Multiple global change stressor effects on phytoplankton nutrient acquisition in a future ocean

Predicting the effects of multiple global change stressors on microbial communities remains a challenge because of the complex interactions among those factors. Here, we explore the combined effects of major global change stressors on nutrient acquisition traits in marine phytoplankton. Nutrient limitation constrains phytoplankton production in large parts of the present-day oceans, and is expected to increase owing to climate change, potentially favouring small phytoplankton that are better adapted to oligotrophic conditions. However, other stressors, such as elevated pCO2, rising temperatures and higher light levels, may reduce general metabolic and photosynthetic costs, allowing the reallocation of energy to the acquisition of increasingly limiting nutrients. We propose that this energy reallocation in response to major global change stressors may be more effective in large-celled phytoplankton species and, thus, could indirectly benefit large-more than small-celled phytoplankton, offsetting, at least partially, competitive disadvantages of large cells in a future ocean. Thus, considering the size-dependent responses to multiple stressors may provide a more nuanced understanding of how different microbial groups would fare in the future climate and what effects that would have on ecosystem functioning. This article is part of the theme issue ‘Conceptual challenges in microbial community ecology’.

Besides the direct effects on primary production, warming is also expected to enhance thermal stratification at low and mid latitudes, preventing nutrients from deep waters entering the well-lit surface mixed layer, thus exacerbating phytoplankton nutrient limitation and reducing primary production [15][16][17]. Moreover, enhanced nutrient trapping in the Southern Ocean due to climatic changes was shown to increase nutrient export to the deep ocean, further strengthening nutrient limitation [18]. Thus, present-day oceanic phytoplankton primary production is already constrained by the availability of key nutrients such as nitrogen, phosphorus and iron [19], and this limitation is expected to increase in a future ocean. Phytoplankton have developed a range of traits to deal with prevailing low-nutrient conditions. These nutrient utilization traits may change in response not only to increased nutrient limitation but to major global change stressors as well, such as higher pCO 2 and temperatures.
How do increased pCO 2 , warming and nutrient limitation interact to modify phytoplankton physiology, ecology and ecosystem impacts? No doubt, the effects are complex and varied. To illustrate this complexity, we focus on how phytoplankton nutrient acquisition may be modified by elevated pCO 2 , warming and higher light availabilities, and what consequences this may have on oceanic ecosystems. Inspired by trait-based approaches in ecology [20][21][22], we propose using traits to understand the combined effects of climate change factors and nutrient limitation on marine phytoplankton. Specifically, we highlight the impacts of climate change on nutrient acquisition at the individual level through phenotypic plasticity, at the population level through genotype-specific responses with potential consequences for evolutionary adaptation, and at the community level through climate-driven species sorting, revealing unexpected scenarios for shifts in community size structure.

