Author Topic: Barometer Calibration 201 – How to use Windy.com for better calibration: ISOBARS  (Read 5953 times)

0 Members and 1 Guest are viewing this topic.

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Barometer Calibration 201 – How to use Windy.com (isobars) for better calibration

Calibrating your station barometric pressure to match your local airport is a given that weather station owners know to do. But what happens when you don’t have a local airport? Or what happens if you are far enough away and between two or more airports? How can you properly calibrate your barometer and know which of the two or more airports to use? Being that these airports can show different pressure values at different times or at other times they seem to show the same pressure values, it can feel daunting to properly calibrate your station! Some will suggest you take your console to the airport and calibrate it there and then take it home. What happens when time goes by and you want to re-calibrate or to determine if your station is still working properly? Here is a better method that doesn’t require driving miles with your console to the airport.

Don't drive to the airport:
 [ You are not allowed to view attachments ]

How to use Windy.com to have it display barometric pressure isobars to then use to assist with calibration

Weather pressure systems are classified as Low and High pressure systems. These pressure systems move and change all the time. Wherever there is a Low or High pressure system from its center as you move away the pressure will gradually change. If you were able to look at a bunch of weather stations at the same time you might start to be able to see a trend and draw a picture of these pressure fronts. You often see the TV weatherpersons showing these blue and red lines with these weird looking triangles and half circles to indicate these pressure fronts in action. Well it doesn’t stop there. There is more data to visualize and much more fine detail to be had. A website like Windy.com is able to show these pressure systems at a more significant level using what is called isobars. In meteorology an isobar is a line of connecting locations that all have the same barometric pressure at any given time.

The way to use isobars to assist in calibrating your station is just a matter of choosing the right airport to calibrate with when conditions at your location match the conditions at the airport no matter how far away the airport is. So it is possible to calibrate with a far away airport if you calibrate when you fall in line with that airport, that is to say that the airport and your location both pass along the same isobar line. Differences in elevation do not matter because the comparative pressures are adjusted to Sea Level. It is still good practice to follow barometer calibration 101 recommended best practices – That is to calibrate when conditions are fair and steady and as close to 1013.25 hPa (29.92 inHg) as possible, and when the average temperature is 59 F, and get new METAR reading at the top of the hour. You can still calibrate when it isn’t optimal but you’ll then want to re-calibrate or fine tune when conditions are better.

Okay so now we need to learn how to use Windy.com to show these isobars. Windy.com can at first seem very complicated with many different controls and settings. You can create a free account to save some settings or you can use it without an account.

Follow these steps to get Windy.com to show you; isobars, cities with barometric pressure, and desired units (these steps work for PC browser – steps will be slightly different if using mobile device):
  • Go to https://www.windy.com (log in if you like – optional)
  • Zoom in to your state (not too zoomed in yet)
  • Select More Layers from the top right menu. Do not use More Layers from the bottom menu which gives different options.
[ You are not allowed to view attachments ]
  • Click the On switch for Pressure to add to Quick Menu list, and also select Pressure after turning on Quick Menu option for pressure. (selecting Pressure is what really does it...enabling Quick Menu On Pressure just makes it an easier choice later as it pins it to the top menu visible choices)
[ You are not allowed to view attachments ]
  • Then select the Forecasted weather icon from the bottom menu (2nd icon from the left.) You should now see the cities with pressure values.
[ You are not allowed to view attachments ]
  • To change to desired units of measure click on the bottom bar till you get inHg or hPa. But it turns out that at least for now using inHg has more precision because they don't show any decimals when set to hPa. See follow-up posts in this thread for details.
[ You are not allowed to view attachments ]
  • Lastly to see the isobars on the map click on the Pressure indicator above the bottom menu bars.
[ You are not allowed to view attachments ]
  • Now make a mental note of what value indication the isobars have around your area. You may need to pan around or zoom out or zoom in. Follow the lines and then finally zoom in closer to your location. You can also look at the nearby cities and their reported pressures.
  • Extrapolate the information you see to mentally draw in gradient isobars. Sure it is better to luck out and have one of the isobars pass right through your location and an airport, but you can approximate these lines. In the following illustration You represents your location as an example, CYZE represents your local airport that I've put on the map (I labeled it - but you can make the airport dot show... read next bullet point to see how.) Lastly the thicker pink lines are isobars that I drew for illustration purposes as an extrapolation. Notice the airport and You locations in this example lie on the same isobar.
[ You are not allowed to view attachments ]
  • You can get your airport to show up on the map if you do a search for it on Windy's search bar. All it does is paint a pulsing dot at that location on the map and it does give some METAR info on the left side but I prefer to look up METAR data elsewhere. Like https://aviationweather.gov/metar and be sure to select decoded and then enter and search for you airport. Here is Newark International Airport in New Jersey KEWR.
[ You are not allowed to view attachments ]
  • You can also now click on the map and drag a pin marker around your location to get a barometric pressure reading for that spot. - Thanks Gszlag for the tip!

