Sandaysoft

What does Cumulus use to forecast "Precipitation likely"?

Situation -- we're now setting new records for "longest dry period" (since 2004). Each day, Cumulus shows "precipitation likely".

I ask because we have a wild fire close by and I'm wondering if the smoke in the valley is confusing the forecast. Or is it something else that is the predictor for Cumulus.

_________________
Linda

http://weather.fortlawrie.com

It uses a simple calculation based on a device called a "Zambretti Forecaster". Basically, all it uses is the pressure trend and the wind direction.

See http://wiki.sandaysoft.com/a/FAQ#Cumulus_got_the_forecast_wrong.21

Edit: I've just realised you have a Davis station. Are you actually using the Cumulus forecast or the Davis one (Forecast panel in the station settings)?

Edit2: I looked at your web site, and you're using the Davis forecast, so ignore everything I've said. Cumulus is getting the forecast from your station; I don't know what algorithm it uses.

_________________
Steve
Sanday Weather
----------------------------------------------------------------------------------------------------------------------------------
Like Cumulus and want to support it? Please donate!

I'll switch back to Cumulus Forecast. I had it originally and can't remember why I went back to Davis.

Thanks, Steve. I'll let you know if I still see the "precip likely".

_________________
Linda

http://weather.fortlawrie.com

I'd be surprised if you get better results from that. But you never know.

_________________
Steve
Sanday Weather
----------------------------------------------------------------------------------------------------------------------------------
Like Cumulus and want to support it? Please donate!

Yea, it still thinks there might be rain later. We would like that, a LOT! But I don't think it's going to. That is, with the Cumulus/Zabretti forecast, it still thinks it might rain.

I remember why I switched -- the Zabretti verbage (e.g., Settled Fine) -- no one understands.

They understand the Davis verbage better.

_________________
Linda

http://weather.fortlawrie.com

You can change the Cumulus forecasts using strings.ini; but I still think you'll get better results from the Davis forecasts.

_________________
Steve
Sanday Weather
----------------------------------------------------------------------------------------------------------------------------------
Like Cumulus and want to support it? Please donate!

Okay. I'll switch back to Davis.

Interesting that both predicted rain -- just wish they were right!

_________________
Linda

http://weather.fortlawrie.com

The brief research I did on what Davis uses for forecasting suggests that it is "based" on Zambretti.

_________________
Linda

http://weather.fortlawrie.com

Linda, I do not know where you might have gotten that information, but I doubt very much that Davis Instruments is/has been using the Zambretti algorithm to produce their weather forecast - it is based on the barometric reading and trend, wind speed and direction, rainfall, temperature, humidity, latitude and longitude, and time of year - it contain some 200 different forecast rules (document included).

I've been looking for something like that for ages, but there's no way there's over 200 different ones.



That is in the list over 40 times.
And other ones are repeated numerous times too.

78

_________________
Mark
Wilmslow Astro Weather

I cannot believe Mark that you had so much time available that you counted the number on dissimilar lines...

It is my understanding that those lines (they can be read from the internal Vantage Pro2 console) are part of a 200 elements array composed of maybe 5-10 groups, thus similar lines used for different forecasts.

A cut'n'paste into Excel - took about 30 seconds

_________________
Mark
Wilmslow Astro Weather