Precision of the daily pollen count. Identifying sources of variation using variance component models.
Precision of the daily pollen count. Identifying sources of variation using variance component models.
复制标题
DOI：
10.1007/bf02311366
复制
发表时间：
1993-01-01
期刊：
影响因子：
2
通讯作者：
Moseholm, L.
中科院分区：
文献类型：
--
作者：
Pedersen, B. V.;Moseholm, L.
三名训练有素的花粉计数器独立检查了位于哥本哈根市中心北郊气象研究所屋顶上距地面 15 m 的 2 个 Burkard 花粉捕捉器中的相同样本。选择在草花粉季节期间从每个诱捕器采集20天的样本。确定了足够数量用于统计分析的四个物种组——云杉/松属、禾本科、荨麻属和蒿属。使用方差分量模型识别变异来源并评估其相对重要性。每组都得出了同等的定性结论：(i) 确定了预期的季节性变化； (ii) 计数器对花粉计数的变化有显着影响； (iii) 陷阱对方差没有贡献，但确定了 d-陷阱相互作用 - 这种相互作用被解释为采样期间仪器变化的问题或气象条件的变化； (iv) 总方差大于最初预期，相对不确定性 >50%。通过减少计数器变化或 d-trap 相互作用，每日花粉捕获计数的方差并未减少太多。花粉计数器之间的差异很小，并且 d-trap 相互作用很难完全理解。通过检查每个样本的更大面积，减少了基本测量误差；结论是，虽然这意味着更高的成本，但它可能是减少花粉计数不确定性的最有效方法。
Three trained pollen counters independently examined the same specimens from 2 Burkard pollen traps which were located 15 m above ground level on the roof of the Meteorological Institute on the northern outskirts of central Copenhagen. Specimens from each trap taken for 20 d during the grass pollen season were selected. Four species groups were identified in sufficient numbers for statistical analysis - Picea/Pinus, Gramineae, Urtica and Artemisia. Identification of sources of variation and assessment of their relative importance were carried out using variance component models. Qualitatively equivalent conclusions were reached for each group: (i) the expected seasonal variation was identified; (ii) the counters contributed significantly to the variation of pollen counts; (iii) the traps did not contribute to the variance, but a d-trap interaction was identified - this interaction was interpreted as a problem of instrumental variation over the time of sampling or as variations in meteorological conditions; and (iv) the total variance was larger than originally expected, with a relative uncertainty >50%. The variance of the daily pollen trap count was not reduced much by reducing the counter variation or d-trap interaction. The variation among pollen counters was small, and the d-trap interaction was difficult to fully comprehend. The basic measurement error was reduced by examining a larger area of each specimen; it is concluded that, although this implies a greater cost, it is probably the most effective method of reducing the uncertainty in pollen counts.