That is pretty much it. Now you can be better informed of when your pressure should actually match that far away airport. Revisit this a few days at a time if isobars aren’t perfectly aligned. Eventually you’ll be in line with that airport. In the future when your pressure reads different than the airport you calibrated with if you check these isobars you’ll now know why you are different.

* I'm not suggesting you use this method solely to calibrate your station. This is a crude method used just to assist with calibration. You still need to look up the airports METAR data and see what the reported pressure is. The purpose of this method is just to know when your pressure and the airport's pressure should match because you lie in the same isobar.
« Last Edit: April 30, 2019, 10:01:51 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline txweather.org

  • Forecaster
  • *****
  • Posts: 1597
    • Texas Weather
Awesome write up and how too. Thanks!

----
Davis Vantage Pro2 Plus +FARS|Meteobridge Nano SD|Meteohub|Meteobridge MR-3020|WU KTXSPRIN75/PWS JRARGWX75/CWOP EW2972/WBB TXWDVUE75/Blitzortung ID: 1142|AWEKAS: 12095
Donations are welcome: https://paypal.me/ffuentesb

Online Platokidd

  • top side land of Lincoln
  • Forecaster
  • *****
  • Posts: 636
Awesome write up and how too. Thanks!

X2 !
Ambient
1-WS-5000 1-WS-2902A 2-WS40/RAIN 1-WH31L 
1-METEOBRIDGE 1-PM2.5 (WH41B) 3-WH31 1-SRX100LX

ECOWITT
2-HP2550 2-HP2560 2-GW2000 2-GW1100
2-WS68 1-WS80 1-WH32EP 10-WH31 1-WH40
1-HP10 2-WH45 4-WH55 5-WH51
1-WN30 1-WH41

1-DAVIS 7714
1-STRATUS
1-Fisher Barometer 1436R-22
PWS at 2 locations.
1- Storm Sensor-Zelda the dog ;)

Offline gszlag

  • Forecaster
  • *****
  • Posts: 431
  • ..have you calibrated your barometer today?
    • Michael's Bay - Manitoulin Island weather

Many thanks for this as I am still fine tuning. Your tutorial is timely.
Since one should calibrate against a known good source, I am having a problem with CYZE in that it [barometer readings] are off almost 3 hPA according to MADIS. Then again MADIS QC seems too high compared to windy.com.

Other airports up here are few and far between. So currently, I am using windy.com to fiine tune.

Your recommendation to use windy.com was brilliant. It has definitely become my go-to weather website.

There is one trick one can try to fine tune the barometric pressure. Once you click on the three pressure switches you can get the barometric pressure for any location just by clicking on it. A flag pin marker will appear and you can drag it on either side of an isobar until the readings change. In windy.com the isobars at best resolve to 1 hPA. If I am lucky (or patient) sometimes I can see the readings change as I drag the flag pin around. For example if the flagpin reads 1020 hPA and I move it a hair, the reading changes to 1021. Aha! I am either in the high 1020's or at 1021 rather than between say, 1020.5.

All good stuff! Great tutorial!

