Understanding the biodiversity and biogeography of plankton in the ocean is essential for predicting responses to environmental changes and informing ocean conservation and management strategies. Species distribution models (SDMs) are a pivotal tool in this regard. This study used data from a global marine ecosystem model as a testbed to assess the reliability of various SDMs, including Generalized Linear Model (GLM), Generalized Additive Model (GAM), Random Forest (RF), Boosted Regression Trees (BRT) and Artificial Neural Network (ANN). We used artificial datasets to replicate the sampling patterns of three datasets: a compiled dataset of global scope, the Tara Ocean dataset, and the Atlantic Meridional Transect (AMT) project. Our findings indicate that tree-based algorithms, RF and BRT, exhibit better predictive accuracy and stability compared to GLM, GAM, and ANN, especially when trained with more spatially resolved datasets. We highlight the significant influence of sampling bias on model performance, with models trained on more comprehensive global datasets outperforming those trained on more latitudinally and longitudinally biased data respectively (Tara and AMT). Furthermore, we demonstrate that broad spatial coverage is a more critical determinant of predictive skill than sample size alone, as simply increasing sampling density within a biased region is insufficient to overcome poor spatial representation. Overall, this research underscores the necessity of careful consideration of sampling strategies and model selection in plankton species distribution modelling.

Marine photoautotrophs have evolved to exploit the ocean's variable light conditions, with chromatic acclimators being able to adjust their pigment content to better match the ambient light color. The impact of chromatic acclimation on phytoplankton distribution and competition is not well understood despite its global importance. This study explores chromatic acclimation's role in shaping the biogeography of Synechococcus, a widespread cyanobacterium. We integrated three pigment types into a global ecosystem model: a green-light specialist, a blue-light specialist, and a chromatic acclimator. Laboratory studies defined each type's specific absorption properties. Our results indicate that chromatic acclimation offers an evolutionary advantage by enabling Synechococcus to adapt to varying light environments. This ability to mimic blue- and green-light specialists and enhance absorption at intermediate states, particularly in areas with high seasonal light variations, increases Synechococcus distribution and biomass. Thus, chromatic acclimation affects ecosystem functioning and biogeochemical processes in the ocean.

The subsurface chlorophyll maximum (SCM) is increasingly recognised as an important but understudied locus of primary production particularly in shelf seas. Here we report the results of a 4 year, repeat station, summer sampling programme (2013 – 2016) of a seasonally recurrent SCM in the Western English Channel. Interannual variability in phytoplankton community structure and chlorophyll distribution and intensity was strongly related to water column stability at the depth interval of the SCM and also to water temperature. The phytoplankton community was statistically distinct in each year. High stability, as evidenced by large Richardson numbers and a well-developed strong thermocline appeared to favour the growth of larger dinoflagellates (autotrophs or mixotrophs) and diatoms. Such conditions led to development of the most intense SCMs and these were sometimes dominated by a single or a few key species most prominently in 2015 with near monospecific concentrations of the dinoflagellate Tripos fusus with average peak SCM chlorophyll concentrations of 7.3 ± 4.4 μg l-1. By contrast, in years with low water column stability and intermittent turbulence at the thermocline (2014, 2016) there was greater chlorophyll dispersal and less intense SCM. In these low stability conditions, red fluorescent nano-phytoplankton, such as naked dinoflagellates, chlorophytes and prymnesiophytes, made a greater contribution to the community, possibly as a result of the advantages that motility and enhanced light utilisation efficiency confer within an SCM exposed to turbulence. It is also likely that turbulence disrupted the stability required by the larger dinoflagellates and diatoms. Several of the key SCM taxa were absent from surface waters including the dinoflagellates Tripos fusus, Tripos lineatus, and most of the Rhizosolenia/ Proboscia diatoms, consistent with adaptations more suited to survival at depth in stratified waters. These traits include luxury nutrient uptake and storage and survival in low light (both groups) and mixotrophy (dinoflagellates). On the other hand, in 2013, diatoms including Pseudo-nitzschia spp. were abundant in both surface, SCM and bottom waters. The relatively cooler waters (11.6 - 12.1 °C on average in 2013 and 2016) were characterised by smaller diatoms (Chaetoceros spp. and Pseudo-nitzschia spp.) whereas the warmer waters (13.1 °C on average in 2014) contained larger diatoms (large Rhizosolenia spp., Lauderia annulata and Leptocylindrus danicus). There did not appear to be continuity of key species between years, other than for the dinoflagellate Tripos lineatus, which was significant in both 2013 and 2014 and present in 2015. In any given year, there was no correspondence between the key spring bloom phytoplankton species as monitored in the nearby Western Channel Observatory L4 station and the key SCM taxa.

