OK, here's my sob-story: I am tired of beating my head against the "Davis-induced SLP Wall" and need an accurate method to "back" the 'raw' barometer sensor Station Pressure "out" of the VP2-calculated Sea Level Pressure (SLP) pressure data that WL-ISP sends to CWOP / MADIS / MesoWest, et.al..
I read the Barometer Pressure section of Davis Application Note #28, Rev. A, dated 5/11/2006, with absolutely NO success because the humidity correction coefficient "C" isn't calculated, but rather comes from a "lookup table." The text wording says the lookup table is "...(provided in the attached table)..."; however, there is NO attached table! (I've e-mailed Davis about this omission) Most likely they meant use the "lookup table" found in one of the footnote referenced documents, but they forgot to specify *which* footnote document!
So, what I did was collect the last seven days of weather data for weather stations KTUS(NWS @ Tucson airport), KDMA(USAF @ DMAFB), CW5619, and DW6988(my station), into an Excel spreadsheet. Data from each station was collected into columnar rows: SLP, ALT, Sta.P, SLP/ALT-ratio, etc.. I then compared the columns of data against each other and found two sets of data pairs with very high correlation coefficients:
• PART I - What's Related to What?
1) Pstn vs. Palt (both in "Hg):
KTUS(2546'): Palt = 1.0764*(Pstn) + 0.5599; R2 = 0.9999
KDMA(2703'): Palt = 1.0850*(Pstn) + 0.4952; R2 = 0.9999
2) K vs. SLP/ALT-ratio (K is mb/meter; SLP/ALT is dimensionless):
KTUS(776m): (Pslp/Palt) = 1.0724 - 0.0085*(K); R2 = 0.9782
KDMA(824m): (Pslp/Palt) = 1.0743 - 0.0087*(K); R2 = 0.9625
NOTE - "scale height" coefficient K, in mb/meter, is found in the following common equation:
Pslp(mb) = Pstn(mb) + (height(m) / K ) ...where K-values range between 8.5 and 9.2
So, what did I learn so far from this exercise? (1) First, the NWS and USAF conversion of Sta.P into ALT is almost perfectly LINEAR! (2) Secondly, the "scale height" coefficient K is LINEARLY related to the ratio of the SLP and ALT values.
• PART II - Putting the Numbers to Work (...TBD...)