a standard sample method for microfossil analyses. Rejected material from core subsampling was collected and homogenised, and was used for standard samples. One standard sample was treated and analysed for every seven sediment samples. Departure of the standard data from the long-term average quantitatively indicates the error. If the error was correlated with time (i.e. the error has a long or short term trend within the time spent for analysis), the additional information gathered from the standard data can then be used to correct raw data to improve data precision. The standard sample method in microfossil analysis is not only very effective in monitoring and improving data quality, but also enhances opportunities for collaboration, as it significantly reduces systematic data offsets between all parties involved in the analysis.