Questions

[daveq]

The all-time records are just that: all time. Each day has a minimum temperature and a maximum temperature. The highest minimum all-time record is the greatest value of the daily minimum recorded at your station, ever, and the lowest maximum all-time record is the lowest value of the daily maximum recorded at your station, ever. And yes, typically the highest minimum would be in summer (during a warm night) and the lowest maximum would be in winter (during a cold day).
Steve

Steve,

I'm seeing the light. One final question. Day is what period of time. 6a - 6p? ... and night 6p - 6a?

Thanks for the time and the explanation.

BTW, I'm liking your program and am hoping to overcome the lack of cron support in order to use Kevin's graphs. I have a satellite internet connection and have limited bandwidth with slow speed compared to other broadband.

--Dave

[daveq]

In essence it's a way of looking for your warmest night (highest min) and coldest day (lowest) max <since records begin>
Long winded but hopefully now as clear as mud!
Dave

Thank you sir! I've got it now. BTW, does this work in 'American'?

--Dave

[daveq]

Also on the rain graph the scale starts at zero but doesn't have another increment. Should there be if the rain is less than an inch for that day?
--Dave

Fixed in latest 1.8.3 build 488. Thanks.

--Dave

[dc1500]

In configuration menu under <station> you can set day to midnight to midnight or 9.00am to 9.00 am. You can also choose units as deg F, inches mph etc.

Dave

[daveq]

In configuration menu under <station> you can set day to midnight to midnight or 9.00am to 9.00 am. You can also choose units as deg F, inches mph etc.

Dave

So that's what Cumulus uses to calculate the highest max and lowest minimum. I have it set to midnight as that is what the other programs use.

--Dave

[steve]

Yes, it simply takes the max or min over the 24 hours, whichever of the midnight or 0900 options you choose. I believe some official meteorological observations take the minimum between 1800 and 0600, and the max between 0600 and 1800 - or something like that.

[kees]

Steve,

On the website of the KNMI (The Royal Dutch Meteorological Institute) there is an article about the
measurement of the max. and min. temperatures.
For climatoligical reports Tmax is measured over the timeframe 0 - 0 hr. UT and for the current
weatherreports (what we see on TV and in the newspapers) over the period 6 - 18 hours UT.
Tmin is measured over the periode 18 - 6 hours UT

[steve]

Thanks - I think that's the same as the UKMO, except they use 09z as the start of the day, not midnight.

[TgT]

Hm

For today, 1st November, meteorological info would look like this:

Temperatures max/min are measured from 20UTC 31.10 till 20UTC 1.11
Three time day observations are at 6,13,20 UTC time and mean temperature is measured as (6UTC+13UTC+(20UTC*2))/4
Rain would go from 6UTC 31.10 till 6UTC 1.11

These are WMO standards.

Anyway, steve, could you add some webtags? min/max/ and time for hum and dew?

[steve]

Anyway, steve, could you add some webtags? min/max/ and time for hum and dew?

I could... but it's not quite as simple as just adding the tags as I don't currently track max and min humidity and dew point, so I'll need to add the code to do the tracking as well. Not difficult, just a bit tedious...

Steve

[steve]

Three time day observations are at 6,13,20 UTC time and mean temperature is measured as (6UTC+13UTC+(20UTC*2))/4

Really? I thought the standard observation times were 00z,06z,12z and 18z.

What you say does ring a bell, though, but I thought the times were one hour later, and were just those used in some countries, rather than being the WMO standard.

Steve

[TgT]

never heard for times you posted, i think those are used as default on numerical model runs like GFS, etc...

[dane]

Rainfall: no scale on graph:

2008-11-02_002115.png

[steve]

Yeah, I know, I spotted it myself earlier. In trying to fix the inches scale, I've screwed it up for millimetres.

Steve

[dane]

OK - no big deal