Improved mountain forecasts

I just looked up Ken Ring's book and I see his prediction for the weather in Wellington last Saturday was "Sunny"

lol, i did say you have to get in sinc with him.

Heh heh. Here's a template that rarely fails if you look at the weather that occurred any day and see how accurate it was. The only catch is that it might be out of sync by up to a day or so either way. It's only fair to give it that kind of slack though, especially given how wide its range is. Apply it to a 30 day month of your choice. Accurate full year almanacs painfully calculated according to my proprietary system are available for an additional fee. 1: Sunny 2: Precipitation 3: Cloudy 4: Sunny 5: Precipitation 6: Cloudy 7: Sunny 8: Precipitation 9: Cloudy 10: Sunny 11: Precipitation 12: Cloudy 13: Sunny 14: Precipitation 15: Cloudy 16: Sunny 17: Precipitation 18: Cloudy 19: Sunny 20: Precipitation 21: Cloudy 22: Sunny 23: Precipitation 24: Cloudy 25: Sunny 26: Precipitation 27: Cloudy 28: Sunny 29: Precipitation 30: Cloudy Yep, 100% accurate within it's bounds. Obviously better than the Met Service, *much* better than Ken Ring. :-P

Interesting that it can be stabbed at over two years out and be more accurate than the met service trying to do it on a day by day basis. Sort of confirms the sceptics point that its predictable. Makes for entertaining Radio and TV when they go head to head.

Well it's not really, I guess. :) The really stupid thing about my system is that it only works backwards and so it's not very useful, kind of like horoscopes. With the out-by-one-day allowance it's just an obscured way of saying that anything could happen every day. You could have three solid weeks of rain, then verify that every single one of those days of rain was predicted by the chart to within a day. What could go completely unnoticed in hindsight is that the same chart predicted that 2/3 of the time was going to be sunny or cloudy without rain, and that never happened, or that for some of those rainy days it was only raining for half the day. So I don't think I'd use it for trip planning. I suppose that at least if forecasters at the met service look over satellite photos and combine the real data with geographic and historic knowledge and experience, then predict that a cold front is expected to hit at some time, chances are that it will actually hit sooner or later. Or with something like the MetVuw, which downloads tens of gigabytes of global weather model data every 6 hours, then applies a New Zealand model to a small number of pixels in the global data to try and extrapolate what's likely to happen over the next few days for its forecast charts. Even MetVuw's only as good as the automated model and it's not perfect, but it's still done from real data and you can be fairly sure something's at least nearby if MetVuw thinks it is. I guess my belief about Ken Ring's methods is that even if he strongly believes that he's onto something (and I'm sure he does), I also think his input data based on the Moon supposedly having a significant effect on weather systems in a way that his theories can predict is just garbage data. Whatever model he applies is unlikely to be very useful, but if people only verify it in hindsight it might seem as if it was. Plus it's possible to score a few extra hits through common sense. If you know in advance that it rains 182 days a year in Milford Sound (which it does according to a random person who edited Wikipedia), you could state 100% rainy days and get a 50% success rate as long as you use your own definition of rain, or probably do much better if your predictions are biased towards the more likely rainy months of January, March, May, October and December (again according to NIWA climate data that a random person pasted into Wikipedia). I'll still sell my charts to anyone who wants to pay me a suitable amount by my own choosing. :-P

Thanks, Militaris for the private wx stations link. I'm sure it will come in useful sometime when I go further afield than the hills here in Canterbury. I'll add it to my article which has the links.

Wow. Radionz has (half) listened to the madpom's winging. Were getting a mountain forcast at a time we can use it to plan the day's tramping: 5am as well as the current 1.30pm. All we now need is the extended forcast (currently too early at 4.30pm when most are still out on the hill) for the rest of nz and I can stop relying on the 5am shipping forcast for a weather outlook when in the hills.

Ive just got back from an over nighter a few hours into the ruahines and just when we come back into cellphone reception my mate's cell got an update from metservice saying that heavy wind and rain warnings were in place and this is by far the nicest day that we have had this week, we cancelled doing an open top tramp because of the warnings in place and its turned out amazing. proves youy cant allways rely on the metservice website etc which changed the forecast that many times coming up to the weekend we werent sure what to do

Yeah that's true, but I do feel sympathy for the forecasters when trying to predict NZ's weather to the kind of exactness typically demanded, then describe it in a way that won't bore and confuse people. They take a lot of flak when people decide they're wrong and rarely get much credit when they're right. When the weather can be so completely different in neighbouring valleys separated by hundreds of metres, it's not clear how it's possible to give an accurate forecast. In the end forecasts come down to the intuition of some experienced people. If it's flip-flopping every hour it probably means that with the information available things could easily get really bad, or not. My partner's wrapping up a PhD which involved lots of weather modelling. The gist of what I've been led to understand, given the amount of data and variables that need to be processed around New Zealand, is that the upcoming weather can sometimes be modelled quite well on some levels, but the processing's only finished several hours after it's happened at the earliest, which makes it sort of pointless. (I'm on second hand info about the specifics of this, though.) Food for thought. :)

I remember Bob McDavitt saying that the only place in the world that it is more difficult to forecast the weather than NZ is the Falkland Islands. One thing that pisses me off is that TV1 is using the weather to keep people watching the news. They come on at the beginning and say "We have a wet weather, wait and see who gets it", then when you get to the actual forecast it is Fiordland who get it, well duh! They seem to be sensationalising some pretty ordinary weather these days. That strikes me as slightly immoral considering how important weather is to us all.