A thought, hopefully not misplaced.

As regards not messing with data, it's the correct approach; but I believe
there's wriggle room in the case of rain.

If it's a momentary value such as temperature, pressure, humidity. They are
absolute values that are cemented to that moment in time.

If it's a value such as rain, while it's still married to the time it was
recorded, it's not cemented.
It's origin is in the past - a bucket starts filling well before it tips,
and depending on the size of the bucket, the time it takes to tip will vary
(eg: Two gauges at the same location, a 0.5mm bucket will tip well after a
0.25mm bucket, yet the rain rate hasn't changed. Which gauge is wrong?)

In short: There are already assumptions (compromises) being made about
recording a tipping buckets value. For the sake of a short time interval (2
secs) my feeling is that it could be adjusted back to midnight without
disturbing its integrity - too much. :-)

That's ignoring values like rain rate and loop interval, let alone archive
interval.




Cheers
 Glenn

rorpi - read only raspberry pi + weewx: now with scripts
<https://github.com/glennmckechnie/rorpi-raspberrypi>

On 31 May 2017 at 10:16, gjr80 <[email protected]> wrote:

> But, say we have a packet that arrives at 00:00:02. How do we know if it's
>> from the previous day? Maybe it is the very first packet of the day?
>
>
> True, maybe we should just not use dayRain (or any other loop aggregates
> for that matter), in fact until I read through this issue properly I
> thought that day (or longer) aggregates coming from a station were not used
> at all.
>
> Gary
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "weewx-user" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to