Similar Resilience Attributes in Lakes with Different Management Practices

Liming has been used extensively in Scandinavia and elsewhere since the 1970s to counteract the negative effects of acidification. Communities in limed lakes usually return to acidified conditions once liming is discontinued, suggesting that liming is unlikely to shift acidified lakes to a state equivalent to pre-acidification conditions that requires no further management intervention. While this suggests a low resilience of limed lakes, attributes that confer resilience have not been assessed, limiting our understanding of the efficiency of costly management programs. In this study, we assessed community metrics (diversity, richness, evenness, biovolume), multivariate community structure and the relative resilience of phytoplankton in limed, acidified and circum-neutral lakes from 1997 to 2009, using multivariate time series modeling. We identified dominant temporal frequencies in the data, allowing us to track community change at distinct temporal scales. We assessed two attributes of relative resilience (cross-scale and within-scale structure) of the phytoplankton communities, based on the fluctuation frequency patterns identified. We also assessed species with stochastic temporal dynamics. Liming increased phytoplankton diversity and richness; however, multivariate community structure differed in limed relative to acidified and circum-neutral lakes. Cross-scale and within-scale attributes of resilience were similar across all lakes studied but the contribution of those species exhibiting stochastic dynamics was higher in the acidified and limed compared to circum-neutral lakes. From a resilience perspective, our results suggest that limed lakes comprise a particular condition of an acidified lake state. This explains why liming does not move acidified lakes out of a “degraded” basin of attraction. In addition, our study demonstrates the potential of time series modeling to assess the efficiency of restoration and management outcomes through quantification of the attributes contributing to resilience in ecosystems.

自20世纪70年代以来，施石灰（liming）已在斯堪的纳维亚及全球其他地区被广泛用于抵消水体酸化（acidification）带来的负面影响。一旦停止施石灰，经施石灰处理的酸化湖泊群落通常会重新回到酸化状态，这表明施石灰无法将酸化湖泊转变为无需后续管理干预的酸化前原始状态。尽管这一现象暗示施石灰湖泊的恢复力（resilience）较低，但影响恢复力的相关属性尚未得到系统评估，这限制了我们对成本高昂的修复管理项目实际效率的认知。
本研究于1997年至2009年间，采用多变量时间序列建模（multivariate time series modeling）方法，对施石灰、酸化及近中性（circum-neutral）湖泊中的浮游植物（phytoplankton）群落指标（多样性、丰富度、均匀度、生物体积）、多变量群落结构以及相对恢复力展开了评估。研究团队识别了数据中的主导时间频率，得以在不同时间尺度上追踪群落动态变化。基于识别出的波动频率模式，我们评估了浮游植物群落相对恢复力的两项核心属性：跨尺度结构与尺度内结构。此外，我们还分析了表现出随机时间动态的物种类群。
研究结果显示：施石灰提升了浮游植物的多样性与丰富度；但经施石灰处理的湖泊，其多变量群落结构与酸化湖泊及近中性湖泊存在显著差异。所有研究湖泊的恢复力跨尺度与尺度内属性均无显著差异，但相较于近中性湖泊，酸化湖泊与施石灰湖泊中表现出随机动态的物种贡献度更高。从恢复力视角来看，本研究结果表明，经施石灰处理的湖泊属于酸化湖泊状态下的一种特定稳态。这一发现解释了为何施石灰无法将酸化湖泊移出‘退化’吸引子盆地（basin of attraction）。此外，本研究证实了多变量时间序列建模可通过量化生态系统中影响恢复力的相关属性，来科学评估生态修复与管理措施的实施成效。