Barometer Calibration 201 – How to use Windy.com (isobars) for better calibration

Calibrating your station barometric pressure to match your local airport is a given that weather station owners know to do. But what happens when you don’t have a local airport? Or what happens if you are far enough away and between two or more airports? How can you properly calibrate your barometer and know which of the two or more airports to use? Being that these airports can show different pressure values at different times or at other times they seem to show the same pressure values, it can feel daunting to properly calibrate your station! Some will suggest you take your console to the airport and calibrate it there and then take it home. What happens when time goes by and you want to re-calibrate or to determine if your station is still working properly? Here is a better method that doesn’t require driving miles with your console to the airport....

Ambient Weather WS-2000
Ecowitt GW1000/GW1100
Ecowitt WS68: Anemometer, UV/solar
Ecowitt WH40: Rain gauge
Ecowitt WH57 Lightning sensor
Ecowitt WH32E: Outside T & H sensor
Stratus Rain Gauge (manual)
Raspberry Pi 3B+ (WeeWX/CumulusMX)
Raspberry Pi Zero 2W (WeeWX/MQTT/Belchertown)
---
Barometer wiki: http://meshka.eu/Ecowitt/dokuwiki/doku.php?id=barometer#barometer
---
http://weather.glenns.ca (pwsdashboard - live)
http://weewx.glenns.ca
http://glenns.ca/cumulusmx2/index.htm
---
Uploading to: AWN, ecowitt.net, Weather Underground, PWSweather.com, AWEKAS, Windy.com, WOW

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Gszlag,
Thank you inspiring me to write this tutorial. Great tip about the pressure pin markers that you can drop in. I've updated the tutorial to show this.

I do see what you mean about CYZE seemingly not calibrated. You can try reaching out to the airport and finding out who is responsible and then ask about this.
https://weather.gladstonefamily.net/site/CYZE

Lots of people though say not to put much faith in analysis like these. They say to stop chasing lines that you know are wrong. It could be that CYZE is correct but there is no good corroborating nearby information to provide verification to yeild better analysis by MADIS. So what is the next closest airport? Did you try that? Or maybe in your case a trip to the airport maybe warranted to find the mystery. You'd have to go to two airports to compare.

In some cases there are factors that may also increase the discrepancies between the airport and your location. In your case though I do think something is up with CYZE. For others here are some of those reasons: When you compare Windy.com with your airport's METAR you are seeing some discrepancies. METAR data is usually only updated at close to the top of the hour unless there is special broadcast when fast changes are occurring. So depending on when you look at the METAR it might not be timely. Also Windy.com displays Sea Level Pressure and some METAR only show Altimeter Pressure which are slightly different things. Also Windy.com doesn't seem to be as Live as i would prefer. Then there is the lack in precision you mentioned of just displaying whole hPa numbers instead of showing tenths. Factor in all that and that is probably why you can't get it to match with what the airport is reporting. But that should not be the objective.

All of these reasons is why I mentioned to only use Windy.com as a guide and to assist in when you should be attempting to fine tune calibrate and when not to. If the isobars are going through you and the airport then you are good to go...otherwise (like you and the airport are on different sides of an isobar) wait for a better time.
« Last Edit: April 09, 2019, 10:12:10 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
To find another airport or reputable government type station you can use Madis Surface Map:
https://madis-data.ncep.noaa.gov/MadisSurface/

But to not be distracted by other CWOP stations on the map which are not to be trusted as primary sources you need to click on Datasets button on the bottom of the map. Then scroll through the list and uncheck APRSWXNET under Mesonets or maybe uncheck all Mesonets.

