Journal of Ecology ( IF 5.3 ) Pub Date : 2022-05-16 , DOI: 10.1111/1365-2745.13926 Kerissa Fuccillo Battle 1, 2 , Anna Duhon 3 , Conrad R. Vispo 3 , Theresa M. Crimmins 4 , Todd N. Rosenstiel 1 , Lilas L. Armstrong‐Davies 2 , Catherine E. de Rivera 1
|
1 INTRODUCTION
A rapidly growing body of studies from around the globe demonstrate clear impacts of changing climate conditions on plant and animal phenology (e.g., Cohen et al., 2018; Menzel et al., 2020; Parmesan & Yohe, 2003; Root et al., 2003). The effect of climate change on the phenology of organisms is of primary ecological concern due to potentially profound ecosystem impacts. The cascading consequences of phenological change include disruptions in interactions among species, ecosystem structure and functioning, and nutrient cycling (Edwards & Richardson, 2004; Rafferty et al., 2015; Thackeray et al., 2016; Visser & Both, 2005). The inherent variability in phenological changes across studies, sites and species makes it difficult to determine which species and interactions deserve priority for further investigation, or to incorporate phenology into ecological forecasts related to impacts on climate change (Peñuelas et al., 2009).
Long-term, repeated observations offer the greatest insight into the direction and magnitude of change in species' phenology; some of our best records of change originate from long-term observation records (e.g., Cook et al., 2008; Miller-Rushing & Primack, 2008; Primack et al., 2009). Records encompassing multiple species enable an understanding of varying rates of change among members of an ecological community. Phenology records collected across multiple sites enable examination of species response to varying local conditions. Phenology records that are long-term in duration, that are comprised of multiple species and that encompass observations from multiple sites are exceptionally rare in the United States; yet, these types of records, when found and integrated properly, offer extraordinary potential for addressing questions relating to the ecological effects of climate-driven shifts.
Here, we present an analysis of newly uncovered historical phenology records across the state of New York from the 19th century and compare them with a contemporary phenology data set with similar taxonomic and geographic dimensionality collected with comparable protocols by citizen scientist observers in the 21st century. This combined phenology data set, encompassing 36 plant species with sufficient data, and a geographic scope of the state of New York (778 km2), enables us to explore long-term changes in plant phenology in the northeastern United States. The historical data set arises from the oldest known example of an organized network of institutions collecting paired phenology and weather data at over 90 locations through participatory science methods in the United States—a network established throughout New York State with data collection spanning 1826–1872. We combine selections from this historical data set with observations from a contemporary network of institutions and individuals collecting data in a comparable way across New York State from 2009 to 2017. The modern data are primarily derived from the New York Phenology Project, a regional affiliate of the USA National Phenology Network. These two data sets share dozens of species in common and were sampled regularly and frequently across the state of New York. To our knowledge, no other North American data set provides a multi-decade, multi-site, standardized collection of phenological data from the early industrial period.
This historical initiative, explicitly designed to collect phenological data and concurrent, standardized meteorological information for an entire region, was unprecedented for North America at the time. Prior to this analysis, most attempts to study multispecies reactions to climate change using historical records have relied either on single-site, single-observer efforts (e.g., Thoreau's records at Concord as summarized by Miller-Rushing & Primack, 2008) or compilations of observations from diverse individuals working independently (Büntgen et al., 2022). The historical data set is especially suited for pairing with modern data because it has the geographic extent that permits association with modern regional records, and yet was conducted under the auspices of a centralized, standardizing project and persisted long enough to create a robust baseline. These strengths, combined with the diversity of species observed, allow us to not only detect broad phenological changes but also, as summarized in the predictions below, to more rigorously explore the influences of urbanization and of ecological traits such as growth form, pollination syndrome and the relative seasonality of leafing and flowering. By exploring these factors within a single data set, our work can give a more complete picture of the overlapping factors determining the ecological impacts of climate change on a regional flora.
We capitalize on the taxonomic and geographic breadth of this novel data set to test the following six predictions shaped by findings of previous studies. First, we predict that most plants in our study will show an advancement in phenology over the period of record, varying in degree of magnitude across species (Prediction 1). Around the globe, phenology in numerous plant species has advanced in recent decades (Menzel et al., 2020; Parmesan & Yohe, 2003; Root et al., 2003). Second, we predict these changes in phenology to be associated with increasing temperatures, again varying across species (Prediction 2). Several recent studies from the northeastern U.S. (Ellwood et al., 2013; Miller-Rushing & Primack, 2008; Primack et al., 2004; Schwartz, 1998) demonstrated strong associations between late winter and early spring temperatures and flower and leaf phenology and a clear advancement in these events in recent decades with increased winter and spring temperatures.
