The implications of increasing average global temperatures for life are difficult to predict with any certainty; even more uncertain is how any response by organisms will be manifest. Microbial marine phytoplankton are the trees and grasses of the ocean, therefore it is vitally important that we understand how a warming ocean may change their physiology to allow prediction of the consequences. Key to achieving this are scientific models that encapsulate how phytoplankton respond to changes in temperature, and therefore enable us to extrapolate the outcomes of such responses to the global ocean. Writing in Nature Climate Change, Andrew Toseland and colleagues1 ask explicitly what happens to these oceanic productivity engines if annual average sea surface temperature is increased by 5 °C.

The world's oceans cover 72% of the planet's surface and harbour microscopic plants and bacteria known as phytoplankton, which are responsible for 98% of the oceans' primary productivity2 and the majority of its geochemical cycles3. Despite considerable investment in understanding how land-based primary productivity may be affected by climate change, we still lack fundamental evidence of how this will be manifest in the oceans. A single group of phytoplankton — diatoms — have been estimated to contribute >25% of global carbon fixation4. However, recent evidence suggests that this carbon sink is being threatened by steadily increasing water temperatures5. To determine the impacts on marine productivity, and therefore nutrient cycling, it is important that we identify how temperature influences phytoplankton physiology. Toseland et al.1 demonstrate this by combining evidence from phytoplankton metatranscriptomes (total community gene transcription) and biochemistry to create a model that encapsulates how phytoplankton alter their cellular chemistry in response to temperature change. The model can then be used to predict any impact on nitrogen (N), phosphorus (P) and carbon cycles.

Previously, researchers have focused on the influence of temperature change on individual components of phytoplankton physiology or on how the ratio of N:P is changing in marine organic matter (for example, refs 6, 7, 8, 9). Toseland et al. shed light on how phytoplankton cells allocate available resources (specifically N and P) so that they can maintain their productivity from the poles to the tropics. The authors employed metatranscriptomics to determine the distribution of phytoplankton and which genes they were transcribing (transcripts) in the Arctic, Antarctic, North Atlantic, North Pacific and equatorial Pacific. They show first that the changes in the relative abundance of transcripts that are destined to make cellular protein (and hence biomass) — in a process called translation — are positively correlated with temperature. Experimental studies in the laboratory show that the translation apparatus of two diatom species worked most efficiently when grown at temperatures close to average equatorial surface waters, and were less efficient at Arctic temperatures as the cold slowed down the molecular machinery. However, Toseland et al. observed that actual cellular productivity in the Arctic and Antarctic was not as repressed as it should be, despite the colder water. They attribute this to a considerable increase in abundance of the cellular translation machinery that helps to build protein, so called ribosomes, which are bound in P-rich RNA (Fig. 1). Hence, to overcome the low water temperatures (average of 2 °C) and concomitant reduction in efficiency, these cells just make more protein factories to maintain their productivity. As this requires more P, the N:P ratio in their cells is reduced.

Figure 1: Temperature-dependent physiology.
figure 1

Phytoplankton have heightened ribosomal efficiency at temperatures experienced around the equator. At colder temperatures they can supplement the reduced efficiency of each ribosome by building more ribosomes. Ribosomes are extremely phosphate rich, shown here in orange (with protein in purple), and hence at colder temperatures a cell needs more phosphorus to make more ribosomes.

This information led to the development of a physiological model of the phytoplankton cell that described how much available P and N the cell would use for creating protein, versus how much it would put into creating RNA. The problem is that RNA uses more P, which is often a limiting nutrient in the world's oceans3; therefore if the cell diverts its resources to create more RNA-laden ribosomes to overcome their reduced efficiency, it needs more P than cells found in warmer water at the equator. The authors placed their model cell in a computer-generated model ocean that replicates the changing temperature, nutrient availability and amount of light that real phytoplankton cells would experience across the global ocean. The model validated the hypothesis that under low temperatures the cells invested more in their cellular machinery to overcome the inefficiency of their factories; whereas under higher temperatures the cells invested in photosynthesis and hence biomass.

In further work they artificially raised the average sea surface temperature by 5 °C, and observed what happened to the phytoplankton cell. As the polar sea warmed up, the phytoplankton cell reduced the production of P-rich ribosomal RNA, changing the cellular N:P ratio, which by definition fundamentally alters this ratio in organic matter. Why does this matter? If the N:P ratio increases then the cell has an increased N requirement, which will cause N to become a limiting resource. Nitrogen limitation could reduce photosynthetic productivity causing an increase in carbon flux from the surface ocean to the atmosphere, thereby resulting in a significant reduction in carbon sequestration by the ocean. Potentially this could result in a catastrophic positive feedback loop, as more atmospheric carbon equals more warming9.

Although this model represents one of the most sophisticated methods for capturing and predicting the result of rising temperature on global oceanic primary productivity, it still has limitations. For example, it doesn't take into consideration the changes in atmospheric carbon levels, which could bolster photosynthetic efficiency and inflate predictions. The model also doesn't account for cyanobacteria, the other major phytoplankton group in the ocean, nor the interactions with other non-photosynthetic bacteria. Future work should focus on the integration of these efforts to create a comprehensive model that will enable us to predict the real outcome of climate change and global warming in this essential system.