Wind Chill Issue

[Peter Fletcher]

We are having very odd weather for the time of year in Western New York State, with peak daytime temperatures nearing 70 ºF. It has also been quite windy. You can see a display of the conditions at my home here.

I am using a Vantage Pro II with the latest version of weewx, and over the last few days the indicated and graphed wind chill temperature has been higher than the indicated heat index temperature, despite winds mostly in the teens (mph). In fact, it looks as if the wind chill temperature is being set to be numerically equal to the uncorrected outside temperature. I now know (having looked it up) that the definition of wind chill requires an air temperature at or below 50 ºF, so that wind chills calculated  by the standard formula for higher temperatures than that are essentially nonsense, but I don't think that substituting the uncorrected outside temperature is better. Would it be possible to display (e.g.) N/A for the wind chill when the temperature is outside the range for which it can be validly calculated, and blank the relevant point/line on the graphs under these conditions?

[Greg Troxel]

"'Peter Fletcher' via weewx-user" <weewx...@googlegroups.com> writes:

> We are having very odd weather for the time of year in Western New York
> State, with peak daytime temperatures nearing 70 ºF. It has also been quite
> windy. You can see a display of the conditions at my home here.

> <https://fletchers-uk.com/weather/index.html>

> I am using a Vantage Pro II with the latest version of weewx, and over the
> last few days the indicated and graphed wind chill temperature has been
> higher than the indicated heat index temperature, despite winds mostly in
> the teens (mph). In fact, it looks as if the wind chill temperature is
> being set to be numerically equal to the uncorrected outside temperature. I
> now know (having looked it up) that the definition of wind chill requires
> an air temperature at or below 50 ºF, so that wind chills calculated by
> the standard formula for higher temperatures than that are essentially
> nonsense, but I don't think that substituting the uncorrected outside
> temperature is better. Would it be possible to display (e.g.) N/A for the
> wind chill when the temperature is outside the range for which it can be
> validly calculated, and blank the relevant point/line on the graphs under
> these conditions?

A few points:

Your wind chill is higher than heat index only because heat index is
lower than temperature. You didn't bring that up, so will assume you
don't find that a problem, but once you accept heat index being lower
than temperature, I don't think you should object to heat index being
lower than wind chill.

Is your weather station calculating wind chill, or weewx? I would
expect this to be from 'hardware' as weewx calls it, from the archive
records from the VP2.

I have a VP2 and am seeing essentially the same thing. My dewpoint is
about 40F.

If the definition of wind chill is that it is equal to temperature
>= 50F, that's who it should be as the standard approach.

I am curious what definition is in use by weewx in your case, and what
the relevant definition is from presumably ukmet.

If you want an option to have some wind-chill-prime definition that
behaves as you want instead, that seems possibly ok, depending on how
many people find that useful vs code size increase/maintenance

[Greg Troxel]

Also, check the values in your skin against what is on your console.

[Peter Fletcher]

Responding to both of your messages: You are right that my comments would equally apply to the heat index being different to the actual temperature. A relatively quick scan of the  relevant pages on the NWS site found the explicit statement that the wind chill was undefined for air temperatures above 50 ºF but I did not notice the slightly less strongly worded caveats about heat index calculations being unreliable below 80 ºF

My console initially showed the same value (60) for Heat index and Wind Chill, with outside temperature 62.0, RH 60, and wind in the low teens. At the same time, the weewx display showed a wind chill of 62.0 and a Heat Index of 60.7. A couple of minutes later, the temperature had dropped to 61 7, the RH had risen to 61, the heat index had risen to 61 (all on the console), with no significant change in the wind. The simultaneous weewx value for heat index was 60.5. It looks, therefore as if weewx is not using the console's calculated values for either derived parameter..

In terms of definitions, at least as per the US NWS, it is not that (wind chill == temperature) above 50 ºF, but that wind chill is undefined above that temperature.They are not quite as clear about heat index, simply saying that "...The Heat Index calculation may produce meaningless results for temperatures and dew points outside of the range depicted on the Heat Index Chart linked below {40<= RH <=100; 80<= ºF <=110)}

All my 'derived value" options in weewx.conf are set up "prefer_hardware', so weewx should presumably be using those transmitted by the console. As far as I can see, it isn't - certainly not for Wind Chill, and probably not for heat index, either!

