3 ms·
I don't think that's a forecast (eclipse is still 20 days away) but rather it's a count of how many days of each type of cloud condition that area has had in th
by margalabargala 3y ago
I don't think that's a forecast (eclipse is still 20 days away) but rather it's a count of how many days of each type of cloud condition that area has had in the past on this date.
- meatmanek 3y agoUsing historical averages is a valid way to forecast things so long as you understand that you're just getting a rough probability distribution.
- margalabargala 3y agoI mean sure, but when people think "weather forecast" they usually think "forecast based on current/recent conditions" not "forecast based on conditions at least one year prior".
- jader201 3y agoThat actually makes me feel better about the data, vs. had it been an attempt to use other models to predict that actual weather, that would have pretty much zero trust in this far out.