Doing this CYSB looks like a possibility, or KDRM, KPZQ, KAPN, or CYVV. Notice that using the map these that I selected are marked with a QC pressure score of S or V, not the ones with a Z or X score for QC. There seems to be a lot of other interesting stations around you with no pressure or bad MADIS pressure analysis, some of these are Maritime stations. CYZE is marked with a QC pressure score of Q...which means questioned.

https://madis.ncep.noaa.gov/madis_sfc_qc_notes.shtml
------------------------------
  MADIS QC Information - Surface
  ------------------------------

  QC Data Descriptor Values
  -------------------------

  No QC available:

   Z - Preliminary, no QC

  Automated QC checks:

   C - Coarse pass, passed level 1
   S - Screened, passed levels 1 and 2
   V - Verified, passed levels 1, 2, and 3
   X - Rejected/erroneous, failed level 1
   Q - Questioned, passed level 1, failed 2 or 3

       where level 1 = validity
             level 2 = internal consistency, temporal consistency,
                       statistical spatial consistency checks
             level 3 = spatial consistency check

« Last Edit: April 09, 2019, 09:46:39 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
You can also add Favorites in Windy.com. These can be a city, a county, a CWOP station, an airport, or other place. Just use the search, enter in some thing relevant, then click on the location results, and then click on the Heart icon and select Add to Favorites.

Favorites show up with yellow text and the corresponding pressure (when proper settings are configured as demonstrated in the first post in this thread).
Notice every place when showing hPa shows up as having the same pressure. See next post for inHg comparison. Impossible to draw isobars with this level of precision.
 [ You are not allowed to view attachments ]
« Last Edit: April 10, 2019, 10:21:05 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Looks like there is more precision with Windy.com if you switch to inHg. This is because they use 2 decimal places with inHg but none with hPa. Had they used just 1 decimal place with hPa it would have been more precise than 2 with inHg. This is rather interesting! Shame...I prefer hPa. Well I can use inHg and then convert.
https://www.convertunits.com/from/hpa/to/inhg

With my Ambient station it is best to calibrate with hPa and then switch to inHg for more precision. So I can still enter in hPa for calibration...I just need to convert first. But I'm not using these numbers anyway as I mentioned they are not timely enough. But this may help pin point more accuracy with the isobars.

Compare this with previous post screenshot that was in hPa. Now every place is not showing the same pressure when switching to inHg.
 [ You are not allowed to view attachments ]

See with that last map you don't even need to zoom out to see the isobars that Windy.com shows. Using this you can draw your own isobars. You can just see it. I drew these not on Windy.com directly but rather on a screenshot graphics/photo program, but you can also mentally visualize these lines by just looking at the numbers...it is where you would draw lines of equal values.
 [ You are not allowed to view attachments ]

Zoom out far enough and sure enough it matches the lines that Windy.com does give you.
 [ You are not allowed to view attachments ]

So right now with me in Winter Springs as you can see on the map...with 3 airports would not be a the best time for me to calibrate, unless I were to just use KORL. I would need to wait till an isobar flows more North-South so that it goes through all the airports and my location for a bit better results. Why not? I'm just fortunate to be in line with 3 airports. But you only need one. So I'd be fine with just one lining up as it is now. But it is a bit off if you really look at it ..as I'm on one side and KORL is on the other side, unless you draw a very thick line. I say wait for better alignment.
« Last Edit: April 29, 2020, 03:18:06 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline gszlag

  • Forecaster
  • *****
  • Posts: 431
  • ..have you calibrated your barometer today?
    • Michael's Bay - Manitoulin Island weather
There is more precision. Great tip!

Looks like there is more precision with Windy.com if you switch to inHg...

Thanks for the MADIS url - lots to explore there and yes, i was wondering what the QC codes were about.

Lots more researching to do. Thanks!

To find another airport or reputable government type station you can use Madis Surface Map:
https://madis-data.ncep.noaa.gov/MadisSurface/
\https://madis.ncep.noaa.gov/madis_sfc_qc_notes.shtml
------------------------------
  MADIS QC Information - Surface
  ------------------------------
  QC Data Descriptor Values
  -------------------------

    Z - Preliminary, no QC

  Automated QC checks:

