Data from: Asynchronous changes in abundance over large scales are explained by demographic variation rather than environmental stochasticity in an invasive flagellate

Environmental stochasticity is important in explaining the persistence and establishment of invasive species, but the simultaneous effects of environmental and demographic factors are difficult to separate. Understanding how demography and environmental factors affect invasive species abundance over large temporal and spatial scales is essential to anticipate populations at risk of becoming established and setting appropriate management measures. Using a hierarchical mixed modeling approach we analyzed the spatial and interannual dynamics of the invasive raphidophyte Gonyostomum semen, a noxious flagellate which is spreading in northern Europe, in response to demographic and environmental variation. We used data from 76 lakes distributed across two biogeographical regions in Sweden (Central Plains in the south and Fennoscandian region in the north) and sampled during 14 years. We found a strong asynchrony in the density dynamics of G. semen populations between the two regions. G. semen showed positive trends (i.e. increasing frequency of high density peaks) in most southern lakes, forming established populations with recurrent blooms in successive years in some of them. In contrast, G. semen populations were smaller and more stochastic in the north. G. semen previous year's abundance, a proxy for cyst production and recruitment, had a strong control on the dynamics, likely contributing to the stability of high density populations in southern lakes. Conversely, the effects of climate and habitat were weaker and their influence varied across regions. Temperature was the limiting factor in the north whereas local habitat was more important in the south. Synthesis: A full understanding of the mechanisms driving abundance changes across large scales can only be gained if endogenous and environmental factors are analyzed together. For phytoplankton species, and specially, noxious microalgae, this implies that proxies for cyst production and recruitment, which are the inoculum for next year population, should be included in e.g. distribution, bloom formation and climate models, as these may modify establishment and population response to environmental variation. Asynchronous changes in abundance across regions also indicate that management plans should be developed for small regions, as inference at a large scale may obscure the mechanisms driving local population changes.

环境随机性（environmental stochasticity）对于阐释入侵物种（invasive species）的存续与定殖机制至关重要，但同时区分环境因素与种群统计因素的协同影响却极具挑战。明晰种群统计与环境因素如何在大时空尺度下调控入侵物种的丰度，对于预判存在定殖风险的种群并制定合理的管理措施而言不可或缺。

我们采用分层混合模型（hierarchical mixed modeling）方法，针对入侵性针胞藻（raphidophyte）物种Gonyostomum semen（精液尾丝藻）——一种正在北欧快速扩散的有害鞭毛藻——的空间与年际动态展开分析，以探究其对种群统计因素与环境变异的响应规律。本研究使用了瑞典境内两个生物地理区域（南部的中部平原（Central Plains）与北部的芬诺斯坎迪亚地区（Fennoscandian region））内76个湖泊的14年监测数据。

研究结果显示，两个区域内G. semen种群的密度动态存在显著的异步性。多数南部湖泊中的G. semen种群呈现正向增长趋势（即高密度峰值出现频率逐年升高），部分湖泊内已形成稳定种群，并在连续年份中反复出现水华。与之相反，北部区域的G. semen种群规模更小，且动态随机性更强。上一年度G. semen的种群丰度——作为孢囊产生（cyst production）与种群补充（recruitment）的替代指标——对其种群动态具有极强的调控作用，这或许是维持南部湖泊高密度种群稳定性的关键因素。与之相对，气候与栖息地因素的影响则相对较弱，且其作用强度存在区域差异：北部区域的限制因子为水温，而南部区域则以本地栖息地条件更为关键。

综合分析：唯有同时考量内源性因素与环境因素，方能全面阐明大尺度下种群丰度变化的驱动机制。对于浮游植物（phytoplankton），尤其是有害微藻（noxious microalgae）而言，这意味着在构建分布模型、水华形成模型与气候模型等研究框架时，应纳入孢囊产生与种群补充的替代指标——这些指标作为次年种群的接种源，可调控物种的定殖能力与种群对环境变异的响应模式。不同区域种群丰度的异步变化同样表明，管理计划应针对小尺度区域制定，因为大尺度的推断可能会掩盖驱动本地种群动态的具体机制。