Cumulus Support

I was wondering if the minimum pressure would be better set at the lowest recorded pressure ever recorded for our area which is fairly low when a hurricane's remnants passes over? Seems like that would throw off the normal, everyday forecast.

I realize the forecast is a general one based on pressure so an exact pressure number is an over specification.

The help section for "pressure extremes" reads "Set the minimum and maximum pressures expected in your part of the world." I assume "expected" is normal minimum low not an extreme. As suggested, I have been diddling with these numbers to see how close Cumulus can hit the mark and it has been as good.

I just installed 1.9 but have not tried the one-hour forecast option.

Randy

I haven't ever studied the code closely enough to understand how it works, but I believe it would make sense to use 'normal' values rather than 'extreme' ones. For example, the 950 and 1050 hPa defaults are intended for the UK; the lowest and highest pressures every recorded in the UK are 926 and 1055 hPa. Having started with some reasonable values, you can then determine what works best for you empirically.

Thanks Steve for the reply. That is certainly logical. It will interesting to see the forecast when the pressure falls below the set minimum.

I went ahead and checked the one hour forecast to give that setting a try.

Randy

As has been stated perhaps too many times, it is beyond the means of any local agency / kit to produce a better than average forecast, indeed even with space based technology, marine automatic and aeroplane / airport reports, surface and subsurface marine temperature and current recorders etc. etc. etc. - NO agency makes more than 40% assurances of even a 12 hour future event.

This (Cumulus forecast) really is just one step above catgut / seaweed / cows lying down, BUT for similar empirical reasons is equally as valid

There seems to be two camps of thought - Those who have got their new toy and why doesn't it tell them there will be 2.1mm of rain in 3 hours time - and those who suddenly realize we haven't got a clue what is going on !!!

Welcome to the machine ..........

beteljuice, ya, I got the picture on this one. Just fun to see how close it gets to being right.

beteljuice wrote:This (Cumulus forecast) really is just one step above catgut / seaweed / cows lying down, BUT for similar epirical reasons is equally as valid

I have decided that the cows aren't as reliable as they used to be so I have tinkered with the strings.ini a bit and am testing a 'percentage' value for the likelihood of rain or fine weather. It seems to be okay especially if updating each hour is chosen. . Expecting showers here tomorrow afternoon so will see what happens.

[Forecast]
Forecast=Forecast
notavailable=Calculating.... Please wait.....
exceptional =Exceptional Weather. 0% chance of precipitation.
forecast1=Settled fine. 0% chance of precipitation.
forecast2=Fine weather. 0% chance of precipitation.
forecast3=Fine weather. 5% chance of precipitation.
forecast4=Fine, becoming less settled. 5% chance of precipitation.
forecast5=Fine, possible shower. 10% chance of precipitation.
forecast6=Fairly fine. 20% chance of precipitation.
forecast7=Fairly fine, possible showers. 30% chance of precipitation.
etc etc etc...............

NOTE:

"Exceptional" is the prefix to 'Fine' OR 'Stormy much precipitation' !

beteljuice wrote:NOTE:

"Exceptional" is the prefix to 'Fine' OR 'Stormy much precipitation' !

I didn't know that. Thanks beteljuice.

Score one for Cumulus forecast. Yesterday, Wunderground and the local news forecast said no chance of rain for the day or even the week. Cumulus forecast "Fairly fine, showery later". I paid no attention to that. Sure enough, later that day my wife called from her Mom's place two miles away to say it was pouring down rain there. I looked at the radar and the only rain in state was at her Mom's place covering an area of maybe a large city block. It sprinkled here at the station's location.

High five!

Got it.

Randy