   C - Coarse pass, passed level 1
   S - Screened, passed levels 1 and 2
   V - Verified, passed levels 1, 2, and 3
   X - Rejected/erroneous, failed level 1
   Q - Questioned, passed level 1, failed 2 or 3

       where level 1 = validity
             level 2 = internal consistency, temporal consistency,
                       statistical spatial consistency checks
             level 3 = spatial consistency check
Ambient Weather WS-2000
Ecowitt GW1000/GW1100
Ecowitt WS68: Anemometer, UV/solar
Ecowitt WH40: Rain gauge
Ecowitt WH57 Lightning sensor
Ecowitt WH32E: Outside T & H sensor
Stratus Rain Gauge (manual)
Raspberry Pi 3B+ (WeeWX/CumulusMX)
Raspberry Pi Zero 2W (WeeWX/MQTT/Belchertown)
---
Barometer wiki: http://meshka.eu/Ecowitt/dokuwiki/doku.php?id=barometer#barometer
---
http://weather.glenns.ca (pwsdashboard - live)
http://weewx.glenns.ca
http://glenns.ca/cumulusmx2/index.htm
---
Uploading to: AWN, ecowitt.net, Weather Underground, PWSweather.com, AWEKAS, Windy.com, WOW

Offline kbellis

  • Forecaster
  • *****
  • Posts: 361
 [ You are not allowed to view attachments ]

Thanks George, good job.

A couple of things worth noting; interpolation, and rounding.

When you have selected the forecast icon as you've outlined above,
 [ You are not allowed to view attachments ]

you're then able to click on a given community for its respective weather forecast.

 [ You are not allowed to view attachments ]

The pressure readings displayed at the geotagged communities' centroid have been interpolated, just like the dragged pin marker (right-click Show weather picker) first mentioned by gszlag. Those interpolations, whether for the community centroid location or the dragged pin marker weather picker, are determined between the locations of the nearest controlling data points. In order to reveal those controlling data points, you'll need to switch from forecast to display wind or temp.

 [ You are not allowed to view attachments ]

Hover your mouse over the displayed wind (or temp) value to reveal information on the control point which is in effect a vertex in an irregular triangular network which forms the basis of the interpolations for the communities, dragged pin markers, etc.

 [ You are not allowed to view attachments ]

Thank goodness windy.com isn't using data from Weather Underground!!, but instead is using vetted data points for their interpolations. Yay windy.com!!

The matter of rounding is something else to keep in mind when the barometric sensor, e.g., for the Fine Offset/ Ambient Weather Osprey has 1) a stated accuracy of ± 0.08 inHg when it's between 27.13 and 32.50 inHg (at unstated temperature); and 2) the resolution of its reading being 0.01 inHg. Is FineOffset rounding at all? Either way, I'm inclined to round the WS-2902A's display to the nearest 0.1 inHg or 1 hPa

In making our comparisons, the question of how is the rounding being performed and from what value, is something else to consider. For example an AWOS using a Vaisala Barocap PTB330 with its stated accuracy of ± 0.10 hPa (Class A, resolution 0.01hPa), or ± 0.20 hPa (Class B, resolution 0.1hPa) when it's between 500 and 1100 hPa at 20 °C, and the METAR /MESO reports pressure to the nearest 0.01 inHg or 0.1 hPa

The differences between these two examples are in one case like making a silk purse from a very fine and respectable sow's ear; and in the other, making a silk purse from silk. But as you've proven, George, adjustments made after an analysis of the data collected over an extended period of time from the Osprey can yield statistically impressive results. Two green thumbs up to you!

One of the handy apps on my phone and tablet: Unit Converter Pro - great for a quick calc
0.01 inHg = 0.338639 hPa
0.1 hPa = 0.002953 inHg
1.0 hPa = 0.02953 inHg

Rounding, one example
29.66 inHg = 1004.40273 hPa <- logically rounds down to 1004
1004 hPa =  29.648107 inHg <- logically rounds up to 29.65

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Thanks Kelly. I've only just begun to see all that Windy.com has to offer.  You took it way to the next level. Love it.

I understand the specifications of our Ambient stations as you pointed out. But when I look at the analysis from MADIS for my CWOP station I'm consistently seeing 0.2 hPa as standard deviation margin of error. That is significantly better than the stated accuracy for our sensor. For this reason I would not be inclined to do any rounding.

