[Bldg-sim] How reliable is weather data from web sources such as the Weather Underground?
Joe Huang
yjhuang at whiteboxtechnologies.com
Sun Mar 4 04:52:33 PST 2018
Since I've been recommending for years web sources such as the Weather Underground for
those wanting site-specific weather data , it behooves me to know how reliable are such
data? It's difficult to answer that question in the absolute sense because there are no
incontrovertible sources against which to compare. However, I did the next best thing,
which is to compare the data against that from a nearby major weather station and then
look to see whether the differences make sense according to general understanding of
terrain factors and urban microclimate variations.
This comparison was not done in a vacuum, but in response to a client who wanted weather
data for a site in Portland Oregon, with a particular interest on the amount of rainfall.
The map below shows the site of interest (red teardrop) and the NSW station at the
Portland Airport 4.5 miles away. The Weather Underground station PORTL437 is located
about 0.5 miles to the east of the site equidistant on the other side of the MLKing Blvd.
I'll let the plots in my reply to the client speak for themselves, but my general feeling
is that the WU data looked quite consistent with the airport data, and that the
differences are quite understandable, e.g., the reduced wind speeds and the lack of
correlation in the wind direction, since the airport station is in a flat open area next
to the Columbia River, while the WU station is in a urban residential area 4-5 miles away.
I'm still withholding judgement whether the 60% difference in the recorded rainfall is
real, a result of the placement of the WU station, or a defect in the measurement
protocol. Come to think of it, I don't know how these automated weather stations record
rainfall (liquid precipitation to be technically correct). I can only think of a rain
gauge that's manually recorded and maintained.
Joe
-------- Forwarded Message --------
Subject: Re: Weather data for Portland with monthly updates
Date: Fri, 23 Feb 2018 16:27:07 -0800
From: Joe Huang <yjhuang at whiteboxtechnologies.com>
To: XXXX
XXXX,
Attached are both weather files for the ISD station at the Portland Airport
(OR_PORTLAND-IAP_726980_18) and the Weather Underground site KORPORTL437
(OR_WU_PORTL437_18). The data are available through Feb. 22 for the ISD station and
through Feb. 23 for the WU station.
Out of curiosity, I compared the data between the two files (see below). The dry-bulb and
dewpoint temperatures are very well correlated, so that I find the hourly variations to be
credible and reflecting microclimate differences. The wind speeds are about three times
higher at the airport, which is also believable, while there is no correlation in the wind
direction. The rainfall amount also differs, with the airport reporting 60 % more than
at the PORTL437 Weather Underground station.
Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data (o)
(925)388-0265 (c) (510)928-2683 "building energy simulations at your fingertips"
On 2/19/2018 12:39 PM, XXXX wrote:
> Joe,
>
> Yeah, the precipitation data as well as relative humidity is crucial to my study so
> that's why I am a bit concerned. If you are willing to separate out ISD and WU that
> would be great because I could compare the two and see if the problem is just isolated
> to precipitation. It did seem like weather underground's RH and temp. data were what I
> expected so I was planning on sticking with that. I would like to see how weather
> underground's wind data compares to ISD because driving rain has some importance me.
> Thanks for the info on the NOAA data.
>
> XXXX
>
> On Mon, Feb 19, 2018 at 12:41 PM, Joe Huang <yjhuang at whiteboxtechnologies.com
> <mailto:yjhuang at whiteboxtechnologies.com>> wrote:
>
> XXXX,
>
> OK. Will do. However, I wouldn't go as far as you in dismissing the WU data simply
> because the rain measurements differ from other sources. For one, I'm not sure what
> mechanisms are used in the automatic weather stations such as the WU stations to
> measure rainfall. It could be that the measurement method is unreliable, or it
> could mean that there is that amount of difference due to other factors - shielding,
> microclimate variations, what not. Are the rainfall measurements of primary
> importance to you? I could give you both ( ISD and WU) if you're interested, no
> difference in price. Also, I don't think of the ISD data as NOAA data, since the
> weather station is a NWS station and the only way NOAA is involved is that the
> NCDC/NCEI that maintains the ISD has now moved from the Dept. of Commerce to NOAA.
>
> Joe
>
> Joe Huang
> White Box Technologies, Inc.
> 346 Rheem Blvd., Suite 205A
> Moraga CA 94556
> yjhuang at whiteboxtechnologies.com <mailto:yjhuang at whiteboxtechnologies.com>
> http://weather.whiteboxtechnologies.com <http://weather.whiteboxtechnologies.com> for simulation-ready weather data
> (o)(925)388-0265 <tel:%28925%29%20388-0265>
> (c)(510)928-2683 <tel:%28510%29%20928-2683>
> "building energy simulations at your fingertips"
>
> On 2/19/2018 8:23 AM, XXXX wrote:
>> Hi Joe,
>>
>> Can you send me another monthly weather update for OR_WU_PORTL437_17 (All of 2017 +
>> January 2018)? I would actually like to switch to NOAA data for this weather file
>> because of the precipitation discrepancy we had discussed earlier. I had originally
>> chosen weather underground because I thought it would give a bit better resolution
>> for the Portland area but it seems maybe there is some error at this location.
>>
>> Thanks,
>>
>> XXXX
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20180304/80caec49/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: map.jpg
Type: image/jpeg
Size: 131342 bytes
Desc: not available
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20180304/80caec49/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Slide1.JPG
Type: image/jpeg
Size: 114105 bytes
Desc: not available
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20180304/80caec49/attachment.jpe>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Slide2.JPG
Type: image/jpeg
Size: 71851 bytes
Desc: not available
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20180304/80caec49/attachment-0001.jpe>
More information about the Bldg-sim
mailing list