Despite the importance of Agulhas Bank (AB) marine productivity in supporting South African coastal fisheries and shelf ecosystems, there are relatively few regional-scale assessments of its spatial and temporal variability, and most productivity studies have been limited in scale. Here we use satellite-derived Net Primary Production (NPP) rates calculated using the Vertically Generalized Production Model (VGPM) to examine the spatial and temporal dynamics of NPP over the 21-year satellite record (1998???2018) on the AB. In calculating VGPM NPP we used the OCCI Chlorophyll-a product, SST from Operational-Sea-Surface-Temperature-and-Sea-Ice-Analysis (OSTIA) and PAR from GlobColour level-3 mapped products as these represent the longest datasets that fit our extended study period. We examine spatial trends between the eastern and central AB, as well as three areas of the bank (around Port Alfred, the Tsitsikamma coast, and the ???cold ridge???) that have been previously identified as contributing significantly to the overall productivity of the AB. The AB shows only a moderate degree of seasonality in NPP calculated from the VGPM, with NPP being highest in austral summer (1.7???1.8 g C m- 2 d-1) and lowest in winter (0.9???1.0 g C m- 2 d-1), and remains relatively high (>1 g C m- 2 d-1) throughout the year, contrasting sharply with other shelf systems. Considered annually, NPP on the bank was 516 g C m- 2 yr- 1 (38 Mt C yr- 1 when scaled to the total shelf area) which is higher than many other shelf systems though lower than the neighbouring Benguela system and is indicative of a moderately productive shelf system fuelled by perennial NPP. Comparing different sections of the AB from east to central bank, and including the three upwelling areas, highlighted that spatial differences in NPP were relatively limited; that these three upwelling areas made similar contributions to their relative proportion of the total shelf area, and that average rates of NPP are spatially similar across the bank, though notable high rates occur in some coastal upwelling areas. Interannual variability in NPP was relatively modest, varying between years by only -15% over the two decades assessed. Over the 21year data set, there was a slight (-0.26% yr- 1) statistically-significant decline in calculated NPP over time for the AB as a whole, which, when examined on a pixel-by-pixel basis, indicated that most of the decline was on the central bank between 100 m and 200 m isobaths. In summer, an increase in NPP occurred on the EAB (26.5???28???E). In conclusion, the AB is a significant site of perennial moderate levels of NPP, varying little interannually and with only a slight decline in NPP over time. These factors lead to a stable environment in terms of ecosystem productivity so that the AB makes a significant contribution to the productivity of South African regional fisheries.

Improvements in physiological understanding, observational capabilities, and computational power have led modelers of phytoplankton in natural systems to represent more phytoplankton classes with better differentiated characteristics. In this chapter, we describe the key phytoplankton processes that are used to differentiate plankton, including phytoplankton absorption, photoadaptation, and elemental stoichiometry ratios. We then briefly survey model applications, in which resolving phytoplankton diversity matter are important: ocean-color, harmful algal blooms, carbon export to the deep ocean, as well as climate change-induced shifts in planktonic communities. Researchers are finding significant benefits from improved representation of phytoplankton types, so it is likely this work will continue to develop in the future.

