ABSTRACT: Nitrogen stable isotope ratios (δ¹⁵N) are routinely used to estimate consumer trophic positions (TPs). This method is very sensitive to the chosen isotopic baseline, which should incorporate all temporal isotopic variability occurring in the base of the food web sustaining the consumers. However, there is still no consensus on the different methods to represent time intervals for isotopic assimilation. It is uncertain if researchers should compute a single baseline obtained by averaging δ¹⁵N source values across all time periods, multiple baselines based on δ¹⁵N source values collected concomitantly with consumers, or consider a time lag between sources and consumers. Aiming to provide empirical evidence to help select the best option, we designed 4 methods using different temporal lags between consumers and producers based on a 9 yr time series of carbon and nitrogen stable isotopes in an estuary. The 4 tested baseline methods considered the δ¹⁵N average of producers sampled in all years (Global); the same season as consumers (Concomitantly); the season before the collection of consumers (Delayed); and the previous season as well as the same season of consumer sampling (Interpolation). Comparison of results with an expected model (based on stomach content data) showed that the Delayed method is the most appropriate. We conclude that time lags in the incorporation of stable isotope values between sources and consumers must be considered in baseline δ¹⁵N calculations to improve TP estimates and provide more reliable modeling results. Additionally, we provide other recommendations to improve sampling designs when using stable isotopes in TP estimations.

中文翻译：

---

改变基线：整合生态和同位素时滞可以改善水生消费者的营养位置估计

摘要：氮稳定同位素比率（δ ¹⁵ N）常规地用于估计消费者营养位置（TPS）。该方法对所选的同位素基线非常敏感，该基线应包括在维持消费者的食物网基础中发生的所有时间同位素变化。但是，对于代表同位素同化的时间间隔的不同方法仍未达成共识。所以这是不确定的研究人员应该计算通过平均δ获得的单个基线^15点N个源的值在所有的时间段，多个基准基于δ ¹⁵与消费者同时收集的N个来源价值，或考虑来源与消费者之间的时间差。为了提供经验证据以帮助选择最佳选择，我们基于河口中碳和氮稳定同位素的9年时间序列，设计了4种方法，使用了消费者和生产者之间的不同时滞。认为是δ4种测试基线方法¹⁵所有年份中平均无N个生产者抽样（全球）；与消费者同一个季节（同时）；收集消费者之前的季节（延迟）；以及上一季和同一季的消费者抽样（插值）。结果与预期模型（基于胃内容数据）的比较表明，“延迟”方法是最合适的。我们得出结论：在源和消费者之间的稳定同位素值的结合当时的滞后必须在基准线被视为δ ^15次N次计算来提高TP估计，并提供更可靠的模拟结果。此外，当在TP估算中使用稳定同位素时，我们提供了其他建议来改进采样设计。