I've read in the Windy.com blogs that they surveyed their users and asked if people would be interested in submitting weather data from their stations to them. It would be interesting if they did that. But I hope it doesn't turn into a WU, because you are right they seem to have better data than what is on WU. If you submit data to CWOP then your data is on Windy.com automatically as they pull it from MADIS. The good thing is that I'm sure they are only looking at good data that passes MADIS analysis. So if they started taking data directly then they would need to do some QC.

I'm also a big fan of that unit converter app that you recommended. I have the entire suite of Smart Tools apps. You can buy the tools separately and end up with separate apps...or you can buy the whole suite and then it is all incorporated into one powerful app. It is only $1 more to buy the whole suite.
https://play.google.com/store/apps/details?id=kr.aboy.tools



Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline kbellis

  • Forecaster
  • *****
  • Posts: 361
 [ You are not allowed to view attachments ]

There are additional considerations that ought to be given when viewing pressure data in the very lovely display at windy.com, particularly if you're staring hard at the hundredths of inHg. The interpolation of data is not a simple linear exercise between control data point locations. At lower altitudes the situation is more complex due to landscape (geophysical turbulence). Additionally, there are differences introduced in these interpolated pressure values related to which forecast model is chosen. Here again, consideration for rounding is a good thing.

 [ You are not allowed to view attachments ]
« Last Edit: April 10, 2019, 10:58:06 AM by kbellis »

Offline kbellis

  • Forecaster
  • *****
  • Posts: 361
 [ You are not allowed to view attachments ]

If you submit data to CWOP then your data is on Windy.com automatically as they pull it from MADIS. The good thing is that I'm sure they are only looking at good data that passes MADIS analysis. So if they started taking data directly then they would need to do some QC.

Personally, I'd vote on keeping WU out of the windy.com picture entirely (for too many reasons to list here).

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Kelly,
Okay I now see your point about rounding pressure on Windy.com and how they change depending on model used...etc. But I would say to folks that being that there are so few markers of actual data points (without looking at centroids that get averages) that the averaging shouldn't matter much. The point of this exercise is not to pull actual numbers from the map, rather the point is just to draw isobars. In my example only when I looked at hundredths of inHg was I able to draw isobars. So when I zoomed out the isobars I drew matched the global zoomed out model showing isobars. If there is averaging going on then it is being averaged everywhere.  Sure there may be a discrepancy in accuracy of those hundredths inHg but they would be equally applied for all points on the map, at least to the point where you may not be able to draw a straight line at fine detail (because of how centroids are designated between data points) but you should be able to draw a straight line at a more zoomed out scale. And that is all that matters...to get a feel for if you are in line with an airport or not at any given moment, to let you know when you should and shouldn't be fine tuning your pressure. And you should be getting the actual pressure from more timely METAR at the top of the hour. You can also use Mesowest and get METAR data at quicker intervals than hourly.


Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline FW8379

  • Member
  • *
  • Posts: 44
I was looking at the pressure systems on Windy.com this morning in my neck of the woods and saw that an isobar was most likely running through an airport METAR (KSLC) and my weather station location. I thought this may be a good time to fine tune my calibration - however the weather is changing. I noticed my REL pressure didn't match the METAR on the same isobar by around 4 hPa (I had fine tuned last night), and something else caught my eye: the differences in isobar "direction" between the 3 forecast models NAM, ECMWF, and GFS. Two seem to agree but the third (NAM) shows them nearly perpendicular to the other two. Interesting. I didn't fine tune given the information and conditions.
(Tried inserting screenshots inline by using the "Insert Image" button but it didn't work. Ended up attaching the screenshots).


Offline gszlag

  • Forecaster
  • *****
  • Posts: 431
  • ..have you calibrated your barometer today?
    • Michael's Bay - Manitoulin Island weather
You are right, windy's models don't all appear to agree today. Best to use local METAR pressure that is observed rather than forecasted and to calibrate during stable pressure conditions. Windy.com has three different pressure forecast models ( in Canada and U.S.) and they are often out. I've seen as much as 5 hPa differences. If you click on the airport icon, you can access METAR info but only Altimeter is listed for your  airport.

Here in Canada, METAR usually includes Altimeter and SLP.  I am not sure how the 1012.6 mb is derived in your region but I would use that number as a starting point for your fine-tune.