The balance of physical and biological processes governing phytoplankton growth rates and the accumulation of biomass is widely debated in the literature, notably during the winter–spring transition. Here we show, in a temperate shelf sea that variability in the depth of the actively mixing surface layer is the leading order control. During a 2‐week period preceding the peak of the spring bloom we observe two distinct regimes; first, growth within the euphotic zone during the day and re‐distribution of new biomass to the seasonal pycnocline at night by convective mixing; then, more rapid biomass accumulation trapped within a shallower, wind‐driven actively mixing layer that was decoupled from the pycnocline below. Our observations of the bloom in the Celtic Sea, Northwest European Shelf, were made using ocean gliders and include measurements of the dissipation of turbulent kinetic energy. A 1‐D phytoplankton growth model driven by our measurements of dissipation and incident irradiance replicates the observed bloom and reinforces the conclusion that physical processes that mediate light availability were key. Day‐to‐day variability in cloud cover and the ability of phytoplankton to acclimate to their light environment were also important factors in determining growth rates, and the timing of the biomass peak. Our results emphasize the need for accurate turbulent mixing parameterizations in coupled hydrodynamic‐ecosystem models. Our findings are applicable to any region where wind‐driven mixing can modify nutrient and light availability, especially across subpolar shelves in the northern hemisphere where light rather than nutrients is typically the limiting factor on phytoplankton growth.

An established one-dimensional Shelf Sea Physics and Primary Production (S2P3) model has been developed into three different new models: S2P3-NPZ which includes a nutrient–phytoplankton–zooplankton (NPZ) framework, where the grazing rate is no longer fixed but instead varies over time depending on different functions chosen to represent the predator–prey relationship between zooplankton and phytoplankton; S2P3-Photoacclim which includes a representation of the process of photo-acclimation and flexible stoichiometry in phytoplankton; and S2P3 v8.0 which combines the NPZ framework and the variable stoichiometry of phytoplankton at the same time. These model formulations are compared to buoy and conductivity–temperature–depth (CTD) observations, as well as zooplankton biomass and in situ phytoplankton physiological parameters obtained in the central Celtic Sea (CCS). Models were calibrated by comparison to observations of the timing and magnitude of the spring phytoplankton bloom, magnitude of the spring zooplankton bloom, and phytoplankton physiological parameters obtained throughout the water column. A sensitivity study was also performed for each model to understand the effects of individual parameters on model dynamics. Results demonstrate that better agreement with biological observations can be obtained through the addition of representations of photo-acclimation, flexible stoichiometry, and grazing provided these can be adequately constrained.

Biodiversity of phytoplankton is important for ecosystem stability and marine biogeochemistry. However, the large-scale patterns of diversity are not well understood and are often poorly characterized in terms of statistical relationships with factors such as latitude, temperature and productivity. Here we use ecological theory and a global trait-based ecosystem model to provide mechanistic understanding of patterns of phytoplankton diversity. Our study suggests that phytoplankton diversity across three dimensions of trait space (size, biogeochemical function and thermal tolerance) is controlled by disparate combinations of drivers: the supply rate of the limiting resource, the imbalance in different resource supplies relative to competing phytoplankton demands, size-selective grazing and transport by the moving ocean. Using sensitivity studies we show that each dimension of diversity is controlled by different drivers. Models including only one (or two) of the trait dimensions will have different patterns of diversity than one which incorporates another trait dimension. We use the results of our model exploration to infer the controls on the diversity patterns derived from field observations along meridional transects in the Atlantic and to explain why different taxa and size classes have differing patterns.