Several studies have also documented larger shifts in phenology among species that are active earlier in the spring season than those active later in the season (CaraDonna et al., 2014; Fitter & Fitter, 2002; Miller-Rushing & Primack, 2008; Panchen et al., 2014; Wolkovich et al., 2014). Species leafing out or flowering earlier in the season are subject to different conditions than species with later-season phenology, including access to light before canopy closure. Because the data we are analysing originate from similar ecoregions, our third prediction is that early spring species will show a greater advancement in the timing of their activity than species undergoing phenological transitions later in the season (Prediction 3). The taxonomic richness in our data set allows us to investigate whether these patterns will hold in observations spanning nearly 200 years.
Fourth, we predict differences in changes in phenology among different plant growth forms (trees, shrubs and forbs; Prediction 4). Growth forms vary in rooting depth as well as access to sunlight and nutrients. As such, changes in the availability of these resources can impact growth forms differentially, resulting in varying impacts to their leaf out and flowering timing. Recent work shows that tree phenology may be shifting more quickly than the phenology of understory forbs in this region (Heberling et al., 2019), though comparisons in phenological change among growth forms has been limited to only a handful of studies (e.g., Calinger et al., 2013; Crimmins et al., 2010, 2011; Heberling et al., 2019).
Fifth, we predict that species in urban locations will show a greater advancement in their phenology than their rural counterparts (Prediction 5). More developed areas tend to exhibit higher temperatures in a phenomenon termed the urban heat island effect. Leaf and flowering phenology are frequently advanced in urban areas compared to nearby less developed areas (Bornstein, 1968; Imhoff et al., 2010; Neil & Wu, 2006; Zhang et al., 2004), and this effect is especially exaggerated at higher latitudes (Li et al., 2019).
Finally, we predict that insect-pollinated species will exhibit greater advancement in first flower dates due to selection promoting earlier flowering to maintain synchrony with pollinators (Prediction 6; Calinger et al., 2013; Fitter & Fitter, 2002). Pollination syndrome is a key factor to explore because of the importance of climate change impacts on plant-pollinator mutualisms (Forrest, 2015; Gérard et al., 2020; Kudo & Cooper, 2019). However, differences in phenological shifts among pollination syndromes are rarely evaluated due to the scarcity of data sets with adequate representation of wind-pollinated species.
中文翻译：
跨越两个世纪的公民科学揭示了美国东北部植物物种和功能群的物候变化
1 简介
