Advancing phenology in limnology and oceanography

Abstract

Phenology, the study of the seasonal timing of natural phenomena, is a central construct in ecology, focusing on interactions between temporal changes in the physical environment and the structuring of annual organismal, population, community, and ecosystem dynamics (Forrest and Miller-Rushing 2010). In aquatic ecology, phenology explicitly or implicitly forms the basis of several foundational concepts. For example, the match/mismatch hypothesis (Cushing 1990) theorizes that the survival of newly hatched fish larvae will depend on their temporal overlap with peak production of their food resources, namely plankton, and was explicitly developed from earlier phenological studies of phytoplankton (Cushing 1967) and fish spawning (Hjort 1914; Cushing 1969). The Plankton Ecology Group (PEG) model (Sommer et al. 1986, 2012) implicitly draws on phenological concepts to explain observed, predictable seasonal succession in plankton communities.

Despite the centrality of phenology in how we understand aquatic ecosystems, the study of aquatic phenology lags behind its terrestrial counterpart. We see three related explanations for slower progress in the aquatic realm. First and most simply, observing phenological phenomena in aquatic systems is difficult because they occur out of sight, and monitoring is costly as a result. Terrestrial research has benefited from the wealth of observations collected by well-coordinated volunteer networks (e.g., National Phenology Network [NPN], European Phenology Network, and the Global Phenological Monitoring Programme) that report observations often at a daily timescale outfitted with little to no equipment. Aquatic representation within these programs is largely limited to observations of the appearance of aquatic birds, large fish, amphibians, or budding/blooming of well-known riparian or wetland vegetation. The relative ease of tracking terrestrial organisms has also allowed deeper investigations of the ecological and evolutionary processes driving terrestrial phenology, including the ability of organisms to adapt to shifting seasonality (Anderson et al. 2012; Kingsolver and Buckley 2015). Thus, it is not surprising that a literature search on the study of phenology reveals a terrestrial bias, with studies dominated by topics such as the timing of bird migration or the appearance of various developmental stages among a range of plant species and locations.

Second, the problem of observing subsurface events or behaviors is compounded by the short life cycles and small body sizes of key aquatic groups. Short generation times mean that notable phenological events occur rapidly and briefly, and small body sizes allow many species to escape notice even under the best of circumstances. Thus, one cannot track the appearance and decline of a spring phytoplankton bloom or the emergence of zooplankton from diapause from shoreline observations or simple camera setups. Recording these and other aquatic phenologies requires sophisticated technology in challenging conditions (remote sensing, autonomous buoys) and/or frequent sampling over decades to be able to assess patterns and change. The end result of these logistical hurdles is that continuous decadal datasets needed for phenological studies are rare or non-existent for certain organisms (Woods et al. 2022).

The third contributing factor to the slow development of aquatic phenological research relates to the question of how aquatic scientists define phenology. The most studied phenological phenomena in marine and inland water research are the onset and melt of seasonal ice and the timing of peak phytoplankton biomass (e.g., Racault et al. 2012; Ji et al. 2013; Henson et al. 2018). Distinct for its century-long records, its rapid change in the face of climate change, and its resulting notoriety, lake ice (Sharma et al. 2016, 2019) has become the cherry blossom of the freshwater world (Aono and Kazui 2008). This example demonstrates a distinction from terrestrial studies: while terrestrial research primarily emphasizes species-level events, aquatic scientists have stretched the definition of phenology to encompass physical and chemical as well as biological events, often focusing on ecosystem processes as well as species dynamics. For example, the onset of stratification and anoxia are routinely cast in the language of phenology (Woolway et al. 2021; Rohwer et al. 2024), in part because of the importance of physical habitat on ecosystem dynamics (Ladwig et al. 2022). A second distinct attribute of aquatic phenological studies is what could be considered a further stretching or perhaps a conflation of phenology (i.e., the timing) with seasonality (i.e., the cyclical nature of events or processes, which can acknowledge or ignore their exact timing). While formal definitions of phenology emphasize the timing of events (“phenology is nature's calendar” sensu NPN), concepts or frameworks such as the PEG and match/mismatch hypothesis combine ideas and observations about seasonality, community succession, and timing. This idiosyncrasy is recurring in modern phenology papers (Johnson et al. 2024; Rodrguez-Velasco et al. 2024). One could argue this laissez-faire approach to defining phenology in aquatic systems limits a unified study of phenology in the field, but we believe it pushes innovative thinking about the temporal aspect of the interplay between physics, chemistry, and biology in limnology and oceanography.

