Minutes of the group meeting held in Cambridge on June 10 2008

Present: Clare, Erik, Róisín, Meike, Sergio, Sophie, James, Marie, Marie-Pierre and Corinne.

General introduction and goals

Corinne presented the main objective of the group:
"to understand and quantify the interactions between marine biogeochemistry and climate.
The group focuses on identifying what are the possible feedbacks between marine ecosystems and climate, not only what is the most likely future. Our approach thus tries to detect hot spots and non-linearities in marine ecosystems.

To achieve this, we need a mixture of model developments and data synthesis and analysis. The group is currently well balanced, with about half of model and half of data specialists.

The current meeting focuses on the developments of PlankTOM5 and PlankTOM10.

PlankTOM5

Erik presented the standard version of PlankTOM5. Since the latest published version of PISCES-P, PlankTOM5 includes ballasting, micro-zooplankton parameters calculated from Sévrine Sailley's database, and river fluxes based on revised observations. The standard model is now ready, and produces reasonable biomass fields.

Sévrine Sailley (through Róisín) presented her database on micro-zooplankton. She showed results from assuming that micro-zoo are entirely ciliate-like or entirely dinoflagelates-like. The resulting ecosystem is very different, and more realistic when ciliates-like rates are used. Sévrine's work also provides preferences for micro-zooplankton grazing. Sévrine also mentioned the nano-zoo database of Joachim Henes, which will be used in the zooplankton comparison.

Corinne showed the results of the climate change experiments. She showed that the ocean is warming by up to 7-8 degrees in the equatorial Pacific and North Atlantic, the ice is melting by 20-30%, the summer MLD is 20-30 m deeper in the Southern Ocean, and the winter MLD is more stratified in the N. Atlantic. The effect on CO2 fluxes is mixed, with the Southern Ocean saturation slowing down, but the Northern Ocean saturation picking up. PP and export increase by 10-15%.

Sergio showed results of ecosystem distribution. All pPFT increase under climate change, with the diatoms increasing most in the Southern Ocean. There is a trade-off between pPFT decrease in the N.E. Pac and increase in the N.W. Pac.

Meike showed that the DMS model was producing reasonable but slightly over-estimated DMS concentration for the control run. Under climate change scenario, the DMS fluxes roughly double, with very high increases in the coastal ocean.

Conclusion: PlankTOM5 is now bug free and looking good. The DMS module will need re-tuning.

PlankTOM10

Corinne stated that PlankTOM10 is now bug free. The PP and export are realistic when using the micro-zooplankton rates of PlankTOM5.0. However some pPFTs go extinct. Róisín's experiments showed that the pPFTs go extinct independently of their grazing pressure if they are not competitive enough. However when they are not extinct, the specific grazing pressure controls the pPFT concentration. The latest problem is that 2 PFTs have T-dependence that is based on a fit to all data, while the rest use the Epley fit which favors the most competitive species.

Róisín showed the results of her sensitivity experiments with growth, mortality, and T-dependence of macro-zooplankton. She showed that the PP and export are little sensitive to macro-zooplankton, but that the other zooplankton respond more. For the moment, only pico-PFTs are surviving, and thus this controls the distribution of the zooplankton. Róisín expects a larger response of the ecosystem when it is more balanced.

Sophie showed the results of her laboratory experiments on growth of diatoms. She showed low standard deviation in the replicates and some variability due to the species. This work is being completed and will be analysed and compared to a database that Sophie is putting together. It will take some months before we can include Sophie's parameters in the model. Sophie will also provide C-cell biomass for the diatoms, which can then be used to convert the database from #/ml to C/ml.

Erik further added that it looks like the Q10 of pPFTs was < Q10 of zoo growth < Q10 of zoo respiration. This has large implications for the response of ecosystems to global warming and needs to be explored in the climate change experiments.

Marie-Pierre presented the results from her data synthesis. She has delivered to Róisín 200,000 macro-zooplankton abundances. She has now gathered 6615 coccolithophorids, 19 trichodesmium, 33,000 diatoms, 500 phaeocystis, 6000 proto-zooplankton (mostly forams). She will then do pico-autotrophs (expected 40,000), and re-visit meso-zooplankton and monera. We discussed the need for cutting out outliers before the mean abundances are calculated. This can be done by computing the median, and accepting all data within the median+/-95% of the standard deviation.

Clare noted that the programme to compare automatically the model results to observations (mean, bias, etc) is moving forward.

Conclusion: PlankTOM10 needs further improvements. The pPFT parameters need to be revised and based closer to observations or optimised. The database is moving along and should provide very valuable constraints on the model results.

Actions

PlankTOM5:

• Clare will update her PlankTOM5 model to include Erik's latest parameters and equations. Then Meike will optimize her DMS code for this version. Clare's code (including the DMS) will then become the standard PlankTOM5 (PlankTOM5.0).
• Clare will run the climate simulations with the forcing fields from ECHAM alone, and with ECHAM+HADLEY. Corinne will look at those. When they are ok, Clare will run again the PlankTOM5 climate change and control simulations (note: revisit the year to use for the control simulation).
• When the climate change simulations are re-run, everyone is welcome to do test experiments. We recommend the experiments to run from 1990 to 2006 (?).
• Clare to add DMS flux, mean surface concentration of DMS, DMSPp, DMSPd in totals.output.

PlankTOM10:

• We need a reasonable PlankTOM10 version and a climate change simulation by mid-September at the latest (deliverable of the Euroceans project). This means that the model should be ready in July, and the climate simulations started in August.
• Erik, Róisín, and Corinne will get together every day to optimize PlankTOM10. Erik will first provide the T-dependence and maxgrowth for the pPFT to Róisín. The three will optimise the code in three steps: (1) ensure the survival of all pPFTs, (2) optimize zPFTs and bacterial growth rates, (3) optimize the zPFT food preferences.
• Meike will have a look at PlankTOM10 rates for phaeocystis and comment.
• Clare will update the DMS on PlankTOM10 (from the PlankTOM5 optimised version). Then Meike or Sergio will have a look at the results, comment if optimisation is feasible, and suggest how to optimise the model if it is.
• Marie-Pierre will pass along databases to Corinne as they get completed. Corinne will look at them and re-format them for use in the model. The C-cell conversions need to be provided by the group.
• Sergio will pass the new DMS database to Clare for the automatic model evaluation.

Others:

• Meike suggested that we do "tracers" presentation during group meetings.

Publication plan:

PlankTOM5:

• Meike: her DMS model (based on the cross)
• Meike and Sergio: DMS in the climate change simulations
• A decision need to be made regarding the publication of the climate change impact on the ecosystem distribution, involving Manfredi, Meike and Sergio (should it be together with the DMS, or should we combine Manfredi's and these results in a new paper?)
• Erik: PlankTOM5
• Róisín or Erik: the zooplankton inter-comparison
• Sévrine: micro-zoo sensitivity to groups in PlankTOM5 (or PlankTOM10)

PlankTOM10:

• Róisín will write up her macro-zooplankton results.
• Meike will produce an outline for PlankTOM10 documentation (by August). She will pass it to Clare with her thesis chapter on the model description. Clare will format it and include all the information she can find from the code, namelist, and web site. Erik will then complete the documents, with the help of others as appropriate.
• Corinne would like to have a publication on PlankTOM10 results as soon as possible. This is to support any new grant applications that want to build on PlankTOM10. We will decide during the summer who will write this publication.

Recommendations for PlankTOM+