全球范围内快速增长的研究表明气候条件变化对植物和动物物候的明显影响（例如，Cohen 等人，2018 年；Menzel 等人， 2020 年；Parmesan & Yohe， 2003 年；Root 等人， 2003 年）。由于潜在的深远生态系统影响，气候变化对生物物候的影响是主要的生态问题。物候变化的级联后果包括破坏物种间相互作用、生态系统结构和功能以及养分循环（Edwards & Richardson, 2004 ; Rafferty et al., 2015 ; Thackeray et al., 2016 ; Visser & Both, 2005）。不同研究、地点和物种的物候变化固有的可变性使得很难确定哪些物种和相互作用值得进一步调查，或将物候纳入与气候变化影响相关的生态预测（Peñuelas 等， 2009）。
长期、重复的观察可以最大程度地了解物种物候变化的方向和幅度；我们的一些最佳变化记录源自长期观察记录（例如，Cook 等人， 2008 年；Miller-Rushing & Primack， 2008 年；Primack 等人，2009 年））。包含多个物种的记录有助于了解生态群落成员之间不同的变化率。在多个地点收集的物候记录可以检查物种对不同当地条件的反应。持续时间较长、由多个物种组成并包含来自多个地点的观测的物候记录在美国极为罕见；然而，这些类型的记录，如果发现和整合得当，为解决与气候驱动的变化的生态影响有关的问题提供了非凡的潜力。
在这里，我们对 19 世纪纽约州新发现的历史物候记录进行了分析，并将它们与 21 世纪公民科学家观察员使用可比协议收集的具有相似分类学和地理维度的当代物候数据集进行了比较。这个组合物候数据集，包含 36 个具有足够数据的植物物种，以及纽约州的地理范围（778 km 2)，使我们能够探索美国东北部植物物候的长期变化。历史数据集来自已知最古老的有组织的机构网络示例，该网络通过参与式科学方法在美国收集 90 多个地点的配对物候学和天气数据——该网络在纽约州建立，数据收集时间跨度为 1826 年至 1872 年。我们将来自该历史数据集的选择与来自当代机构和个人网络的观察结果相结合，从 2009 年到 2017 年在纽约州以可比方式收集数据。现代数据主要来自纽约物候学项目，该项目的区域附属机构美国国家物候网络。这两个数据集共有几十个物种，并在纽约州定期和频繁地进行采样。据我们所知，没有其他北美数据集提供了早期工业时期物候数据的数十年、多地点、标准化的集合。
这项历史性举措明确旨在收集整个地区的物候数据和同时标准化的气象信息，这在当时的北美是前所未有的。在此分析之前，大多数使用历史记录来研究多物种对气候变化的反应的尝试要么依赖于单一地点、单一观察者的努力（例如，由 Miller-Rushing & Primack 总结的梭罗在 Concord 的记录， 2008 年）或来自独立工作的不同个人的观察结果（Büntgen 等人， 2022）。历史数据集特别适合与现代数据配对，因为它具有允许与现代区域记录相关联的地理范围，而且是在一个集中的标准化项目的主持下进行的，并且持续时间足够长以创建一个稳健的基线。这些优势与观察到的物种多样性相结合，不仅使我们能够检测广泛的物候变化，而且正如以下预测所总结的那样，更严格地探索城市化和生态特征（如生长形式、授粉综合征和叶和开花的相对季节性。通过在单个数据集中探索这些因素，我们的工作可以更全面地了解确定气候变化对区域植物群的生态影响的重叠因素。
我们利用这一新数据集的分类学和地理广度来测试由先前研究结果形成的以下六个预测。首先，我们预测我们研究中的大多数植物将在记录期间表现出物候学的进步，不同物种的程度会有所不同（预测 1）。在全球范围内，近几十年来，许多植物物种的物候学都取得了进步（Menzel 等人，2020 年；Parmesan 和 Yohe， 2003 年；Root 等人， 2003 年）。其次，我们预测这些物候学的变化与温度升高有关，同样因物种而异（预测 2）。美国东北部最近的几项研究（Ellwood 等人， 2013 年；Miller-Rushing & Primack， 2008 年）; Primack 等人， 2004 年；Schwartz， 1998 年）证明了晚冬和早春温度与花和叶物候之间的强烈关联，以及近几十年来随着冬季和春季温度的升高，这些事件的明显进展。
几项研究还记录了在春季较早活动的物种之间的物候变化比在季节后期活动的物种更大（CaraDonna 等，2014；Fitter & Fitter， 2002；Miller-Rushing & Primack， 2008；Panchen 等等人， 2014 年；Wolkovich 等人， 2014 年）。在本季较早开花或开花的物种与具有后期物候的物种相比，受到不同条件的影响，包括在树冠关闭之前获得光照。因为我们正在分析的数据来自类似的生态区，所以我们的第三个预测是，早春物种的活动时间将比本季后期经历物候转变的物种有更大的进步（预测 3）。我们数据集中的丰富分类使我们能够研究这些模式是否会在近 200 年的观察中保持不变。
第四，我们预测了不同植物生长形式（树木、灌木和杂草；预测 4）之间物候变化的差异。生长形式在生根深度以及获得阳光和养分方面各不相同。因此，这些资源可用性的变化会对生长形式产生不同的影响，从而对它们的落叶和开花时间产生不同的影响。最近的研究表明，树木物候可能比该地区林下杂草的物候变化更快（Heberling 等人， 2019 年），尽管生长形式之间物候变化的比较仅限于少数研究（例如，Calinger等人， 2013 年；Crimmins 等人，2010 年，2011 年；Heberling 等人， 2019 年）。
第五，我们预测城市地区的物种将比农村地区的物种表现出更大的物候进步（预测 5）。在被称为城市热岛效应的现象中，较发达的地区往往表现出更高的温度。与附近的欠发达地区相比，城市地区的叶子和开花物候通常更先进（Bornstein， 1968；Imhoff 等， 2010；Neil 和 Wu， 2006；Zhang 等， 2004），这种影响在较高的地区尤为夸张。纬度（Li et al., 2019）。
最后，我们预测昆虫授粉物种将在第一次开花日期方面表现出更大的进步，因为选择促进了早期开花以保持与传粉者的同步（预测 6；Calinger 等人， 2013 年；Fitter & Fitter， 2002 年）。由于气候变化对植物-传粉媒介共生的影响非常重要，因此授粉综合症是一个值得探索的关键因素（Forrest， 2015 年；Gérard 等人， 2020 年；Kudo & Cooper， 2019 年）。然而，由于缺乏充分代表风传粉物种的数据集，很少评估授粉综合征之间物候变化的差异。
京公网安备 11010802027423号