Sir

James E. Lovelock and Chris G. Rapley, in their Correspondence 'Ocean pipes could help the Earth to cure itself' (Nature 449, 403; doi:10.1038/449403a 2007) propose a variant on some well-publicized schemes to remove carbon dioxide from the atmosphere, by fertilizing the surface waters of the ocean (see also Nature doi:10.1038/news070924-8 2007). All such schemes suffer from a major problem, because simply enhancing the growth of phytoplankton is not enough. It is the sinking flux of particulate organic carbon into the deep ocean — and ideally into the sediments (usually a small fraction of the total primary production) — that must be enhanced for sequestration to be effective.

Several recent open-ocean experiments have attempted to quantify the level of growth enhancement and sequestration caused by purposeful fertilization: for example, by iron, an essential micro-nutrient. Despite successfully increasing plant biomass, these have not demonstrated sequestration of carbon into the deep ocean (below 1,000 metres), which is essential if it is to be isolated from the atmosphere for centuries or longer. The sinking particles carrying this carbon are degraded rapidly by respiration, and mostly remineralized within the upper ocean. It is likely that almost all the CO2 taken up is released back to the atmosphere within a year. Also, this scheme would bring water with high natural pCO2 levels (associated with the nutrients) back to the surface, potentially causing exhalation of CO2.

We support efforts to find ways of sequestering carbon, but the likely consequences of geo-engineering schemes should be thoroughly researched before they are promoted as solutions. We do not consider ocean fertilization to be a promising approach, and on a large scale it would constitute major interference with an ecosystem which is still poorly understood. Fertilization is likely to alter the phytoplankton community composition and succession, and thus the structure of the oceans' food webs. It might damage these remote and possibly fragile ecosystems, trigger unexpected feedbacks and even reduce their ability to sequester carbon. We cannot, therefore, support this approach, until it can be shown that there would be demonstrable benefits which would outweigh the potential impacts.