Challenges and prospects for interpreting long-term phytoplankton diversity changes in Lake Zurich (Switzerland)
Challenges and prospects for interpreting long-term phytoplankton diversity changes in Lake Zurich (Switzerland)
复制标题
解释苏黎世湖（瑞士）浮游植物多样性长期变化的挑战和前景
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
R. Ptáčník
中科院分区：
文献类型：
--
作者：
F. Pomati;C. Tellenbach;B. Matthews;P. Venail;B. Ibelings;R. Ptáčník
摘要1.分析和解释浮游植物的长期时间序列带来了一些挑战，这些挑战来自数据收集和生物分类鉴定方面的潜在历史不一致。在之前的一篇论文中，Pomati等人（2012年）发现浮游植物多样性显著增加，这与苏黎世湖32年来的贫营养化和变暖相吻合。这些发现最近受到了挑战，因为在时间序列中检测限和分类学分类的潜在偏倚（Straile，Jochimsen & K€ 2013）。我们同意，谨慎的长期浮游植物数据系列是非常重要的，但认为，在苏黎世湖检测到的丰富度的增加不能仅仅是由于方法的偏差。2.在对苏黎世湖浮游植物数据集进行进一步分析后，我们发现Straile等人（2013年）报告的分类单元检测限的变化没有得到数据的支持，并且源于这些作者可用数据集中密度计算的舍入误差。我们发现，下降的比例丰度为常见的类群，增加了类群的年度流行率和减少社区营业额的时间序列。总的来说，这些数据清楚地表明了优势度下降的趋势，而更多的分类群同时共存。我们还认为，分类学分类一直是强大的（至少在家庭层面上），并提出了一个诊断图，可以帮助检测浮游生物丰富度随时间的变化的无偏信号。3. Straile等人（2013年）观察到苏黎世湖和附近的瓦伦湖之间物种出现的完美同步。虽然我们同意，这种完美的同步在社区组成可以反映数据库编译的偏见，它也可以是一个重要的生态信号的变化，区域物种池，值得进一步分析。4.我们得出结论，Pomati et al.（2012）的结果是稳健的，没有受到Straile et al.（2013）的批评的实质性破坏。更一般地说，确实有可能从长期浮游植物监测数据集中提取生物多样性变化的有意义的信号，只要清楚地了解在时间序列的历史上如何对数据进行采样、记录和分析。
SUMMARY 1. Analysing and interpreting long-term phytoplankton time series present a number of challenges, arising from potential historical inconsistencies in data collection and taxonomic identification of organisms. In a previous paper, Pomati et al. (2012) found a remarkable increase in phytoplankton diversity that coincided with oligotrophication and warming of Lake Zurich over a 32-year period. These findings were recently challenged on the basis of potential biases in detection limits and taxonomic classification over the time series (Straile, Jochimsen & K€ 2013). We agree that being cautious with long-term phytoplankton data series is extremely important, but argue that the increase in richness detected in Lake Zurich cannot be due only to methodological bias. 2. Following additional analysis of the Lake Zurich phytoplankton dataset, we found that the shift in taxon detection limits reported by Straile et al. (2013) is not supported by the data and stems from a rounding error in the calculation of density in the dataset available to those authors. We found a decline in the proportional abundance for common taxa, an increase in the annual prevalence of taxa and reduced community turnover over the time series. Taken together, the data clearly indicate a trend of decreasing dominance, while more taxa coexist simultaneously. We also argue that the taxonomic classification has been robust (at least at the family level) and propose a diagnostic plot that can help detect an unbiased signal of change in plankton richness through time. 3. Straile et al. (2013) observed perfect synchrony in species occurrence between Lake Zurich and nearby Lake Walen. While we agree that such perfect synchrony in community composition can reflect a bias in the database compilation, it can also be an important ecological signal of changes in regional species pools that deserves further analysis. 4. We conclude that the results of Pomati et al. (2012) are robust and not substantially undermined by the criticisms of Straile et al. (2013). More generally, it is indeed possible to extract meaningful signals of biodiversity change from long-term phytoplankton monitoring datasets, provided there is a clear understanding of how the data have been sampled, recorded and analysed over the history of the time series.