The idea (see ref. 1) is that vertical pipes could be used to pump up nutrient-rich water from deeper in the ocean. Phytoplankton growth in most parts of the ocean is limited by scarcity of nutrients. Supplying more nutrients is likely to stimulate more phytoplankton growth. The phytoplankton will take up corresponding amounts of carbon as they synthesise new cells. The new growth will therefore consume carbon and lead to CO2 entering the ocean from the atmosphere, in order to replace the carbon lost into new algal cells.
You can make a first test of this idea in the carbon model. Click on 'Carbon model' on the home page and then click on 'Model Parameters' on the control panel. Select the 'Physical' tab and then double KSM to 0.10. 'Apply and Close' and then return to the main control panel and run the model. You will see that atmospheric pCO2 actually increases somewhat, rather than decreases. This is because deep water naturally contains more carbon than surface water (look at the DIC plot on the bottom-left of page 1 of the graphs). Therefore the lifting of deep water to the surface (via pipes) elevates the amount of CO2 and prompts initial outgassing (release of CO2 to the atmosphere). This was not considered when formulating the pipes proposal and rather drastically reduces the overall efficacy. Pipes could even, potentially, exacerbate the rise of CO2 in the atmosphere. You can repeat this analysis but make it take place in the fossil fuel era, if you compare two runs, one with fossil fuels and a standard value of KSM, the other with fossil fuels and a doubled value of KSM.
1. J.E. Lovelock & C.G. Rapley (2007) Ocean pipes could help the Earth to cure itself. Nature, 449: 403 (27 September 2007).
2. J.G. Shepherd, D. Iglesias-Rodriguez & A. Yool (2007) Geo-engineering might cause, not cure, problems. Nature 449: 781 (18 October 2007).