Continental margins are disproportionally important for global primary production, fisheries and CO2 uptake. However, across the Northeast Atlantic shelves, there has been an ongoing summertime decline of key biota—large diatoms, dinoflagellates and copepods—that traditionally fuel higher tropic levels such as fish, sea birds and marine mammals. Here, we combine multiple time series with in situ process studies to link these declines to summer nutrient stress and increasing proportions of picophytoplankton that can comprise up to 90% of the combined pico‐ and nanophytoplankton biomass in coastal areas. Among the pico‐fraction, it is the cyanobacterium Synechococcus that flourishes when iron and nitrogen resupply to surface waters are diminished. Our field data show how traits beyond small size give Synechococcus a competitive edge over pico‐ and nanoeukaryotes. Key is their ability to grow at low irradiances near the nutricline, which is aided by their superior light‐harvesting system and high affinity to iron. However, minute size and lack of essential biomolecules (e.g. omega‐3 polyunsaturated fatty acids and sterols) render Synechococcus poor primary producers to sustain shelf sea food webs efficiently. The combination of earlier spring blooms and lower summer food quantity and quality creates an increasing period of suboptimal feeding conditions for zooplankton at a time of year when their metabolic demand is highest. We suggest that this nutrition‐related mismatch has contributed to the widespread, ~50% decline in summer copepod abundance we observe over the last 60 years. With Synechococcus clades being prominent from the tropics to the Arctic and their abundances increasing worldwide, our study informs projections of future food web dynamics in coastal and shelf areas where droughts and stratification lead to increasing nutrient starvation of surface waters.

The Western English Channel is a seasonally stratified temperate coastal sea where a subsurface chlorophyll maximum (SCM) is typically detectable within the seasonal thermocline. The SCM often develops as a thin layer ( < 5 m) that may contain elevated concentrations of phytoplankton (subsurface chlorophyll maximum thin layer; SCMTL). During summer 2013 a study was conducted offshore of Falmouth, UK to assess spatial and short-term temporal variability in SCM thickness in relation to water column structure and physical conditions and to evaluate any associated changes in phytoplankton community structure. SCMTLs were observed in 18 of 52 vertical profiles, typically characterised by higher chlorophyll concentrations than broader SCMs. SCMTLs were generally associated with a 'stepped' thermocline, likely representing the presence of one or more shallow mixed layers forming above/within the seasonal thermocline, and related to increased stratification compared to broader SCMs. Pseudo-nitzschia was almost exclusively the dominant diatom taxon in SCMs, yet statistically distinct differences in community structure existed between SCMTLs and broader SCMs. Within the phytoplankton, the distinction was largely due to a greater biomass of Proboscia alata and other rhizosolenid diatoms, and the dinoflagellate Ceratium lineaturn in SCMTLs, and a smaller population of the diatom Chaetoceros spp. There was also a distinction amongst heterotrophic dinoflagellates, with enhanced biomass of Gyrodinium spp. in SCMTLs and a reduction in Diplopsalis lenticula. We propose that this observed difference resulted from promotion of phytoplankton better adapted to environmental conditions more specific to SCMTLs compared to broader SCMs. With more intense and prolonged stratification projected for the NW European shelf, there may be increased prevalence of SCMTLs and the associated larger-sized specialised taxa, with implications for increased carbon export. This study adds to a growing body of evidence of the importance of SCMTLs in coastal and shelf seas, and highlights the requirement for improved understanding of physical forcing and the ecology and physiology of key taxa, particularly as predicted changes in stratification could alter the role of SCM phytoplankton in a future influenced by climate change.

Monitoring changes in marine phytoplankton is important as they form the foundation of the marine food web and are crucial in the carbon cycle. Often Chlorophyll-a (Chl-a) is used to track changes in phytoplankton, since there are global, regular satellite-derived estimates. However, satellite sensors do not measure Chl-a directly. Instead, Chl-a is estimated from remote sensing reflectance (R-RS): the ratio of upwelling radiance to the downwelling irradiance at the ocean's surface. Using a model, we show that R-RS in the blue-green spectrum is likely to have a stronger and earlier climate-change-driven signal than Chl-a. This is because R-RS has lower natural variability and integrates not only changes to in-water Chl-a, but also alterations in other optically important constituents. Phytoplankton community structure, which strongly affects ocean optics, is likely to show one of the clearest and most rapid signatures of changes to the base of the marine ecosystem.

