It's not that I'm harping on this subject, it's only that I have a sustained attention span, and since weather is what I'm working on when I'm not posting, working, eating, sleeping, having sex and so on, I'm going to write a little bit more about weather. I know those gentle readers with ADHD will suffer a little, but brighten up - it looks to be a short post.
A number of people have suggested sources associated with modern, scientific weather prediction models and prognostic climate charts, etcetera, and while I appreciate the interest I have to tell you that anything associated with meteorological prediction models is of no use to me at all. I'd like to explain why.
The purpose of mathematical prediction of the weather as carried forth by scientific meteorologists is to produce a wide-scale model in order to gauge the effect upon a large number of people, or upon crop production, transport, communications and so on. If temperature goes up, how will that affect the creation of hurricanes, say, and how will that affect the eastern seaboard of North America? Can we expect a bad winter if there's increased sunspot activity? Whatever. In any event, like I said, it is of no use to me.
I am not attempting in any way to build a weather model that takes into account the weather for the whole coastal region of, say, Lower Saxony and Denmark, which surrounds the online party. The only thing I need to know is what is the weather like in the fifty foot circle surrounding the party. Beyond that, I don't care. Because of this, the weather chart I make has to be party-centric. I need to know if it is raining, or snowing, or hailing, or sunny. I don't need to know if there's a complicated storm front moving across the whole continent and how that is slamming against the low originating in the Norwegian Sea.
There might be something to be said for devastation caused by a particular storm ... but the facts are that parties in the 17th century move pretty slowly and over a small area in a given length of time, so if the party itself isn't actually devastated by the storm, I don't care if parts of Denmark are. I can assume they'll be fixed by the time the party gets there.
There is a tendency to overthink tables in general. My trade tables, for another example, are designed to do one thing: produce a price at the counter the party needs to pay. They do provide some information about the movement of goods and services in general, but this information - I have to say - gets a little wonky if you look at it too closely. If you try to apply the 'sources' of trade exactly as written, you will discover two towns of the same size right next to each vastly differing in production and services, and therefore real value ... which means the system breaks down if used to try to determine per capita income. I don't worry about that. The system wasn't designed to deliver that kind of information. It is designed to have an axe cost this much in this market, and that much in that market, and that is all. And in this function, it works marvelously.
A few days ago I made reference to the frontages movie studios made of towns for the camera to see when filming the actors walk along a street. For the purpose of the game, I don't need tables that looks behind the buildings any more than the camera does. This is game design, not abstract modelling for the real world, and it helps to remember the purpose of that design.
If you let yourself get bogged down in things that really don't matter, you'll never get anything you try to do done. Which won't matter anyway, because it won't serve a purpose your players will care about.