[Peter Fletcher]

A further note, from Davis's documentation on their wind chill calculations:

"The formula below was adopted by both Environment Canada and the U.S. National Weather Service to ensure a uniform wind chill standard in North America. The formula is supposed to more closely emulate the response of the human body when exposed to conditions of wind and cold than the previous formula did:

35.74 + 0.6215T - 35.75 * (V^0.16 ) + 0.4275T * (V^0.16 )

This relationship takes into account the fact that wind speeds are measured "officially" at 10 meters (33 feet) above the ground, but the human is typically only 5 to 6 feet (2 meters) above the ground..... This newer version of the formula addresses the fact that the latest National Weather Service (NWS) formula was not designed for use above 40°F. The result of the straight NWS implementation was little or no chilling effect at mild temperatures. This updated version provides for reasonable chilling effect at mild temperatures based on the effects determined by Steadman (1979) (see THSW Index section), but as with the new NWS formula, no upper limit where chilling has no additional effect."

On heat index, they say:

" Heat Index uses temperature and relative humidity to determine how hot the air actually “feels.” When humidity is low, the apparent temperature will be lower than the air temperature, since perspiration evaporates rapidly to cool the body. However, when humidity is high (i.e., the air is saturated with water vapor) the apparent temperature “feels” higher than the actual air temperature, because perspiration evaporates more slowly.

Formulas: Heat Index is based upon a lookup table presented by Steadman (1979) and loosely derived from the methodology outlined by Steadman (1998). Thus, air temperatures below 50°F follow this 1998 procedure. Air temperatures above 68°F follow his procedure outlined in 1979 (since the US NWS continues to use this). Davis has made a smooth transition between the two methods between 50°F and 68°F.

[gjr80]

Just to clear up obtaining heatindex and windchill from Davis hardware. Irrespective of any other settings, the only way you will get hardware provided heatindex and windchill from a Davis station is if you have WeeWX use the LOOP2 packet data from the console/logger. This is because heatindex and windchill are not included in the ARCHIVE record and LOOP packet provided by the console/logger. To use LOOP2 data the user must make a change to weewx.conf, a default install will use LOOP packet and ARCHIVE record data from the console/logger and not LOOP2 data.

Based upon what you have posted so far my guess is the WeeWX is not receiving heatindex/windchill data from the console/logger and hence WeeWX is calculating both.

Gary

[Peter Fletcher]

@gjr80: I am sure that you are right. I did the fairly obvious check of forcing both derived parameters to "hardware" in weewx.conf, upon which both show up as "N/A" (and blank on the graph) on the web page. I have a vague memory that there are potential issues with having weewx use LOOP2 packets (otherwise, why not default to this). Can you elucidate?

[gjr80]

There is no real issue as such, it’s just a case of LOOP and LOOP2 providing slightly different data. The main differences are LOOP provides data for extraTemps, extraSoilTemps/Wetness and leafTemp/Wetness as well as a number of rain aggregates and alarms whereas LOOP2 does not, but instead provides a number of more ‘detailed’ obs related to wind and rain (eg 2 minute average wind speed and last 15 minute rain) as well as heatindex, windchill, dew point and THSW as well as various other things. This is by no means an exhaustive list of the differences.

The impetus that caused LOOP2 support to be added to WeeWX was due in part to being able to access THSW. For the average user THSW is probably the only real point of difference driving the use of LOOP2.

Gary

[Greg Troxel]

thanks, those summaries are very helpful!

[Greg Troxel]

gjr80 <gjrod...@gmail.com> writes:

> The impetus that caused LOOP2 support to be added to WeeWX was due in part
> to being able to access THSW. For the average user THSW is probably the
> only real point of difference driving the use of LOOP2.

It would be nice if when one specified loop_request=3 to alternate, then
the software calculation for Observation A would be disabled on loop
packets without A, if a packet with A has been seen recently. That sort
of automates how you have to "prefer hardware".

My other question is if these are in archive records, but I can go look
that up and read the code and/or logs...

[Peter Fletcher]

I have made the necessary changes to my weewx.conf, and everything is now working with the derived values coming from the hardware. FWIW, the current values of outside temperature and RH are respectively 55.5 °F  and 78 %, with the wind having moderated a bit to 5 mph. Wind chill and heat index are both shown as 55.0 °F, which makes more sense to me, in the light of the Davis documentation. Thanks to all for the input!

[Tom Keffer]

Version 4.3 will have a feature to allow users to select between the new NWS algorithm, and their older algorithm. 

--
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 weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/b1d6abe7-ffbf-4635-8985-4f1be215da3en%40googlegroups.com.