I’m new to all things electronic and SparkFun. Just finished building my WIMP wireless weather station from a combination of new (e.g., Weather Shield V12, RedBoard, Sunny Buddy) and retro equipment (Electric Imp Shield I tracked down in Canada). Streaming the data to Wunderground (KCOFORTC617) and everything is working great. Whenever I look at the Wunderground page I feel guilty about that blank UV Index widget on the API and want to add UV functionality to my system. My first novice thought was to add a UV sensor via breakout board, but after discussing it with a non-novice friend a simpler possibility came up — can I measure the solar input on the Sunny Buddy (attached to 9W solar panel) and scale the voltage reading to approximate a UV index? Seems that the necessary parts are in place, I just haven’t been able to find examples of implementation…or why it wouldn’t work. Thanks for humoring me as I dip my toes!
I found this post, that might be helpful. 
While the general principle of “more sun - more UV” like that blog post says is true, I wouldn’t use that to report to Wunderground. The correlation has a lot of noise in it because UV is scattered differently than visual light (where most of the Sun’s power is) by air, moisture, etc. If the Sun is low on the sky, the decrease in UV will be larger than in the visible, etc.
There’s another problem: I don’t think it’ll work because the Sunny Buddy attempts to maximize the power from the solar cell by operating at the max power point, which essentially means “draw enough current that keeps the voltage at the set value”. So if you’re using the Sunny Buddy to charge a battery, the voltage essentially won’t change, only the current.
For my weather station I added a ML8511 UV sensor that Sparkfun used to sell, but it looks like they no longer carry it.
to approximate a UV index?
No, just use a UV sensor.