S1. Ecosystem model parametrization S1.1. Phytoplankton Growth:Phytoplankton growth rates were parameterized as functions of maximum photosynthetic rate, local light, nutrients and temperature.As in Dutkiewicz et al (2015a), we follow Geider et al (1998) such that the growth rate for phytoplankton j is equal to the carbon-specific photosynthesis rate:

This article provides a proof of concept for using a biogeochemical/ecosystem/optical model with a radiative transfer component as a laboratory to explore aspects of ocean colour. We focus here on the satellite ocean colour chlorophyll a (Chl a) product provided by the often-used blue/green reflectance ratio algorithm. The model produces output that can be compared directly to the real-world ocean colour remotely sensed reflectance. This model output can then be used to produce an ocean colour satellite-like Chl a product using an algorithm linking the blue versus green reflectance similar to that used for the real world. Given that the model includes complete knowledge of the (model) water constituents, optics and reflectance, we can explore uncertainties and their causes in this proxy for Chl a (called derived Chl a in this paper). We compare the derived Chl a to the actual model Chl a field. In the model we find that the mean absolute bias due to the algorithm is 22 % between derived and actual Chl a. The real-world algorithm is found using concurrent in situ measurement of Chl a and radiometry. We ask whether increased in situ measurements to train the algorithm would improve the algorithm, and find a mixed result. There is a global overall improvement, but at the expense of some regions, especially in lower latitudes where the biases increase. Not surprisingly, we find that region-specific algorithms provide a significant improvement, at least in the annual mean. However, in the model, we find that no matter how the algorithm coefficients are found there can be a temporal mismatch between the derived Chl a and the actual Chl a. These mismatches stem from temporal decoupling between Chl a and other optically important water constituents (such as coloured dissolved organic matter and detrital matter). The degree of decoupling differs regionally and over time. For example, in many highly seasonal regions, the timing of initiation and peak of the spring bloom in the derived Chl a lags the actual Chl a by days and sometimes weeks. These results indicate that care should also be taken when studying phenology through satellite-derived products of Chl a. This study also reemphasizes that ocean-colour-derived Chl a is not the same as the real in situ Chl a. In fact the model derived Chl a compares better to real-world satellite-derived Chl a than the model actual Chl a. Modellers should keep this is mind when evaluating model output with ocean colour Chl a and in particular when assimilating this product. Our goal is to illustrate the use of a numerical laboratory that (a) helps users of ocean colour, particularly modellers, gain further understanding of the products they use and (b) helps the ocean colour community to explore other ocean colour products, their biases and uncertainties, as well as to aid in future algorithm development.

To improve our understanding of the role of phytoplankton for marine ecosystems and global biogeochemical cycles, information on the global distribution of major phytoplankton groups is essential. Although algorithms have been developed to assess phytoplankton diversity from space for over two decades, so far the application of these data sets has been limited. This scientific roadmap identifies user needs, summarizes the current state of the art, and pinpoints major gaps in long-term objectives to deliver space-derived phytoplankton diversity data that meets the user requirements. These major gaps in using ocean color to estimate phytoplankton community structure were identified as: (a) the mismatch between satellite, in situ and model data on phytoplankton composition, (b) the lack of quantitative uncertainty estimates provided with satellite data, (c) the spectral limitation of current sensors to enable the full exploitation of backscattered sunlight, and (d) the very limited applicability of satellite algorithms determining phytoplankton composition for regional, especially coastal or inland, waters. Recommendation for actions include but are not limited to: (i) an increased communication and round-robin exercises among and within the related expert groups, (ii) the launching of higher spectrally and spatially resolved sensors, (iii) the development of algorithms that exploit hyperspectral information, and of (iv) techniques to merge and synergistically use the various streams of continuous information on phytoplankton diversity from various satellite sensors' and in situ data to ensure long-term monitoring of phytoplankton composition.

