Abstract In the context of extreme event ecology, identification of pointer years has become a central aspect of tree-ring research. However, the variety of methods employed for pointer year detection since the introduction of the concept in 1979 impedes a direct comparison among studies. Moreover, most commonly used methods partly rely on arbitrarily selected thresholds, resulting in a potentially inconsistent application of those means. To overcome these discrepancies, we here introduce the “standardized growth change” method SGC, which relies on probability density functions of standardized year-to-year ring width differences and internationally accepted significance levels. To evaluate the performance of SGC, it is applied to 1000 pseudo-populations with known properties as well as to an existing Scots pine tree ring data set and compare the results derived from SGC to the four most frequently applied pointer year detection methods. Our comparative evaluation indicates SGC to supersede the other considered methods. In particular, it identified all artificially introduced pointer years in the pseudo-populations, whereas the other methods missed between 3 and 96 percent of known events. A detailed evaluation of misclassifications by the other approaches points out method-specific weaknesses. Finally, we provide technical aspects and recommendations for the application of SGC in a broader context.

中文翻译：

---

一种识别指针年的新方法

摘要 在极端事件生态学的背景下，指针年份的识别已成为树木年轮研究的一个核心方面。然而，自 1979 年引入这一概念以来，指针年份检测所采用的各种方法阻碍了研究之间的直接比较。此外，最常用的方法部分依赖于任意选择的阈值，导致这些方法的应用可能不一致。为了克服这些差异，我们在此引入了“标准化增长变化”方法 SGC，该方法依赖于标准化的年度环宽差异和国际公认的显着性水平的概率密度函数。为了评估 SGC 的性能，它应用于 1000 个具有已知特性的伪种群以及现有的苏格兰松树年轮数据集，并将 SGC 得出的结果与四种最常用的指针年份检测方法进行比较。我们的比较评估表明 SGC 取代了其他考虑过的方法。特别是，它识别了伪种群中所有人为引入的指针年份，而其他方法错过了 3% 到 96% 的已知事件。其他方法对错误分类的详细评估指出了特定于方法的弱点。最后，我们为在更广泛的背景下应用 SGC 提供技术方面和建议。我们的比较评估表明 SGC 取代了其他考虑过的方法。特别是，它识别了伪种群中所有人为引入的指针年份，而其他方法错过了 3% 到 96% 的已知事件。其他方法对错误分类的详细评估指出了特定于方法的弱点。最后，我们为在更广泛的背景下应用 SGC 提供技术方面和建议。我们的比较评估表明 SGC 取代了其他考虑过的方法。特别是，它识别了伪种群中所有人为引入的指针年份，而其他方法错过了 3% 到 96% 的已知事件。其他方法对错误分类的详细评估指出了特定于方法的弱点。最后，我们为在更广泛的背景下应用 SGC 提供技术方面和建议。