As per aviationweather.gov:

 Data at: 1328 UTC 25 Oct 2020
METAR for:   KSLC (Salt Lake City Intl, UT, US)
Text:   KSLC 251316Z 34013KT 2SM -SN BR BKN005 BKN023 OVC038 00/00 A2990 RMK AO2 P0000 T00000000 VISNO NW RNWY $
Temperature:   0.0°C ( 32°F)
Dewpoint:   0.0°C ( 32°F) [RH = 100%]
Pressure (altimeter):   29.90 inches Hg (1012.6 mb)
Winds:   from the NNW (340 degrees) at 15 MPH (13 knots; 6.7 m/s)
Visibility:   2.00 sm ( 3.22 km)
Ceiling:   500 feet AGL
Clouds:   broken clouds at 500 feet AGL, broken clouds at 2300 feet AGL, overcast cloud deck at 3800 feet AGL
Weather:   -SN BR (light snow, mist)
QC Flag:   SOME DATA ABOVE MAY BE INACCURATE!!!"$" is an indication the sensor requires maintenance.
Text:   KSLC 251258Z 35013KT 1SM R16L/5000VP6000FT -SN BR BKN005 OVC014 00/M01 A2990 RMK AO2 TWR VIS 1 1/2 P0000 T00001011 VISNO NW RNWY $
Ambient Weather WS-2000
Ecowitt GW1000/GW1100
Ecowitt WS68: Anemometer, UV/solar
Ecowitt WH40: Rain gauge
Ecowitt WH57 Lightning sensor
Ecowitt WH32E: Outside T & H sensor
Stratus Rain Gauge (manual)
Raspberry Pi 3B+ (WeeWX/CumulusMX)
Raspberry Pi Zero 2W (WeeWX/MQTT/Belchertown)
---
Barometer wiki: http://meshka.eu/Ecowitt/dokuwiki/doku.php?id=barometer#barometer
---
http://weather.glenns.ca (pwsdashboard - live)
http://weewx.glenns.ca
http://glenns.ca/cumulusmx2/index.htm
---
Uploading to: AWN, ecowitt.net, Weather Underground, PWSweather.com, AWEKAS, Windy.com, WOW

Offline FW8379

  • Member
  • *
  • Posts: 44
I found it interesting to see the isobars so different (90° angles!) for the same times and locations between the three models. Like galfert has pointed out, Windy.com can be used to determine when to fine tune (when a pressure isobar runs through your station and airport), but not necessarily to adjust to the number shown on Windy.

KSLC METAR (closest airport to me) does actually report both Altimeter and Sea Level Pressures. Sea Level Pressure is only reported at the top of the hour for this particular METAR station and is what I use for fine tune calibration when the conditions are "right". It doesn't seem like the isobars run through both my station and the closest airport METAR (KSLC) very often. Or at least when I look at them.
Quote
I am not sure how the 1012.6 mb is derived
This is an Altimeter pressure for KSLC that gets reported more frequently.
Here is a snippet of what it looks like when both ALT and SLP get reported:

METAR text:    KSLC 251354Z 31011KT 1SM R16L/3500VP6000FT -SN BR SCT006 BKN014 OVC035 00/00 A2992 RMK AO2 SLP124 P0002 T00000000 VISNO NW RNWY $
Conditions at:    KSLC (SALT LAKE CITY , UT, US) observed 1354 UTC 25 October 2020
Temperature:    0.0°C (32°F)
Dewpoint:    0.0°C (32°F) [RH = 100%]
Pressure (altimeter):    29.92 inches Hg (1013.3 mb)
[Sea-level pressure: 1012.4 mb]
Winds:    from the NW (310 degrees) at 13 MPH (11 knots; 5.7 m/s) .......

So if conditions were right for me to fine tune I would fine tune to the 1012.4 mb value in this case.
« Last Edit: October 25, 2020, 10:46:26 AM by FW8379 »

Offline gszlag

  • Forecaster
  • *****
  • Posts: 431
  • ..have you calibrated your barometer today?
    • Michael's Bay - Manitoulin Island weather
Perfect!
Looks like you have figured it out.
In my neck of the woods the SLP is included in all reports ncluding interim reports. Thanks for the heads up about the on-the-hour updates in the U.S.

