<P> In 1884, Samuel Pierpont Langley attempted to estimate the solar constant from Mount Whitney in California . By taking readings at different times of day, he tried to correct for effects due to atmospheric absorption . However, the final value he proposed, 2.903 kW / m2, was much too large . </P> <P> Between 1902 and 1957, measurements by Charles Greeley Abbot and others at various high - altitude sites found values between 1.322 and 1.465 kW / m2 . Abbot showed that one of Langley's corrections was erroneously applied . Abbot's results varied between 1.89 and 2.22 calories (1.318 to 1.548 kW / m2), a variation that appeared to be due to the Sun and not the Earth's atmosphere . </P> <P> In 1954 the solar constant was evaluated as 2.00 cal / min / sq cm ± 2% . Current results are about 2.5 percent lower . </P> <P> The actual direct solar irradiance at the top of the atmosphere fluctuates by about 6.9% during a year (from 1.412 kW / m2 in early January to 1.321 kW / m2 in early July) due to the Earth's varying distance from the Sun, and typically by much less than 0.1% from day to day . Thus, for the whole Earth (which has a cross section of 127,400,000 km2), the power is 1.730 × 10 W (or 173,000 terawatts), plus or minus 3.5% (half the approximately 6.9% annual range). The solar constant does not remain constant over long periods of time (see Solar variation), but over a year the solar constant varies much less than the solar irradiance measured at the top of the atmosphere . This is because the solar constant is evaluated at a fixed distance of 1 Astronomical Unit (AU) while the solar irradiance will be affected by the eccentricity of the Earth's orbit . Its distance to the Sun varies annually between 147.1 10 km at aphelion and 152.1 10 km at perihelion . </P>

How is the energy that reaches earth from the sun produced