Warming-induced changes in seasonal priority effects drive shifts in community composition

Shifting community assembly dynamics are an underappreciated mechanism by which warming will alter plant community composition. Germination timing (which can determine the order in which seedlings emerge within a community) will likely shift unevenly across species in response to warming. In seasonal environments where communities reassemble at the beginning of each growing season, changes in germination timing could lead to changes in seasonal priority effects, and ultimately community composition. We test this expectation by assembling mesocosms of 15 species in one of two orders-"ambient" assembly order or "warmed" assembly order-based on the order in which the constituent species germinated under ambient and warmed conditions. Community composition differed significantly between mesocosms assembled in ambient versus warmed orders. The impact of assembly order on species mean biomass was largely explained by how much earlier (or later) a species arrived in the warmed-order treatment relative to the ambient-order treatment. Species whose germination phenology advanced more under warmed conditions relative to ambient conditions showed greater relative increases in biomass under the warmed assembly treatment. These findings demonstrate that warming can drive community assembly and shape community composition by reordering the relative timing of germination among species. These findings enhance our ability to predict which species are likely to benefit from warming and which may decline based on how warming may shift assembly order, ultimately informing how warming may alter plant communities.

Long-term data reveal widespread phenological change across major US estuarine food webs

Climate change is shifting the timing of organismal life-history events. Although consequential food-web mismatches can emerge if predators and prey shift at different rates, research on phenological shifts has traditionally focused on single trophic levels. Here, we analysed >2000 long-term, monthly time series of phytoplankton, zooplankton, and fish abundance or biomass for the San Francisco, Chesapeake, and Massachusetts bays. Phenological shifts occurred in over a quarter (28%) of the combined series across all three estuaries. However, phenological trends for many taxa (ca. 29-68%) did not track the changing environment. While planktonic taxa largely advanced their phenologies, fishes displayed broad patterns of both advanced and delayed timing of peak abundance. Overall, these divergent patterns illustrate the potential for climate-driven trophic mismatches. Our results suggest that even if signatures of global climate change differ locally, widespread phenological change has the potential to disrupt estuarine food webs.

Limited evidence for phenological differences between non-native and native species

Although many species shift their phenology with climate change, species vary significantly in the direction and magnitude of these responses (i.e., phenological sensitivity). Studies increasingly detect early phenology or high phenological sensitivity to climate in non-native species, which may favor non-native species over natives in warming climates. Yet relatively few studies explicitly compare phenological responses to climate between native vs. non-native species or between non-native populations in the native vs. introduced range, limiting our ability to quantify the role of phenology in invasion success. Here, we review the empirical evidence for and against differences in phenology and phenological sensitivity to climate in both native vs. non-native species and native and introduced populations of non-native species. Contrary to common assumptions, native and non-native plant species did not consistently differ in mean phenology or phenological sensitivity. However, non-native plant species were often either just as or more sensitive, but rarely less sensitive, to climate as natives. Introduced populations of non-native plant species often show earlier reproduction than native populations of the same species, but there was mixed evidence for differences in phenological sensitivity between introduced and native plant populations. We found very few studies comparing native vs. invasive animal phenology. Future work should characterize phenological sensitivity to climate in native vs. non-native plant and animal species, in native vs. introduced populations of non-native species, and across different stages of invasion, and should carefully consider how differences in phenology might promote invasion success or disadvantage native species under climate change.

Phenological displacement is uncommon among sympatric angiosperms

Interactions between species can influence successful reproduction, resulting in reproductive character displacement, where the similarity of reproductive traits - such as flowering time - among close relatives growing together differ from when growing apart. Evidence for the overall prevalence and direction of this phenomenon, and its stability under environmental change, remains untested across large scales. Using the power of crowdsourcing, we gathered phenological information from over 40 000 herbarium specimens, and investigated displacement in flowering time across 110 animal-pollinated species in the eastern USA. Overall, flowering time displacement is not common across large scales. However, displacement is generally greater among species pairs that flower close in time, regardless of direction. Furthermore, with climate change, the flowering times of closely related species are predicted, on average, to shift further apart by the mid-21^st century. We demonstrate that the degree and direction of phenological displacement among co-occurring closely related species pairs varies tremendously. However, future climate change may alter the differences in reproductive timing among many of these species pairs, which may have significant consequences for species interactions and gene flow. Our study provides one promising path towards understanding how the phenological landscape is structured and may respond to future environmental change.

Using Convolutional Neural Networks to Efficiently Extract Immense Phenological Data From Community Science Images

Community science image libraries offer a massive, but largely untapped, source of observational data for phenological research. The iNaturalist platform offers a particularly rich archive, containing more than 49 million verifiable, georeferenced, open access images, encompassing seven continents and over 278,000 species. A critical limitation preventing scientists from taking full advantage of this rich data source is labor. Each image must be manually inspected and categorized by phenophase, which is both time-intensive and costly. Consequently, researchers may only be able to use a subset of the total number of images available in the database. While iNaturalist has the potential to yield enough data for high-resolution and spatially extensive studies, it requires more efficient tools for phenological data extraction. A promising solution is automation of the image annotation process using deep learning. Recent innovations in deep learning have made these open-source tools accessible to a general research audience. However, it is unknown whether deep learning tools can accurately and efficiently annotate phenophases in community science images. Here, we train a convolutional neural network (CNN) to annotate images of Alliaria petiolata into distinct phenophases from iNaturalist and compare the performance of the model with non-expert human annotators. We demonstrate that researchers can successfully employ deep learning techniques to extract phenological information from community science images. A CNN classified two-stage phenology (flowering and non-flowering) with 95.9% accuracy and classified four-stage phenology (vegetative, budding, flowering, and fruiting) with 86.4% accuracy. The overall accuracy of the CNN did not differ from humans (p = 0.383), although performance varied across phenophases. We found that a primary challenge of using deep learning for image annotation was not related to the model itself, but instead in the quality of the community science images. Up to 4% of A. petiolata images in iNaturalist were taken from an improper distance, were physically manipulated, or were digitally altered, which limited both human and machine annotators in accurately classifying phenology. Thus, we provide a list of photography guidelines that could be included in community science platforms to inform community scientists in the best practices for creating images that facilitate phenological analysis.