When the conditions are right - meaning when you are in the same pressure zone as your airport, you can match your reading with METAR.

A high pressure zone that is stable is ideal. If pressure is climbing really rapidy or dropping rapidly you'll have a hard time keeping up - you won't be able to push the buttons fast enough!

PS. windy.com IS useful when you are not close to an airport or in my case, the closest METAR was not calibrated properly for a few months.
Ambient Weather WS-2000
Ecowitt GW1000/GW1100
Ecowitt WS68: Anemometer, UV/solar
Ecowitt WH40: Rain gauge
Ecowitt WH57 Lightning sensor
Ecowitt WH32E: Outside T & H sensor
Stratus Rain Gauge (manual)
Raspberry Pi 3B+ (WeeWX/CumulusMX)
Raspberry Pi Zero 2W (WeeWX/MQTT/Belchertown)
---
Barometer wiki: http://meshka.eu/Ecowitt/dokuwiki/doku.php?id=barometer#barometer
---
http://weather.glenns.ca (pwsdashboard - live)
http://weewx.glenns.ca
http://glenns.ca/cumulusmx2/index.htm
---
Uploading to: AWN, ecowitt.net, Weather Underground, PWSweather.com, AWEKAS, Windy.com, WOW

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
It has been fairly quiet high px last few days at home. Has consistently shown my reading as around 4 to 5  high by extrapolating. Today I am dead on 30.12 bar. I read 30.19.

It will get a resetting when I get back home.

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
I am sitting right on an IsoBar today. Which number do I compare it to?

Thanks

David

 [ You are not allowed to view attachments ]

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
You compare it to the first number 29.72 inHg. The other number in parenthesis is your Station Absolute pressure...which you don't compare to anyone else.
Same thing in the other column if you choose to do it with hPa which are mb or millibars, you look at the first number 1006.3 hPa.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
Cool, thanks for your help.

The reading on Windy was 29.71.

I must be close. Will continue to monitor for now.

Regards

David

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
Running plus or minus a couple hundredths of an inch.

Can’t be too upset about that lol

“New” used station has a sluggish Hum sensor.....easy fix.

Offline davidmc36

  • He who dies with the most toys wins!
  • Forecaster
  • *****
  • Posts: 1246
  • FN25ie61jw
    • MorewoodW34
Today is not the time to compare. Says Rising Rapidly on WFL page.

Today just watch for fun.

 \:D/
« Last Edit: November 18, 2020, 05:19:51 AM by davidmc36 »

Offline gszlag

  • Forecaster
  • *****
  • Posts: 431
  • ..have you calibrated your barometer today?
    • Michael's Bay - Manitoulin Island weather
Today is not the time to compare. Says Rising Rapidly on WFL page.

Today just watch for fun.

 \:D/

Today might be a good day especially this afternoon as high pressure system has rolled in. Time to use METAR. you have Ottawa, Kemptville, Cornwall and others to fine tune. You should be 1032-ish Good luck!
Ambient Weather WS-2000
Ecowitt GW1000/GW1100
Ecowitt WS68: Anemometer, UV/solar
Ecowitt WH40: Rain gauge
Ecowitt WH57 Lightning sensor
Ecowitt WH32E: Outside T & H sensor
Stratus Rain Gauge (manual)
Raspberry Pi 3B+ (WeeWX/CumulusMX)
Raspberry Pi Zero 2W (WeeWX/MQTT/Belchertown)
---
Barometer wiki: http://meshka.eu/Ecowitt/dokuwiki/doku.php?id=barometer#barometer
---
http://weather.glenns.ca (pwsdashboard - live)
http://weewx.glenns.ca
http://glenns.ca/cumulusmx2/index.htm
---
Uploading to: AWN, ecowitt.net, Weather Underground, PWSweather.com, AWEKAS, Windy.com, WOW