The differences among phytoplankton carbon ($C_{phy}$) predictions from six ocean colour algorithms are investigated by comparison with \textit{in situ} estimates of phytoplankton carbon. The common satellite data used as input for the algorithms is the Ocean Colour Climate Change Initiative merged product. The matching \textit{in situ} data are derived from flow cytometric cell counts and per-cell carbon estimates for different types of pico-phytoplankton. This combination of satellite and \textit{in situ} data provides a relatively large matching dataset (N$>$500), which is independent from most of the algorithms tested and spans almost two orders of magnitude in $C_{phy}$. Results show that not a single algorithm outperforms any of the other when using all matching data. Concentrating on the oligotrophic regions ($B$ \textless 0.15 mg\,Chl\,m$^{-3}$), where flow cytometric analysis captures most of the phytoplankton biomass, reveals significant differences in algorithm performance. The bias ranges from -35\% to +150\% and RMSD (unbiased) from 5 to 10 mg\,C\,m$^{-3}$ among algorithms, with chlorophyll-based algorithms performing better than the rest. The backscattering-based algorithms produce different reasults at the clearest waters and these differences are discussed in terms of the different algorithms used for $b_{bp}$ retrieval.

Particulate Organic Carbon (POC) plays a vital role in the ocean carbon cycle. Though relatively small compared with other carbon pools, the POC pool is responsible for large fluxes and is linked to many important ocean biogeochemical processes. The satellite ocean-colour signal is influenced by particle composition, size, and concentration and provides a way to observe variability in the POC pool at a range of temporal and spatial scales. To provide accurate estimates of POC concentration from satellite ocean colour data requires algorithms that are well validated, with uncertainties characterised. Here, a number of algorithms to derive POC using different optical variables are applied to merged satellite ocean colour data provided by the Ocean Colour Climate Change Initiative (OC-CCI) and validated against the largest database of $\textit{in situ}$ POC measurements currently available. The results of this validation exercise indicate satisfactory levels of performance from several algorithms (highest performance was observed from the algorithms of \cite{stramski2008} and \cite{loisel2002}) and uncertainties that are within the requirements of the user community. Estimates of the standing stock of the POC can be made by applying these algorithms, and yield an estimated mixed-layer integrated global stock of POC between 0.77 and 1.3 Pg C of carbon. Performance of the algorithms vary regionally, suggesting that blending of region-specific algorithms may provide the best way forward for generating global POC products.

We present a numerical model of the ocean that couples a three-stream radiative transfer component with a marine biogeochemical–ecosystem component in a dynamic three-dimensional physical framework. The radiative transfer component resolves the penetration of spectral irradiance as it is absorbed and scattered within the water column. We explicitly include the effect of several optically important water constituents (different phytoplankton functional types; detrital particles; and coloured dissolved organic matter, CDOM). The model is evaluated against in situ-observed and satellite-derived products. In particular we compare to concurrently measured biogeochemical, ecosystem, and optical data along a meridional transect of the Atlantic Ocean. The simulation captures the patterns and magnitudes of these data, and estimates surface upwelling irradiance analogous to that observed by ocean colour satellite instruments. We find that incorporating the different optically important constituents explicitly and including spectral irradiance was crucial to capture the variability in the depth of the subsurface chlorophyll a (Chl a) maximum. We conduct a series of sensitivity experiments to demonstrate, globally, the relative importance of each of the water constituents, as well as the crucial feedbacks between the light field, the relative fitness of phytoplankton types, and the biogeochemistry of the ocean. CDOM has proportionally more importance at attenuating light at short wavelengths and in more productive waters, phytoplankton absorption is relatively more important at the subsurface Chl a maximum, and water molecules have the greatest contribution when concentrations of other constituents are low, such as in the oligotrophic gyres. Scattering had less effect on attenuation, but since it is important for the amount and type of upwelling irradiance, it is crucial for setting sea surface reflectance. Strikingly, sensitivity experiments in which absorption by any of the optical constituents was increased led to a decrease in the size of the oligotrophic regions of the subtropical gyres: lateral nutrient supplies were enhanced as a result of decreasing high-latitude productivity. This new model that captures bio-optical feedbacks will be important for improving our understanding of the role of light and optical constituents on ocean biogeochemistry, especially in a changing environment. Further, resolving surface upwelling irradiance will make it easier to connect to satellite-derived products in the future.