Ten-year trends reveal declining quality of seeded pollinator habitat on reclaimed mines regardless of seed mix diversity

Plant-pollinator interactions represent a crucial ecosystem function threatened by anthropogenic landscape changes. Disturbances that reduce plant diversity are associated with floral resource and pollinator declines. Establishing wildflower plantings is a major conservation strategy targeting pollinators, the success of which depends on long-term persistence of seeded floral communities. However, most pollinator-oriented seeding projects are monitored for a few years, making it difficult to evaluate the longevity of such interventions. Selecting plant species to provide pollinators diverse arrays of floral resources throughout their activity season is often limited by budgetary constraints and other conservation priorities. To evaluate the long-term persistence of prairie vegetation seeded to support pollinators, we sowed wildflower seed mixes into plots on a degraded reclaimed strip-mine landscape in central Ohio, USA. We examined how pollinator habitat quality, measured as floral abundance and diversity, changed over 10 years (2009-2019) in the absence of management, over the course of the blooming season within each year, and across three seed mixes containing different numbers and combinations of flowering plant species. Seeded species floral abundance declined by more than 75% over the study, with the largest decline occurring between the fifth and seventh summers. Native and non-native adventive flowering plants quickly colonized the plots and represented >50% of floral community abundances on average. Floral richness remained relatively constant throughout the study, with a small peak one year after plot establishment. Plots seeded with High-Diversity Mixes averaged two or three more species per plot compared with a Low-Diversity Mix, despite having been seeded with twice as many plant species. Within years, the abundance and diversity of seeded species were lowest early in the blooming season and increased monotonically from June to August. Adventive species exhibited the opposite trend, such that complementary abundance patterns of seeded and adventive species blooms resulted in a relatively constant floral abundance across the growing season. Seeded plant communities followed classic successional patterns in which annual species quickly established and flowered but were replaced by perennial species after the first few summers. Long-term data on establishment and persistence of flower species can guide species selection for future-oriented pollinator habitat restorations.

Plant phenological responses to experimental warming-A synthesis

Although there is abundant evidence that plant phenology is shifting with climatic warming, the magnitude and direction of these shifts can depend on the environmental context, plant species, and even the specific phenophase of study. These disparities have resulted in difficulties predicting future phenological shifts, detecting phenological mismatches and identifying other ecological consequences. Experimental warming studies are uniquely poised to help us understand how climate warming will impact plant phenology, and meta-analyses allow us to expose broader trends from individual studies. Here, we review 70 studies comprised 1226 observations of plant phenology under experimental warming. We find that plants are advancing their early-season phenophases (bud break, leaf-out, and flowering) in response to warming while marginally delaying their late-season phenophases (leaf coloration, leaf fall, and senescence). We find consistency in the magnitude of phenological shifts across latitude, elevation, and habitat types, whereas the effect of warming on nonnative annual plants is two times larger than the effect of warming on native perennial plants. Encouragingly for researchers, plant phenological responses were generally consistent across a variety of experimental warming methods. However, we found numerous gaps in the experimental warming literature, limiting our ability to predict the effects of warming on phenological shifts. In particular, studies outside of temperate ecosystems in the Northern Hemisphere, or those that focused on late-season phenophases, annual plants, nonnative plants, or woody plants and grasses, were underrepresented in our data set. Future experimental warming studies could further refine our understanding of phenological responses to warming by setting up experiments outside of traditionally studied biogeographic zones and measuring multiple plant phenophases (especially late-season phenophases) across species of varying origin, growth form, and life cycle.

Extirpated prairie species demonstrate more variable phenological responses to warming than extant congeners

PREMISE

Shifting phenology in response to climate is one mechanism that can promote population persistence and geographic spread; therefore, species with limited ability to phenologically track changing environmental conditions may be more susceptible to population declines. Alternatively, apparently nonresponding species may demonstrate divergent responses to multiple environmental conditions experienced across seasons.

METHODS

Capitalizing on herbarium records from across the midwestern United States and on detailed botanical surveys documenting local extinctions over the past century, we investigated whether extirpated and extant taxa differ in their phenological responses to temperature and precipitation during winter and spring (during flowering and the growing season before flowering) or in the magnitude of their flowering time shift over the past century.

RESULTS

Although warmer temperatures across seasons advanced flowering, extirpated and extant species differed in the magnitude of their phenological responses to winter and spring warming. Extirpated species demonstrated inconsistent phenological responses to warmer spring temperatures, whereas extant species consistently advanced flowering in response to warmer spring temperatures. In contrast, extirpated species advanced flowering more than extant species in response to warmer winter temperatures. Greater spring precipitation tended to delay flowering for both extirpated and extant taxa. Finally, both extirpated and extant taxa delayed flowering over time.

CONCLUSIONS

This study highlights the importance of understanding phenological responses to seasonal warming and indicates that extirpated species may demonstrate more variable phenological responses to temperature than extant congeners, a finding consistent with the hypothesis that appropriate phenological responses may reduce species' likelihood of extinction.