Early solar output
Early in the Earth's history, the Sun's output would be only 70% as intense during that epoch as it is during the modern epoch. In the current environmental conditions, this solar output would be insufficient to maintain a liquid ocean. Astronomers Carl Sagan and George Mullen pointed out in 1972 that this is contrary to the geologic and paleontological evidence.
According to the Standard Solar Model, stars similar to the Sun should gradually brighten over their main sequence life time. However, with the predicted solar luminosity 4 billion (4 × 109) years ago and with greenhouse gas concentrations the same as are current for the modern Earth, any liquid water exposed to the surface would freeze. However, the geological record shows a continually relatively warm surface in the full early temperature record of the Earth, with the exception of a cold phase about 2.4 billion years ago. Water-related sediments have been found that date to as early as 3.8 billion years ago. Hints of early life forms have been dated from as early as 3.5 billion years, and the basic carbon isotopy is very much in line with what is found today. A regular change between ice ages and warm periods is only to be found since one billion years.
When it first formed, Earth's atmosphere may have contained more greenhouse gases. Carbon dioxide concentrations may have been higher, with estimated partial pressure as large as 1,000 kPa (10 bar), because there was no plant photosynthesis to convert the gas into oxygen. Methane, a very active greenhouse gas which reacts with oxygen to produce carbon dioxide, may have been more prevalent as well, with a mixing ratio of 10−4 parts per million by volume.
Based on a study of geological sulfur isotopes, in 2009 a group of scientists including Yuichiro Ueno from the University of Tokyo proposed that carbonyl sulfide (OCS) was present in the Archean atmosphere. Carbonyl sulfide is an efficient greenhouse gas and the scientists estimate that the additional greenhouse effect would have been sufficient to prevent the Earth from freezing over.
Following the initial accretion of the continents after about 1 billion years, geo-botanist Heinrich Walter and others believe that a non-biological version of the carbon cycle provided a negative temperature feedback. The carbon dioxide in the atmosphere dissolved in liquid water and combined with metal ions derived from silicate weathering to produce carbonates. During ice age periods, this part of the cycle would shut down. Volcanic carbon emissions would then restart a warming cycle due to the greenhouse effect.
According to the Snowball Earth hypothesis, there may have been a number of periods when the Earth's oceans froze over completely. The most recent such period may have been about 630 million years ago. Afterwards, the Cambrian explosion of new multicellular life forms started.
A minority view, propounded by the Israeli-American physicist Nir Shaviv, uses climatological influences of solar wind, combined with a hypothesis of Danish physicist Henrik Svensmark for a cooling effect of cosmic rays, to explain the paradox. According to Shaviv, the early Sun had emitted a stronger solar wind that produced a protective effect against cosmic rays. In that early age, a moderate greenhouse effect comparable to today's would have been sufficient to explain an ice-free Earth.
The temperature minimum around 2.4 billion years goes along with a cosmic ray flux modulation by a variable star formation rate in the Milky Way Galaxy. The reduced solar impact later results into a stronger impact of cosmic ray flux (CRF), which is hypothesized to lead to a relationship with climatological variations.
An alternative model of solar evolution has been proposed as an explanation for the faint young sun paradox. In this model, the early Sun underwent an extended period of higher solar wind output. This caused a mass loss from the Sun on the order of 5−10% over its lifetime, resulting in a more consistent level of solar luminosity. (As the early Sun had more mass, resulting in more energy output than was predicted.) In order to explain the warm conditions in the Archean era, this mass loss must have occurred over an interval of about one billion years. However, records of ion implantation from meteorites and lunar samples show that the elevated rate of solar wind flux only lasted for a period of 0.1 billion years. Observations of the young Sun-like star π1 Ursa Majoris matches this rate of decline in the stellar wind output, suggesting that a higher mass loss rate can not by itself resolve the paradox.