Help!

I'm looking for someone to help me make more sense of the Atmospheric Opacity data available on The MER Analyst Notebook.

SOLAR_DISTANCE and SOLAR_LONGITUDE - the values listed agree pretty well (+- .01%) with the values I get using the formulas in Telling Time on Mars by Michael Allison.

LOCAL_TIME - I can't establish if this refers to Hybrid time (Mission time) or is adjusted for local true solar time.

AIR_MASS - This is the value I really want to get the formula for. The label notes state "Airmass factor relative to the zenith. It can be approximated as the secant of zenith angle for most observations, but is computed using a 13-km scale height to improve accuracy for low-Sun observations."

The secant of Zenith at the LOCAL_TIME doesn't work (matches OK initially but drifts gradually) so I assume that means LOCAL_TIME is Mission Time but there is a more complex formula being used so it becomes increasingly inaccurate as the Zenith angle rises (e.g. for late afternoon or early morning measurements). I'd like to understand the calculation as it has direct relevance on my attempts to accurately model Insolation for my Solar Power chart.

I've found a number of Airmass formulas (e.g. this one for Mars Odyssey ) but while they are a much better match they still deviate fairly seriously when the Zenith angle is greater than 65deg.

Local time -- local true solar time, definitely.

Airmass--Solar elevation angle isn't reported. That'd be useful. Airmass reflects not just the seasonal difference between hybrid time and true solar time, but the seasonal variation in the max elevation the Sun gets to. Generally, for airmass >60 degrees, the fact that the planet is round and the atmosphere is extended matters. If you go with sec(Z), airmass goes infinite at sunset/rise--for Mars airmass goes to 20-ish, so there's obviously a bit of a correction. I'll try to find a reference to deriving airmass in a spherical atmosphere.

Of course, exp(-tau*airmass) won't be the whole story if you're looking at solar power at the surface. A lot of scattered light gets to the surface at all times of day (even when the Sun isnt up yet--Pathfinder solar panels measured light >2 hours before sunrise).

PS, OK try http://scienceworld.wolfram.com/astronomy/Airmass.html, or Google "airmass spherical atmosphere". Don't assume 13 km is accurate; MER papers have mentioned an 11 km dust scale height.

Yep the local time is LTST - the drift problem was an error elsewhere in my code that was halving my calculated solar declination value. Should have spotted that earlier.

I know that I'll need to model how tau affects the ratio of direct\diffuse radiation in addition to the attenuation of direct radiation via exp(-tau*AM). Not sure how I'm going to do that yet but I have some estimates from documents I've found already.

So with my timing error fixed Secant(ZA) now matches reported AM to within +-1.5% for all angles which will suffice for now as it is no worse than reported Tau error.

Thanks Dilo much appreciated.

Interesting. On Earth, with our thick atmosphere, astronomical twilight (which would seem to be too faint to register on solar panels) starts at 1-1/2 hours before sunrise. So it seems that the Martian twilight and dawn are much brighter. What could cause this brightening? Dust in the high atmosphere? A sunrise/set aurora caused by the solar wind impinging on the upper atmosphere?

Strange place.

--Bill