Plasticity of nutrient acquisition traits (a) Elevated pCO 2 and warming
Nutrient acquisition in phytoplankton approximates a hyperbolic function, with uptake rates and growth rates steeply increasing at low nutrient concentrations toward saturation when nutrient is in excess [23][24][25][26]. Key nutrient acquisition traits include the maximum uptake rate (V max ) or maximum growth rate (µ max ), and the half-saturation constant (K 1/2 ), which describes the concentration of a nutrient where nutrient uptake or growth equals half of the maximum rate. Nutrient uptake or growth affinity (α) combines both traits, representing the initial slope following V max /K 1/2 and µ max /K 1/2 , respectively (figure 1a) [24,27]. Climate-driven increases in nutrient limitation may thus likely benefit phytoplankton with high nutrient uptake or growth affinities, attained either through plastic responses or through evolutionary selection. In addition to the uptake traits, the minimum nutrient requirement, minimum nutrient quota Q min , is important in determining nutrient competitive abilities, which can be expressed as scaled uptake affinity V max =K 1=2 Q min [28]. In general, smaller-celled species tend to have better competitive abilities [28], so that they would have a competitive advantage in the future more oligotrophic ocean.
Phytoplankton possess high phenotypic plasticity and can strongly modulate their physiology in response to elevated pCO 2 and temperature. Warming may reduce the energetic and elemental costs for overall metabolism, and with elevated pCO 2 the energetic costs of carbon acquisition could be reduced, notably by down-regulation of energy-demanding CCMs [4,5,29]. Consequently, cells may reallocate energy and/or elements to enhance the uptake of a limiting resource, leading to higher nutrient uptake or growth affinities (figure 1b). Indeed, at higher temperatures phytoplankton seem to have higher nitrogen uptake rates (for ammonium and urea, but not nitrate) [30], and higher nutrient growth affinities (figure 2a) [27]. Similarly, elevated pCO 2 also led to higher net nitrogen assimilation rates (i.e. nitrogen quota multiplied by growth rate) in two dinoflagellate species. This was accompanied, however, by a disproportional increase in their K 1/2 for nitrogen, highlighting a potential trade-off between the rate at which nitrogen is assimilated and the relative affinity for nitrogen [31]. Consequently, nitrogen growth affinities (i.e. µ max /K 1/2 ) were at an optimum or decreased with elevated pCO 2 (figure 2b). These findings were mainly explained by a shift toward higher investments in nitrogenrich functional compounds, such as alkaloid toxins and chlorophyll a. Alternatively, the CO 2 -driven down-regulation of CCMs may enhance photo-oxidative stress, leading to increased energetic and elemental costs (e.g. nitrogen) associated to photo-inhibition [36][37][38], which may, in turn, lead to reduced nitrogen growth affinities.
A major source of (bioavailable) nitrogen in the open ocean is the N 2 fixed by diazotrophic cyanobacteria [39] and released into the water column. Thus, changes in N 2 fixation could significantly alter N budgets in oligotrophic oceans. Nitrogen fixation has been shown to be strongly temperature-dependent, with optimum rates in warm, low latitude tropical and subtropical regions (figure 2c) [32,33]. Although warming generally enhances N 2 fixation rates, it may also cause oxygen inhibition of the responsible enzyme nitrogenase, thereby possibly leading to a decline in N 2 fixation rates at high temperatures [33,40]. Nitrogen fixation rates were shown to generally increase at higher pCO 2 , nutrient concentration Figure 1. Conceptual overview of nutrient acquisition traits (a) and putative impacts of climate change (warming and elevated pCO 2 ) on these traits in marine phytoplankton (b). Nutrient acquisition traits include maximum growth (µ max ) or uptake (V max ) rate, half-saturation concentration (K 1/2 ) and the nutrient or growth uptake affinity (α). Brown arrows in (b) indicate potential effects of climatic change (darker shades) on nutrient acquisition traits.
royalsocietypublishing.org/journal/rstb Phil. Trans. R. Soc. B 375: 20190706 from present-day levels of around 400 ppm to approximately 750 ppm [41][42][43], beyond which N 2 fixation rates levelled off (figure 2d). These patterns show that N 2 fixation can be limited by CO 2 , but also that there is a maximum rate at CO 2 levels above 1000 ppm [34]. Although elevated pCO 2 was shown to be beneficial for N 2 fixation, a decrease in pH may possibly inhibit it owing to a decrease in nitrogenase efficiency, resulting in declined growth and N 2 fixation rates [44,45, but see 46]. Besides nitrogen and phosphorus, iron is a major limiting resource for oceanic primary production as well, particularly in the Southern Ocean [19]. Similar to nitrogen and phosphorus, the uptake and assimilation of iron were also shown to be affected by temperature. Specifically, along with enhancing N 2 fixation, warming increased iron use efficiency in a marine diazotroph (Trichodesmium), and this could even offset the effect of iron limitation [47]. At the same time, however, the inhibitory effect of decreasing pH was most apparent under Fe-limiting conditions [44,45]. Whether elevated pCO 2 would promote N 2 fixation may, thus, depend on the availability of Fe, and further work is needed to elucidate the interactive effects of Fe and CO 2 on N 2 fixation in marine diazotrophic cyanobacteria.

