Author Topic: Barometric Pressure question  (Read 3165 times)

0 Members and 1 Guest are viewing this topic.

Offline A1cntrler

  • Senior Member
  • **
  • Posts: 98
Barometric Pressure question
« on: July 01, 2015, 02:23:14 PM »
I've got a 5-in-1 system.  I upload my data via a Meteobridge.  I've been chasing this pressue reporting issue for a while now and it is driving me crazy.  I'm sure I just don't fully understand how it is reported and what is reported. 

I recently had an issue where I had to reload my Meteobridge onto the device and lost my settings that were in it.  I think that my settings were fine before (about a couple weeks ago) as I routinely got two thumbs up and the green check on my Quality Reports.  Since the reload of the meteobridge I've had poor ratings on barometric pressure.  I went as far as to even match the two closest reporting stations to me (withing about a 1/2 mile radius) by adjusting the Pressure Correction setting inside Meteobridge.  One station to the West of me was reporting 29.93, and one due east of me was reporting 29.95.  I made the adjustments to show that Weather Underground  from my station was showing 29.94.  The next day I got the Quality Report and it showed that I had pressure errors of as high as +19.68 and a low of -2.84.  I let it go for a couple days and I still keep getting the errors.  This was with a correction on Meteobridge of 1.62, which brought it to the 29.94.  I lowered the correction to 1.00 and still got the errors.  Today I set it to a correction of 0 just to see what tomorrow brings.  With ti at 0 my station is now reporting to Weather Underground a pressure of 28.31, while the other 2 are at 29.93 and 29.96. 

What am I doing wrong here?  Altitude is set correctly.  I'm between the two stations I'm comparing myself to in altitude, with a difference of 188 ft between the one to the West to the one to the East of me.  Only the station the the West of me shows up in the CWOP nearby stations report (I think it is at the firehouse in town here looking at the marker on the map)

Edit:  Just checked NWS pressure readings from around my area and they are all in line with the 29.92-29.95.  Mine is reporting 28.31 which is too low.  If I add the correction to bring it inline with the NWS data then I get errors from CWOP.  Do I just need to add the correction in and let it ride for a week or so and hope the numbers balance out?  Is it possible that CWOP isn't getting the corrected version of the data that is being reported to Weather Underground from my Meteobridge?
« Last Edit: July 01, 2015, 02:34:17 PM by A1cntrler »
Ambient WS-5000 (KVACHEST93)

Offline Josiah

  • Add-InWx Software Author
  • Forecaster
  • *****
  • Posts: 449
    • Add-InWx
Re: Barometric Pressure question
« Reply #1 on: July 01, 2015, 05:38:15 PM »
I know you're using Meteobridge and not Cumulus, but the general concept regarding the different Pressure Correction formulas that CWOP and WUnderground want the Pressure put thru is the same.
Let me point you to a thread (on another forum site) that may be helpful: http://sandaysoft.com/forum/viewtopic.php?f=6&t=11262
Quote from Steve on that thread:
Quote
"CWOP require altimeter pressure rather than MSLP, so Cumulus calculates (an approximation to) that using information it gets from the console."

Offline A1cntrler

  • Senior Member
  • **
  • Posts: 98
Re: Barometric Pressure question
« Reply #2 on: July 01, 2015, 08:05:21 PM »
So if I'm reading the post correctly, it looks like CWOP uses the actual pressure (Labeled Indoor Press on my Meteobridge) where CWOP then calculates it based off the station altitude vs Sea Level.   

I've been playing with my numbers and trying things with the pressure correction and it seems that I've got it pretty close to the analysis line on the  Gladstone site.  I'm maybe 2mb higher than the analysis, but with it there the pressures look fairly reasonable on WU as well.  I'm going to let it ride for a few days and see how the errors look, and maybe tweek it a bit further. 

Overall I'm pretty happy with my station, but the QC checks drive me crazy.  If I continue to have problems with it I'll just stop the CWOP all together.  I've always been the type of person to like things to work as they should, and having CWOP telling me I'm consistently wrong gives me headaches.  I don't want to for some reason be that guy that is throwing off the numbers for everyone else.
Ambient WS-5000 (KVACHEST93)

Offline Josiah

  • Add-InWx Software Author
  • Forecaster
  • *****
  • Posts: 449
    • Add-InWx
Re: Barometric Pressure question
« Reply #3 on: July 01, 2015, 08:36:36 PM »
So if I'm reading the post correctly, it looks like CWOP uses the actual pressure (Labeled Indoor Press on my Meteobridge) where CWOP then calculates it based off the station altitude vs Sea Level.
Yeah, that pretty much sums it up nicely.

One thing to keep in mind is that it may take a few weeks and possibly even a whole month for the CWOP Red "X" to change to a check mark. The reason for that is its based on the "Error Standard Deviation" which is based on data for the past 30 days. So what I would do is look at how close the 2 graph lines (I'm referring to the Station Pressure and Analysis Pressure) on the Pressure graph are following each-other.
Another thing to keep in mind is that you, like many others, may be in a Micro-Climate and the Measured pressure may not follow the Analysis Pressure very closely.

Offline A1cntrler

  • Senior Member
  • **
  • Posts: 98
Re: Barometric Pressure question
« Reply #4 on: July 01, 2015, 08:46:51 PM »

Another thing to keep in mind is that you, like many others, may be in a Micro-Climate and the Measured pressure may not follow the Analysis Pressure very closely.

Micro Climate is an understatement for here.  I lived in VA for over 30 years and moved here to California 5 years ago.  Still not used to all the different micro climates we have around here.  The weather on TV news at night shows the 7 day forecast, and there are 7 different 7 days they have to show.  It can be in the 60's at the coast, and you moved 20 miles inland and you're in triple digits.  It can be 70 here and a Winter Storm Warning issued for a point 10 miles to the north for heavy snow.   

The stations I was trying to calibrate myself to were withing 1/2 miles of me and less than 200' spread between the three of us in elevation. 

Thanks for your help and helping me to understand more about the readings and reporting of them. 

I can see the blue line (observed at my station) and compare it to the red line (The comparison) and I'm now running slightly above that line but they're pretty much mirroring each others with the rise and fall of pressure.
Ambient WS-5000 (KVACHEST93)

Offline Josiah

  • Add-InWx Software Author
  • Forecaster
  • *****
  • Posts: 449
    • Add-InWx
Re: Barometric Pressure question
« Reply #5 on: July 01, 2015, 09:22:18 PM »
Oh cool, multiple stations all within 1/2 mile, that's really nice for calibration references.

Well, Now that you've got the pressure mirroring the rising and and falling, and you now understand the reporting differences, it should be pretty simple from here on out.

One tip i read, (not sure where, it may have been on this forum even) when calibrating the pressure, to avoid over compensating don't make frequent large changes. Instead, make smaller ones and wait awhile before making another adjustment. Unless of course you know its WAY off. :)

Offline nincehelser

  • Forecaster
  • *****
  • Posts: 3337
Re: Barometric Pressure question
« Reply #6 on: July 01, 2015, 09:55:45 PM »
OK.  I'm confused.

I didn't see exactly what Acurite gear you are using, but why not take the absolute pressure (aka "station pressure") from the Acurite sensor, then have the Meteobridge calculate the altimeter setting using your elevation, and report that to CWOP?  It would also be reasonable to report the same altimeter setting to wunderground... they don't seem to really care, though.

Why are you making all the manual adjustments to other stations around you?  Is the altimeter calculation way off from them?


 

anything