Rainfall Rate

[nking]

How long does Cumulus continue to report rainfall rate? I haven’t timed it but it is obviously less than 1 hour

This may have already been requested but, would it be possible to have a webtag to reflect rainfall rates as per the definitions below. I realize there will be some compromise due to the F/O bucket size but it would be a valuable enhancement to the reporting

How fast rain is falling is classified according to the rate or precipitation.
Very light rain precipitation rate is < 0.25 mm/hour
Light rain precipitation rate is between 0.25mm/hour and 1.0mm/hour
Moderate rain precipitation rate is between 1.0 mm/hour and 4.0 mm/hour
Heavy rain recipitation rate is between 4.0 mm/hour and 16.0 mm/hour
Very heavy rain precipitation rate is between 16.0 mm/hour and 50 mm/hour
Extreme rain recipitation rate is > 50.0 mm/hour

[steve]

How long does Cumulus continue to report rainfall rate? I haven’t timed it but it is obviously less than 1 hour

I don't understand the question, but maybe this is the answer anyway: For Fine Offset stations (which annoyingly don't provide a rainfall rate), Cumulus uses a very simple calculation for rainfall rate using the amount of rain that fell in the previous five minutes. For example, 0.3 mm fell in the previous 5 minutes, current rainfall rate = 3.6 mm/hr.

This may have already been requested but, would it be possible to have a webtag to reflect rainfall rates as per the definitions below. I realize there will be some compromise due to the F/O bucket size but it would be a valuable enhancement to the reporting

I could, but this sort of thing comes under a category which I like to call "Things you can easily do for yourself with what is provided already". I'd like to concentrate on facilities in the category "Things you can't do for yourself and I have to write some code in Cumulus".

[daj]

You could easily do this with PHP or Javascript based on the value of the <#rrate> webtag

[nking]

Hi Steve/David,

It was a very badly worded question which you did manage to answer, in part i.e. previous 5 mins. Does this mean that Cumulus will only show the rate for 5 mins if it doesn’t get a second 0.3mm?

Is the logic really as simple as that, if so, this would appear to be a very misleading piece of data? To more accurately (I use this advisedly) answer the rate of rain, wouldn’t you need at least two buckets and elapsed period between each.

Perhaps an alternative for the current “Rate” field would be just a flag set with time the bucket last filled. The rainfall rate is taken care of by the “Last Hour” field if I understand it correctly (which I probably don’t).

I take yours and David’s point that the tagging can be done elsewhere but I wonder if <#rhour> might be better than <#rrate>. The more I think about it the less easy an answer appears and in the scheme of things, does it really matter

[steve]

Originally I just used rain rate = rain in last hour, but I got fed up with people querying this, so I tried to do something about it. If you can be bothered, you can search back in the forum from some time ago and find threads where I tried to implement a more complicated algorithm like Davis use, based on the time between tips. There were problems with this giving silly values under some circumstances and I gave up trying to fix it, and settled for rate = rain in last 5 minutes scaled up. It's hard to get meaningful rain rates from TBRs anyway, so I just left it at that. Even with the Davis method, it returns to zero quite soon after the last tip. Maybe one day I'll look at it again...

[Gina]

Whilst I agree that a rain rate would be nice to have, I agree with Steve that it's very difficult to get any meaningful data from FO and similar weather stations. With the standard FO station and light rain or drizzle you're lucky to get a bucketfull or two an hour. ATM I can't see any better way of showing rain rate than the way Steve has done it in Cumulus. You could calculate the average rate over an hour but we tend to get a lot of short sharp showers in the UK and the average over an hour wouldn't be that helpful. That data is there anyway - users could edit or create XML templates to display it in any way desired.

I have given this some thought during my own software development and not come up with a solution so far. It may get some more attention when I've got other things to my satisfaction.