(b) Increased light availabilities
Enhanced thermal stratification of the oceanic waters may lead to shallowing of the upper mixed layer, which may, together with sea ice retreat, enhance the relative light availability in the ocean surface layers and thereby stimulate primary production [48,49]. With higher relative light availabilities, the costs of light capture may be reduced and, thus, could allow reallocation of energy and/or elements towards nutrient acquisition. Indeed, higher light levels were shown to enhance N 2 fixation rates in diazotrophic cyanobacteria [33,50]. Moreover, cellular chlorophyll a content in various phytoplankton species decreased with higher light intensities [51], which may reduce nitrogen demand for synthesizing these pigments [52]. Increasing light availability can also directly decrease nutrient demands of phytoplankton by reducing their elemental quota [53], though these responses may vary among species [54]. Higher light availabilities combined with elevated pCO 2 , however, may cause photo-oxidative stress, thereby leading to reduced primary production [37,55, but see 56]. Despite being beneficial to photosynthesis, the impact of enhanced light levels will thus depend on the availability of other resources, and may possibly become detrimental.

Favouring the small…
Climate-driven depletion of nutrients may shift phytoplankton communities towards dominance by species with low nutrient requirements, high nutrient uptake efficiencies, and a high flexibility to shunt excess energy towards royalsocietypublishing.org/journal/rstb Phil. Trans. R. Soc. B 375: 20190706 nutrient acquisition. Being small seems a particularly good strategy to deal with nutrient depletion, as (absolute) nutrient requirements are proportional to size [25,57]. Moreover, because of their high surface-to-volume ratio, small cells have higher growth and nutrient uptake affinities, and are less likely to become diffusion-limited [25,[58][59][60]. Consequently, smaller-sized phytoplankton generally dominate phytoplankton biomass in the open ocean where nutrients are depleted and primary production is low, while largercelled phytoplankton are generally more dominant in more productive coastal waters [61,62].
With climate-driven declines in nutrient availabilities, phytoplankton communities may thus possibly shift towards small-celled species. Indeed, the size of diatom frustules, indicative of diatom cell size, was shown to be inversely correlated with temperature variations over the past approximately 65 Myr (figure 3a). In other words, warmer periods had smaller diatoms dominant, which could have resulted from the reductions in nutrient availability due to enhanced thermal stratification [64]. Also in contemporary marine phytoplankton, cell size usually decreases with temperature (figure 3b) [63]. Similarly, experimental warming led to a shift in the phytoplankton community toward smaller phytoplankton species, which was most prominent under high-nutrient stress (figure 3c) [65]. This is in line with climate change scenarios tested with a global Earth system model, which projected a shift toward smaller phytoplankton species, particularly at higher latitudes [66].

…and the large?
Although nutrient depletion generally favours small phytoplankton, climate-driven reallocation of energy and/or elements may be particularly beneficial for large species. First, large species are more diffusion-limited compared with small species and may thus benefit relatively more from enhanced CO 2 diffusion rates. For example, elevated pCO 2 was shown to favour growth of larger diatoms [67], and to shift phytoplankton communities to larger species [67,68]. Second, large species have relatively high elemental investments in light capturing, because of the lower absorption efficiencies compared with small-celled phytoplankton [69]. Consequently, large species may benefit relatively more from increased light availability caused by shallower mixing layer depths in a warmer ocean, as they can reallocate more resources and energy from light harvesting to nutrient acquisition. Third, large species also tend to be more flexible in size, with a proportionately greater possible reduction compared with smaller species, because smaller-celled species are closer to their minimum structural demands and, therefore, have limited cell size flexibility [70]. Large species may, thus, have a greater ability to reduce cell size and benefit from the associated increases in surface-to-volume ratio. Fourth, some larger phytoplankton taxa possess vacuoles that increase their surface-to-volume ratio, thereby enhancing the effective surface area for nutrient transport [71,72]. Fifth, these vacuoles serve as storage compartments for nutrients, particularly advantageous in fluctuating nutrient conditions [73]. With climate change, storm intensities are predicted to increase [74], which may temporarily enhance nutrient concentrations in the surface waters by mixing with nutrient-rich deeper waters, and was shown, as consequence, to promote primary production and favour large diatoms [75,76]. Lastly, cell size is generally correlated with genome size [77,78], and processes such as adaptive gene loss and genomic streamlining may optimize nutrient acquisition traits in small phytoplankton species, particularly in more stable environments [79]. Conversely, it is conceivable that larger cells may have a greater gene redundancy, leading to more resilient traits [8], which may provide a competitive advantage in dynamic environments. In summary, the higher flexibility of larger phytoplankton species in response to direct and indirect effects of warming and elevated pCO 2 may, at least partially, offset their competitive disadvantage in nutrient acquisition.