Recently, L&O Letters published a special issue on phenology in aquatic systems. Given the magnitude of global change we are experiencing, the special issue was initiated to address the urgent need to better understand if and how aquatic phenology is changing and the implications of those changes (Feiner et al. 2022; Woods et al. 2022). What is changing or likely to shift, what abrupt changes might we expect, and what is at risk of collapsing? For example: How will the disappearance of lake and ocean ice influence trophic interactions? How will earlier leaf-out affect light and organic matter availability in streams? Will prolonged stratified summer conditions promote harmful algal blooms in lake and coastal systems? What are the implications of phenological change on the genetic structure of populations and the resilience of aquatic ecosystems to adapt to new climate regimes? How do we manage aquatic systems in a future of changing phenology? Contributions in the special issue focus on phenology in both inland and marine aquatic ecosystems across multiple levels of biological organization. The articles collectively show how changes in phenology have the power to alter typically predictable hydrological, chemical, and biological processes in aquatic ecosystems, with important influences on how humans interact with and manage these resources (Rosemartin et al. 2014). In this editorial, we highlight key findings from this new collection of articles and summarize future research directions proposed by authors.

In the special issue, many studies focused on marine phytoplankton dynamics, leveraging both long-term monitoring data (Cloern et al. 2024; Jan et al. 2024; Stevens et al. 2024) and autonomous observations from platforms such as the BGC-Argo float network (Vives et al. 2024), a cabled ocean observatory (Stevens et al. 2024), and satellite remote sensing (Cyr et al. 2024). A commonality of these studies was that in many cases, changing phytoplankton phenology is not solely linked to climate dynamics, but regulated by grazing pressure. This was also found in lakes, where Rohwer et al. (2024) showed that a species invasion altered phytoplankton phenology by triggering a trophic cascade, and Bailey and Hood (2024) showed that biotic controls, especially grazer-defended phytoplankton, modulated zooplankton phenology. Similarly, Straile and Rothhaupt (2024) leveraged decades of long-term monitoring data from Lake Constance to test PEG model predictions (Sommer et al. 1986) with respect to the interplay of oligotrophication and climate warming on the timing of the spring clear-water phase. They found that oligotrophication both advanced the spring phytoplankton bloom and muted the clear-water phase.

Building on a wealth of phenological research focused on changing lake ice conditions, Barta et al. (2024) and Hazukov et al. (2024) leveraged long-term datasets to investigate some of the ecological ramifications of changing ice conditions. Hazukov et al. (2024) found that in West Greenland, early-ice off years led to longer spring mixing than late ice-off years, which increased the temperature and dissolved oxygen concentrations in the hypolimnia of the study lakes. Barta et al. (2024) showed that across 194 Midwest US lakes, the spawning phenology of walleye (Sander vitreus) was outpaced by changing ice phenology, and that large deviations in the timing of spring spawning from historic averages were negatively associated with offspring survival. Changing spring conditions in lakes were also investigated by Gardner et al. (2024), who showed that in Lake Michigan, larval fish are at risk of diminished nearshore resources in warm years that trigger earlier offshore transport.

