[wxqc] Davis rain gauge adjustment

Paule paulecwop at gentlevi.router5.com
Mon Oct 3 19:56:22 CDT 2011


Ted, thanks for the more detailed comment re analog issues
with 4".  I also weigh the collected rain on a triple beam,
which will give me a value to the thousandths of inch.
Davis says, error is +/- so many per cent, or 0.01" so
there seems to be some "expectation" that even the best obs
will be +/- 0.01  :)
We had a nice rainfall other day, 0.74" in both my 4" and
with my 8" funnel, and Davis reported a respectable 0.73"
I reported 0.74" to cocorahs.
paule DW8010

but as to your comments below, sure those are all sources of
error.

On Tue, September 27, 2011 19:52, Ted Lum wrote:
> Paying particular attention to "4 Where do errors and uncertainties
> come
> from?" in particular "Operator skill" as it might pertain to "Parallax
> Errors" for example. How does the operator insure that the gauge is
> held
> level and that the eye is level with the fluid? How does the operator
> account for variation due to meniscus? How does the operator choose to
> reflect the analog reading obtained... something like 0.25 and 5/8 of
> a
> hundredth... does that round up to 0.26... down to 0.25... expressed
> as
> 0.255? Exactly how repeatable is the operators technique and
> judgement?
> Humans just don't make the best quantizers.
>
> The 4" gauge has additional challenges if the catch exceeds 1" because
> you then have to take multiple measurement for each 1" increment and
> deal with transferring the fluid from one cylinder to another without
> any spillage or "left behind" fluid.
>
> The digital gauge is not without its flaws. However, one things that
> it
> is that humans are not is repeatable with characteristic curves.
> Tipping
> buckets are known to have rate related errors because the physical act
> of tipping fails to collect and account for some fluid because tipping
> takes time, its not instantaneous. However, their response to various
> rates can be characterized in order to develop rate dependent
> correction
> curves that can be applied to correct readings because while they have
> errors the errors are repeatable, can be characterized, and as a
> result
> can be corrected.
>
> Of course, a lot of this is theoretical because I'm not aware of any
> software that actually tries nullify rate errors, so while possible I
> don't believe that functionality is commercial available. At the same
> time, as long as you have the raw data you could produce corrected
> reading yourself rather easily in Excel or equiv. So, frankly, is see
> greater uncertainty related to operator based quantization errors.
>
> The 4" is not necessarily more accurate. It is 4 times less sensitive
> and is taking a statically smaller sampling... if you've taken
> statistics then you're aware of the effect of sample size on means and
> averages. Additionally, its aerodynamically inferior. In wind
> conditions, which are known to significantly affect performance, the
> size and shape of the 8" gauge performs better, though both are going
> to
> have errors.
>
>
> On 9/27/2011 4:32 PM, Paule wrote:
>> Ted, thanks for the link to the pdf re "uncertainty"
>> skimmed it, the sections on uncertainty of measurement.
>> I've had several stats courses so I'm somewhat familiar
>> with confidence level, etc.  In the few pages I read I
>> saw no mention of uncertainty in reference to analog.
>> Perhaps my question was not clearly enough stated.
>> why would use of a 4" analog cocorahs collector increase
>> uncertainty compared to a digital collector.
>> That's how I understood the context of your comment in
>> comparison to Davis.  Davis folks have told me directly
>> that when Davis differs from an analog like 4" collector,
>> that user should rely on 4" analog.  There is some implication
>> that 4" is more accurate with "better" uncertainty.  Perhaps
>> I'm linguisticly rusty.  Perhaps by increased uncertainty,
>> you mean (meant) better uncertainty.
>>
>> On Sat, September 24, 2011 21:06, Ted Lum wrote:
>>> http://www.wmo.int/pages/prog/gcos/documents/gruanmanuals/UK_NPL/mgpg11.pdf
>>>
>>> On 9/24/2011 7:14 PM, Paule wrote:
>>>> Ted, not sure what you mean when you said "4" is an analog
>>>> device which increases the uncertainty"
>>>>
>>>> Not sure what you mean by "uncertainty" in this context?
>>>>
>>>>
>>>> ***Ted_wrote***
>>>>> (and BTW the 4" gauge is an analog device
>>>>> which increases uncertainty).
>>>
>
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>
> _______________________________________________
> wxqc mailing list
> Post messages to wxqc at lists.gladstonefamily.net
> To unsubcribe or change delivery options, please go to:
> http://server.gladstonefamily.net/mailman/listinfo/wxqc
> To search the archives:
> http://www.google.com/coop/cse?cx=008314629403309390388%3Aknlfnptih9u
>
> The contents of this message are the responsibility of the author.
>




More information about the wxqc mailing list