Genetic signatures of ecological diversity along an urbanization gradient

Despite decades of work in environmental science and ecology, estimating human influences on ecosystems remains challenging. This is partly due to complex chains of causation among ecosystem elements, exacerbated by the difficulty of collecting biological data at sufficient spatial, temporal, and taxonomic scales. Here, we demonstrate the utility of environmental DNA (eDNA) for quantifying associations between human land use and changes in an adjacent ecosystem. We analyze metazoan eDNA sequences from water sampled in nearshore marine eelgrass communities and assess the relationship between these ecological communities and the degree of urbanization in the surrounding watershed. Counter to conventional wisdom, we find strongly increasing richness and decreasing beta diversity with greater urbanization, and similar trends in the diversity of life histories with urbanization. We also find evidence that urbanization influences nearshore communities at local (hundreds of meters) rather than regional (tens of km) scales. Given that different survey methods sample different components of an ecosystem, we then discuss the advantages of eDNA—which we use here to detect hundreds of taxa simultaneously—as a complement to traditional ecological sampling, particularly in the context of broad ecological assessments where exhaustive manual sampling is impractical. Genetic data are a powerful means of uncovering human-ecosystem interactions that might otherwise remain hidden; nevertheless, no sampling method reveals the whole of a biological community.

尽管环境科学与生态学领域已有数十年研究积淀，但量化人类活动对生态系统的影响仍极具挑战。这一方面源于生态系统各组分间存在复杂的因果关联链，另一方面则因难以在足够的空间、时间及分类学尺度上采集生物数据，令该难题进一步加剧。本研究证实了环境DNA（environmental DNA, eDNA）在量化人类土地利用与邻近生态系统变化间关联的应用价值。我们对近岸海草床群落水体中采集的后生动物eDNA序列展开分析，评估了该类生态群落与周边流域城市化程度之间的关联。与传统认知相悖的是，我们发现随着城市化水平提升，物种丰富度显著升高，而β多样性则呈下降趋势；生活史多样性亦呈现出随城市化水平提升的类似趋势。此外，我们还发现城市化对近岸群落的影响主要发生在局地尺度（数百米级），而非区域尺度（数十公里级）。鉴于不同调查方法仅能采集生态系统的不同组分，我们进而探讨了eDNA作为传统生态采样补充手段的优势——本研究中我们借助eDNA可同时检测数百个分类群——尤其适用于无法开展全面人工采样的大型生态评估场景。遗传数据是揭示人类活动与生态系统间潜在关联的有力工具，此类关联若不借助遗传数据或被隐藏；但即便如此，尚无任何一种采样方法能够完整展现生物群落的全貌。