Evolution of nutrient acquisition traits
Impacts of climate change on marine phytoplankton will depend not only on their plastic responses, but also on their potential to adapt evolutionarily through selection on standing genetic variation or novel mutations [80,81]. Adaptation to elevated pCO 2 and warming was observed in royalsocietypublishing.org/journal/rstb Phil. Trans. R. Soc. B 375: 20190706 various phytoplankton species across major marine phytoplankton groups [82][83][84][85][86][87][88]. Yet, evolutionary responses to elevated pCO 2 seem to be diverse, and may, furthermore, differ in direction compared with the observed plastic responses of phenotypes [89]. However, evolutionary changes observed in coccolithophores that adapted to elevated pCO 2 were consistent with their plastic responses and, at least partially, offset fitness losses [82,89,90]. Evolutionary adaptation toward elevated pCO 2 was particularly evident under environmental conditions that decreased fitness [89]. It is therefore conceivable that a decline in fitness due to increased nutrient limitation might be compensated by adaptation through increased nutrient uptake affinities. Existing intraspecific genetic and phenotypic diversity of marine phytoplankton populations is substantial and thus provides the basis for adaptation through the selection of best-fit genotypes [91][92][93][94]. With regard to nutrient acquisition, populations of the dinoflagellate Alexandrium ostenfeldii were shown to exhibit a large intraspecific variation in nutrient uptake kinetics, demonstrating a wide range of nitrogen uptake affinities (figure 4) [95]. This suggests a large potential for selection of clones with higher nutrient uptake affinities when nutrients become (more) limiting.

Concluding remarks and future directions
We have described how a trait-based ecological approach may help understand the interactive impacts of climate change factors and nutrient limitation on marine phytoplankton, highlighting possible shifts in nutrient acquisition through elevated pCO 2 , warming or changes in light availabilities. We hypothesize that climate-driven exacerbation of nutrient limitation may be partially counteracted by the concomitant increases in pCO 2 , temperature and relative light availabilities, which may benefit large phytoplankton species capable of reallocating greater resources to nutrient acquisition more than small species and thus, at least partially, offset their competitive disadvantages. This could lead to different outcomes for phytoplankton size distributions, which, in turn, would have different effects on ecosystem processes and food-web dynamics. The next step would be to incorporate energy or resource reallocation in mechanistic models, investigating the magnitude of possible direct and indirect effects of simultaneously acting stressors, and link these to food-web and ecosystem models, thereby generating process-based predictions for oceanic ecosystems.
Obviously, global environmental changes involve a multitude of factors that may affect phytoplankton in diverse ways, maybe differently from what we propose here. By highlighting the complex interplay of several global change stressors on phytoplankton nutrient acquisition, we argue that investigating how multiple stressors may interact to modify phytoplankton traits should be an urgent research priority, requiring collaborations of phytoplankton physiologists, ecologists and modellers. Also, we note that taking into account interacting stressors may yield different predictions compared with when stressors are considered in isolation. For example, recent work showed that nutrient limitation may make phytoplankton more vulnerable to rising temperatures by decreasing their temperature optima and impeding evolutionary adaptation to warming [97,98]. Using a trait-based framework for a better mechanistic understanding of trait flexibility in different phytoplankton size classes under the combined changes in pCO 2 , temperature and resource availabilities, as well as other anticipated environmental change stressors, should further improve our predictions of the future oceanic primary production and ecosystem dynamics. Competing interests. We declare we have no competing interests. Funding. This work was supported by the NSF grant no. OCE-1638958 (to E.L.) and the German Centre for Integrative Biodiversity Research (iDiv; to E.L.).