Of course we receive less sunlight in winter, because the sun is lower in the sky, so the light has to travel through more atmosphere, and loses strength. Does this mean that the blue light is more scattered out?
Are the frequencies we receive in the winter relatively the same as in summer or is there a change in the received spectrum? Are there less UV-B rays in winter than there are in summer relative to UV-A? If so what is the cause of that?
As you suggest, longer the path, larger the proportion of scattered radiation. Since the shorter wavelengths are more strongly affected by the Rayleigh scattering, sunlight appears more red when the sun is low.
Lower solar elevation angle will also result in longer path through the ozone layer, and hence stronger absorption at the UV-B wavelength range. Together with the Rayleigh scattering, this will result in somewhat less UV-B compared to UV-A, all other things being equal.
The online version of NCAR's TUV radiative transfer model can evaluate the clear sky radiation for a given location and time. At noon, 45° north, I get UV-B/UV-A ratio of about 0.03 for 30th June and 0.01 for 30th January.
In addition to the sun elevation, the radiative fluxes will be affected by a number of factors like clouds, surface albedo (snow!), aerosol extinction, and variations in the ozone layer.
As an addition to the already mentioned effect of the atmosphere for lower solar elevations. As you seem to care about minute differences too. You might want to consider Doppler effect.
The Earth-Sun distance during aphelion and perihelion differs by 5,000,000 km. Therefore, the mean radial velocity of Earth with respect to the Sun is 1,141 km/h or 317 m/s. Something that would lead to a frequency change of about 0.0001% on the solar radiation seen from Earth. This effect would be nil during aphelion and perihelion, and maximum as we approach perihelion. When that would happen in terms of winter/summer will depend on the hemisphere you are at.
This effect might seem negligible, but even much smaller Doppler variations are commonly measured to detect planets around other stars.
As a final note, this seasonal effect will be even smaller than the Doppler shifts due to the rotation of the Earth. Such rotation have associated radial speeds up to ~1,666 km/h.
