What We Publish

The role of terrestrial plants in limiting atmospheric CO2 decline over the past 24 million years

Mark Pagani, Ken Caldeira, Robert Berner, and David J. Beerling · Nature 460, 85-88 · 2009

Key finding. Because C3 plant productivity, root biomass, and canopy transpiration all collapse as atmospheric carbon dioxide falls below about 200 parts per million, the biological amplification of silicate-rock weathering shuts down; adding a critical threshold of 150 to 250 parts per million to a carbonate-silicate model reproduces the observed floor in atmospheric carbon dioxide over the past 24 million years.

A line plot of predicted atmospheric carbon dioxide in parts per million, from 0 to 500, against effective weatherable surface area relative to 50 million years ago, from 1 to 5. Six coloured curves correspond to critical carbon dioxide thresholds of 250, 200, 150, 100, 50, and 0 parts per million, each with a matching dashed horizontal line at its threshold. Every curve falls steeply as weatherable area increases and then flattens just above its own threshold, so the higher the threshold the higher the floor at which carbon dioxide settles.
As mountain building exposes more weatherable rock, atmospheric carbon dioxide is drawn down — but only until it approaches the concentration at which plants stop accelerating weathering. Each curve flattens just above its own threshold, which is how a biological limit becomes a floor under atmospheric carbon dioxide. Figure 3 from Pagani et al. (2009), Nature 460, 85-88. Reproduced under author reuse rights. Extracted from the published PDF and resized for web display.

What question did this research address?

Mountain building over the past 24 million years — the Himalaya, the Andes, New Guinea, the Southern Alps of New Zealand — exposed enormous quantities of fresh silicate rock, and warm wet conditions should have made that rock weather quickly. Silicate weathering consumes atmospheric carbon dioxide, so the expectation is a steep long-term drawdown.

Proxy records show no such drawdown. Atmospheric carbon dioxide fell from 1,000 to 1,500 parts per million in the late Eocene to near-modern values by 24 million years ago, and then stopped, never dropping below roughly 200 to 250 parts per million for the rest of the Cenozoic.

Holding carbon dioxide near a floor for more than 20 million years requires either a coincidence in the carbon fluxes too improbable to credit, or a negative feedback that switches on at low concentrations. This paper asked whether terrestrial plants supply that feedback.

What did we find?

Rooted vascular plants and their mycorrhizal fungal partners accelerate chemical weathering by a factor of 1.5 to about 10, through root fracturing of minerals, acidification of soil water by root-respired carbon dioxide and organic acids, chelation by secreted organic ligands, and transpiration that lengthens the residence time of soil water.

That amplification depends on plant health, and C3 photosynthesis is compromised at low carbon dioxide. Rubisco also reacts with oxygen, and photorespiration rises both as the carbon dioxide to oxygen ratio falls and as temperature rises — so the penalty is worst in exactly the warm, low-carbon-dioxide world of the Neogene.

Dynamic global vegetation model simulations run between 50 and 500 parts per million show tropical and global forest primary production and root biomass declining abruptly below about 200 parts per million, with canopy transpiration strongly diminished as well.

Independent evidence puts the ecological compensation point — where net ecosystem exchange reaches zero — between about 100 and 190 parts per million under cool Pleistocene conditions, and higher under the warmth of the early to middle Miocene.

Standard geochemical carbon cycle models assume C3 plants keep promoting silicate weathering even as carbon dioxide approaches zero. Replacing that assumption with a critical threshold of 150 to 250 parts per million damps carbon dioxide variability and yields minimum concentrations matching those observed over the past 24 million years.

The replacement of forests by C3 and later C4 grasslands from the latest Oligocene onward probably weakened the feedback further, since grasses have shallow roots and arbuscular mycorrhizal partners that do not secrete the organic acids driving weathering under ectomycorrhizal forests.

The authors describe the modelling as semi-quantitative. It does not capture feedbacks such as the effect of ecosystem turnover on erosion, soil stability, rock exposure, or albedo.

Why does it matter?

It supplies the missing stabilizer at the low end of the carbon cycle. The temperature dependence of weathering is the textbook thermostat, but it does not explain why carbon dioxide stopped falling during a period of intense uplift; a biological threshold does.

The mechanism is a floor, not a ceiling. It says that plant failure protected the Earth from a severe icehouse over the Neogene, and implies nothing that would limit a rise in carbon dioxide — the feedback only engages near the bottom of the range.

It makes ecological carbon dioxide thresholds a first-order term in long-term carbon-cycle models rather than a detail, since a model that lets plants weather rock down to zero carbon dioxide will predict a drawdown the geologic record contradicts.

Citation

Mark Pagani, Ken Caldeira, Robert Berner, and David J. Beerling (2009). The role of terrestrial plants in limiting atmospheric CO2 decline over the past 24 million years. Nature 460, 85-88.

Related