by Jef Huisman and Ben Sommeijer
Global warming of the surface layers of the oceans reduces the upward transport of nutrients. Computer simulations predict that plankton growth will become unstable when the supply of nutrients is reduced. This may have a negative impact on the food chains of the oceans and on uptake of the greenhouse gas carbon dioxide into the oceans. Scientists of the Universiteit van Amsterdam and CWI (the Netherlands) and the University of Hawaii (USA) presented their results in Nature of 19 January 2006.
Plankton Processes
Plankton plays a key role in the oceans. It forms the basis of the marine food web. Moreover, phytoplankton (microscopically small algae) consumes the greenhouse gas carbon dioxide during photosynthesis. Because the oceans cover more than 70% of the earth’s surface, marine phytoplankton is quantitatively important for reducing the greenhouse effect on earth. Phytoplankton growth depends on light and on nutrients such as nitrogen and phosphorus. These nutrients are supplied from deeper ocean layers, and are slowly mixed upwards. In large parts of the oceans, phytoplankton is concentrated at about 100 meters depth. Phytoplankton grows well at this depth, because there is a sufficient supply of light from above and a sufficient supply of nutrients from below.
Stratification of Ocean Waters
However, warm surface layers reduce mixing of the ocean waters. This vertical stratification of the water column is widespread in the oceans. A larger temperature difference between two water layers implies less mixing of chemicals between these water layers. Global warming of the surface layers of the oceans, owing to climate change, strengthens the stratification and thereby reduces the upward mixing of nutrients.

To study how reduced upward mixing of nutrients affects the growth of marine plankton, we developed advanced computer simulations. Surprisingly, these simulations predict that plankton populations will show strong oscillations and chaos when vertical mixing of nutrients is reduced (see Figure 1). The model is forced by seasonal changes in the incident light intensity. Different values of the mixing parameter result in essentially different behaviour of the deep chlorophyll maximum (DCM). The top panel of Figure 1 (well mixed situation) shows that the DCM tracks the seasonal variability. The middle panel (moderate mixing) shows double periodicity of DCM locked in a seasonal environment. The lower panel (low mixing) shows a chaotic DCM. In all figures, the left panel shows phytoplankton dynamics; the right panel shows nutrient dynamics.

This model prediction was rather unexpected, because it contradicts conventional wisdom that deep plankton in the oceans would represent a stable system. Therefore, the scientists compared their model predictions with data from longterm time series of plankton in the subtropical Pacific Ocean (see Figure 2), carried out by David Karl of the University of Hawaii. The subtropical Pacific Ocean is strongly stratified, with a low supply of nutrients into the surface layers. Phytoplankton in the subtropical Pacific indeed exhibits complex population fluctuations, consistent with the computer predictions. These results have recently been published in Nature (19 January 2006) in the article “Reduced mixing generates oscillations and chaos in the oceanic deep chlorophyll maximum”.
Mathematical Model and Solution Methods
The new model predictions are based on mathematical simulation of the dynamics of the plankton species and the nutrients in the ocean. The model consists of a set of integropartial differential equations of advectiondiffusionreaction type. The ‘integro’part in the equations originates from a nonlocal integral term describing the penetration of light into the water, subject to absorption of light by photosynthesizing phytoplankton. The numerical solution of the model was based on a finite volume method, with spatial discretisation of the differential operators as well as the integral term. The advection terms were discretised by a socalled thirdorder upwind biased formula, the diffusion terms by a symmetric secondorder formula, and the integral term by the repeated trapezoidal rule. The resulting system of stiff ordinary differential equations was integrated over time by means of an adapted version of the widelyused computer code VODE (http://www.netlib.org/ode/) which is based on an implicit time integration method to cope with the stiffness of the system.
Computational advances increasing the efficiency of numerical solutions of the model were essential to analyze these intriguing fluctuations in the phytoplankton as a result of global warming.
The Netherlands Organization for Scientific Research (NWO), the Dutch BSIK/BRICKS project, the American National Science Foundation (NSF), and the Gordon and Betty Moore Foundation supported the investigations.
Links:
http://www.cwi.nl/projects/pdels/Phytoplankton/
http://www.science.uva.nl/ibed/amb
http://hahana.soest.hawaii.edu/hot/hotdogs
http://www.nature.com/nature
Please contact:
Ben Sommeijer, CWI, The Netherlands
Email: B.P.Sommeijercwi.nl