Robust direct effect of carbon dioxide on tropical circulation and regional precipitation

August 7, 2013

Global warming simulations suggest that wet regions (where precipitation exceeds evaporation) will become wetter and dry regions drier by the end of the 21st century (e.g., Held and Soden 2006), with larger contrasts expected between dry and wet seasons (Chou et al., 2013). This ‘rich-get-richer’ behavior is consistent with a large increase in the moisture content of atmosphere, leading to enhanced horizontal moisture fluxes across regions.

Yet ‘rich-get-richer’ does not apply to all changes in hydrology as climate warms: one counter-example is when considering precipitation changes in the buffer zones separating dry and wet regions, where local changes in the local wind pattern dominate over that of the moisture content (e.g., Boos 2012). Another limitation of ‘rich-get-richer’ is evident when investigating hydrologic changes on shorter timescales (e.g., years to decades) than that required for climate to reach steady state (e.g., centuries). In particular, Bony et al. (2013) demonstrates that following a sudden increase in greenhouse gases (GHG) concentrations, it may take many years before the ’rich-get-richer’ behavior can be observed. Instead, wet regions become drier and dry regions wetter during the first part of the adjustment to a warmer climate.

A rise in moisture content of the atmosphere is driven primarily by a rise in surface temperature, which can take decades to adjust over the oceans due to their large thermal inertia; on the other hand, wind patterns in the atmosphere can respond within days to radiative changes resulting from an increase in GHG concentrations. Changes in hydrology due to changes in moisture content often oppose those due to changes in the wind pattern, and the degree to which ‘rich-get-richer’ becomes noticeable depends on the relative magnitude of both components. Thus, it is perhaps unsurprising that Bony et al. (2013) finds, in a set of simulations in which climate is left to adjust following a massive GHG release, results that are opposite in the first few years of climate change to those obtained in steady state.

In summary, Bony et al. (2013) shows that changes in hydrology following GHG release can have opposite signs on short (first few years) and long (centennial) timescales.


Bony, S. et al., 2013: Robust direct effect of carbon dioxide on tropical circulation and regional precipitation, Nature Geoscience, 6, 447–451. (

Boos, W.R., 2012: Thermodynamic Scaling of the Hydrological Cycle of the Last Glacial Maximum, Journal of Climate, 25, 992-1006. (

Chou, C. et al., 2013: Increase in the range between wet and dry season precipitation, Nature Geoscience, 6, 263–267. (

Held, I.M. and Soden B.J., 2006: Robust Responses of the Hydrological Cycle to Global Warming, Journal of Climate, 19, 5686-5699. (