After being alerted by a user (thanks Reko!), I’ve found a problem in the UV index forecasts, mainly when using wret to generate them (perhaps output to a lastret file). This program was accidentally using older algorithms that had UV index too high. These were not agreeing with the better, newer algorithm in WXSIM itself (I also found that the display on the data entry form was using the old algorithm, though the WXSIM output was not).
While fixing this, I decided to take a new, close look at the “new” algorithm, in light of any additional data I could find. I found somewhat conflicting data, but a consensus weighted towards what seem to be the most authoritative data (some from TEMIS) suggested even the “new” algorithm had values too high, so I am altering the algorithm to decrease values by between 5 and 10%, depending somewhat on the situation (I belive I’ve improved the elevation model, too, by basing it more on surface pressure).
These changes will be in the new version.
Tom