Rivers are also highlighted in this special issue by linking the concept of river regimes to phenology. Johnson et al. (2024) compiled silica concentrations from 200 streams and identified a limited number of distinct seasonal patterns of silica concentrations, but frequent shifting among pattern types within individual sites. Marzolf et al. (2024) identified 59 rivers in the United States where long-term daily ecosystem metabolism could be estimated. From this dataset, they found that while annual river GPP was highly variable, metabolism phenology was mostly unchanging. Furthermore, the phenology of stream thermal regimes, its influence on leaf litter breakdown, and the role of groundwater in buffering phenological changes in both temperature and leaf decomposition in a changing climate were highlighted by Hare et al. (2024).

Overall, the concept of phenology and its application in limnology and oceanography is broad. Raes et al. (2024) showed that, despite methodological differences, seasonal trends in microbial community richness and evenness remain consistent across time series sites in both the Northern and Southern Hemispheres and that microbial diversity is linked to daylength. Botrel et al. (2024) review the phenology of benthic primary production (BPP) and argue that greater focus on BPP is needed to begin to understand how phenological changes in BPP might translate to ecosystem function. Finally, with their complementary studies of reservoirs and greenhouse gas (GHG) dynamics, Rodrguez-Velasco et al. (2024) and Martnez-Garca et al. (2024) push the boundaries of phenological research into systems and topics that have rarely been considered in this context. Rodrguez-Velasco et al. (2024) highlight the inherent variability in both the strength and the shape of seasonal dynamics and driver-response relationships for emissions of different GHGs. Martnez-Garca et al. (2024) demonstrate that some of the variability may result from lags in driver-response relationships. Specifically, these authors identified a delay between inputs of carbon to reservoir sediments and the subsequent generation of CO₂ and CH₄, and argued that such lags may be common in lakes and reservoirs.

As with most research, the conclusions reached in this special issue inevitably led to the identification of more questions needing further research. We surveyed the special issue papers and identified more than two dozen questions raised by the authors, which fit into a handful of broad categories. Most commonly, authors identified the need to understand the mechanisms driving many of the recorded observations. For example, significant questions remain about the specific environmental factors driving phenological cycles in aquatic biogeochemistry (e.g., particulate organic carbon [Martnez-Garca et al. 2024], silica [Johnson et al. 2024], and gas emissions [Rodrguez-Velasco et al. 2024]) and the interaction of changing cues in spurring major phenological events in aquatic organisms (e.g., benthic primary producers, Botrel et al. 2024; zooplankton, Bailey and Hood 2024; Jan et al. 2024; and fish, Barta et al. 2024; Gardner et al. 2024). Multiple studies recommended experimental approaches to understand how changes to important demographic rates (e.g., reproduction, growth, and mortality) could assist in identifying drivers shaping phenological succession in lake, reservoir, and marine ecosystems (Botrel et al. 2024; Martnez-Garca et al. 2024; Bailey and Hood 2024; Stevens et al. 2024). Stevens et al. (2024) begin to address this gap by measuring the response of cell division rates in the phytoplankton Synechococcus to temperature and finding that this relationship explains spatial variability in Synechococcus on the Northeast U.S. Shelf. Similar work, measuring the response of demographic rates to environmental conditions through the lens of phenological shifts, could allow for more accurate modeling of community turnover in a changing climate.

The second significant uncertainty apparent in many of the publications in the special issue centered on the broader ecological consequences of some of the observed phenological shifts and changes. It is well understood that changes to one component of an aquatic ecosystem can cause trophic cascades up and down the food web to substantially affect production of higher and lower trophic levels (Carpenter and Kitchell 1996). So, it is natural to question how, for example, observed changes in the phenology of different benthic primary producers in lakes (Botrel et al. 2024) or variable shifts among different plankton taxa in marine ecosystems (Cyr et al. 2024; Jan et al. 2024) will alter overall ecological function in these systems. Barta et al. (2024) provide one example by finding a strong association between variability in a phenological phenomenon (fish spawning) with subsequent fish population productivity (the number of juveniles surviving to their first fall). Expanding beyond the observation of phenological change to a systematic understanding of how such changes will alter whole-ecosystem processes presents an area of critical need in future research.

