New research published in the April 2010 issue of Nature by Rosing et al. (introduced by Kasting) casts a new light on the faint young Sun paradox. Previously, scientists invoked extreme concentrations of greenhouse gases in the Archean atmosphere to account for the presence of liquid water on the Earth during a time when the Sun was less bright than it is today. This new study, however, suggests that lower albedo was the primary driver for an ice-free, early Earth.

What is the faint young Sun paradox?

Temperature-wise, liquid water exists on the surface of the Earth within very narrow window, but geological evidence indicates the presence of liquid water spanning 4 billion years. Given that the Sun was 25-30% less luminous 3-4 billion years ago, the Earth would have been frozen over with ice during the Archean under present-day conditions.

The long-standing solution to this paradox is to assume that conditions in the Archean were much different than they are today, and given what we know about the Precambrian world, it’s a reasonable assumption to make. Within this context of a changing Earth, the idea is that higher concentrations of greenhouse gases (i.e. CO2, CH4, and NH4) allowed liquid water to persist during this period of lower insolation.

Problems with the GHG solution

Creating a scenario where increased tectonic activity on the early Earth raised atmospheric CO2 levels and led to warmer temperatures is easy to accept. The problem, however, is that evidence such as Archean paleosols, weathering rinds on fluvial clasts, and evaporites point to lower atmospheric CO2 than is required to keep the early Earth ice-free.

Additionally, the presence of iron minerals like magnetite and siderite within Precambrian banded iron formations places a constraint on atmospheric CO2 to an upper limit of three times the present-day level (i.e. ~900 ppm). Minimally, this assumes a coupling (if not equilibrium) between the atmosphere and ocean in which banded iron formations were deposited. As a consequence, Rosing et al. suggest that greenhouse gas concentrations in the Archean atmosphere were too low to keep liquid water stable at that time. Increased methane as a forcing mechanism is also discounted along similar lines of reasoning.

Planetary albedo as a solution?

The authors assert that in the present day, vegetation and cloud cover limit the albedo difference between land and ocean with respect to the land/ocean difference on the early Earth. We understand that land plants did not appear until Ordovician-Silurian time (approximately 460-440 Ma), so the Precambrian is certainly different when it comes to the influence of vegetation on reducing albedo.

In terms of cloud cover, it is argued that in the present day, increased biotic input from plants and eukaryotic algae creates more cloud condensation nuclei and leads to smaller water droplets in clouds (~12 um). Smaller droplets means more scattering of incoming solar radiation, and thus, higher albedo. During the Precambrian when this biotic influence was absent, the authors suggest that cloud droplets were larger (~20-30 um) on account of fewer nucleation sites, leading to lower albedo. As we understand it, planetary albedo is inversely proportional to the average surface temperature of the Earth.

Given that there is an albedo difference between land and water, continental growth through time clearly exerts a strong control on planetary albedo. As indicated in Figure 1a, continental landmass has increased in time, even if debate over the specifics remains. According to Rosing et al., continent formation was initiated by 4 Ga, experienced rapid growth throughout the Archean and Paleoproterozoic from 3.5 to 1.5 Ga, and leveled off at approximately 1 Ga. In turn, planetary albedo also increased throughout this interval (Fig. 1b.). Figure 1c. accounts for greenhouse gas and cloud influence creating an overall increase in planetary albedo throughout geological time.

Fig. 1. a. Land/ocean ratio through time (i.e. 3.8 Ga to present day) under the assumption of continental growth. b. Estimated surface albedo where increased land leads to an increase in albedo to the present day. c. Average planetary albedo including greenhouse gas and cloud effects, indicating an overall increase. Adapted from Rosing et al. (2010).

All of this comes together in Figure 2, where the authors’ model indicates that albedo forcing at 900 ppm CO2 and CH4 with 20-30 um cloud droplets is effective in ensuring the occurrence of liquid water on the Earth as far back as 3.8 Ga.

Fig. 2. Green: modeled temperature at 900 ppm CO2 and CH4 with droplet size of 20 um (solid) and 30 um (dashed). Blue: temperature at 365 ppm CO2 and CH4 with droplet size 20 um (dashed) and 12 um (present-day; dotted). Adapted from Rosing et al. (2010).

Does the model work?

You decide. It’s clear that significant problems arise when invoking extreme concentrations of greenhouse gases. However, as this paper illustrates, it’s not unreasonable to estimate higher concentrations of greenhouse gases than the present atmospheric level. Because I’m not well-read on the faint young Sun literature, and given that, as Kasting notes, the paradox is almost 40 years old, I wonder how this hypothesis diverges – if at all – from the standing literature. I’m just guessing, but surely this idea has been on the radar before? In any case, as far as I can tell, the authors have provided a pretty elegant response to the faint young Sun paradox.

Is it resolved? I guess we’ll just have to wait and see!


Rosing, M.T., Bird, D.K., Sleep, N.H., and Bjerrum, C.J., 2010. No climate paradox under the faint early Sun. Nature, vol. 464, pg. 744-747.

Kasting, J.F., 2010. Faint young Sun redux. Nature, vol. 464, pg. 687-689.