Understanding the effects of temporal variability on estimates of overall ecosystem processes was another significant uncertainty in several special issues papers covering lotic (Hare et al. 2024; Johnson et al. 2024), lentic (Botrel et al. 2024; Martnez-Garca et al. 2024; Rodrguez-Velasco et al. 2024), and marine systems (Jan et al. 2024; Stevens et al. 2024; Vives et al. 2024). Each of these papers pointed out how the presence of significant seasonal cycles in ecosystem processes may influence conclusions about aquatic ecosystem dynamics that are reached using data at a coarse temporal resolution. This is a significant problem in aquatic research, as relatively few high-resolution datasets exist, and temporal coverage may be particularly poor during winter or during shoulder seasons (i.e., spring and fall) when thinning ice and unfavorable weather can make field sampling difficult or dangerous (Stanley et al. 2019; Bailey and Hood 2024). This represents an important gap, as these seasons are often the times of the year when the onset of different phenological processes (i.e., lake mixing, plankton blooming, fish spawning) are at their height and, therefore, may be most important for understanding the legacy effects on the remainder of the growing season (Dugan 2021; Feiner et al. 2022; Hazukov et al. 2024). Increasing the use of continuous, autonomous sampling or remote sensing data may provide a solution in some cases (Cyr et al. 2024; Stevens et al. 2024; Vives et al. 2024). Broadly, attempts must be made to understand seasonal processes, especially when failing to account for them can lead to inaccurate predictions when attempting to upscale whole-ecosystem processes (e.g., GHG emissions; Rodrguez-Velasco et al. 2024). The importance of observing phenological events over the entire annual cycle was also stressed by Jan et al. (2024), showing the importance of summer and potential autumn blooms for biological productivity.

Finally, during our survey of special issue papers, it became apparent that, like so much scientific research, the articles in this special issue were dominated by studies based on North American and European ecosystems. Indeed, only three studies in the special issue included sites outside of North America and Europe, including one study of streams in Asia (Russia, Johnson et al. 2024), one study of the Southern Ocean (Vives et al. 2024), and one study that included coastal marine sites in Australia (Raes et al. 2024). A similar Western bias shows up in the literature reviews performed by Botrel et al. (2024) and Rodrguez-Velasco et al. (2024). The lack of research and/or recognition of existing research in Asian, African, and, more generally, Southern Hemisphere ecosystems may limit our ability to generalize what we know about aquatic phenology outside of north-temperate ecosystems. This is not a new problem, nor one limited to aquatic ecology (Pyek et al. 2008; Martin et al. 2012; Trimble and van Aarde 2012; Archer et al. 2014). We highlight this simply to point out that future research in understudied climatic zones could unlock a broader ability to predict how phenology may change with a changing climate, and how a wider range of ecosystem processes may respond.

Overall, aquatic sciences will benefit from further research into the intricate interplay between biological timing and environmental shifts, notably driven by climate change, but also by other internal and external factors. This knowledge could provide critical insights for conservation and management strategies. For example, incorporating information about phenological processes can identify communities or species that may be the most vulnerable to trophic mismatches and therefore merit additional attention (e.g., Enquist et al. 2014), recognize key life history periods for accurate forecasting of future population or ecosystem states (e.g., Teichert et al. 2020), or assist in the design and timing of harvest regulations to prevent overharvest of vulnerable subpopulations (e.g., Peer and Miller 2014). Thus, there is a clear need to understand how aquatic organisms, from microbes to phytoplankton to fish, shift life history events, and what cascading effects phenological changes have on food web dynamics and energy flow within aquatic systems. Invasive species, habitat alteration, and anthropogenic pollution further complicate how life history events might change. In addition, exploring the evolutionary responses of aquatic organisms to changing phenological cues will shed light on species adaptive capacities in the face of environmental instability. Using new tools and methods to integrate phenological research in the aquatic sciences is essential for capturing dynamic ecosystem interactions, and will advance our ability to predict and mitigate the impacts of environmental change on aquatic ecosystems.