ch...@chrismaness.com writes:
> I am nerding out calibrating my barometer today. I have a handheld
> precision barometer that I can take co a vertical survey marker close
> to a nearby major airport. I can then call the automated message at
> the tower and get real time AS (altimeter settings) from the airport
> barometer. I just am not sure what height these AS settings are
> referenced to. There are two listed in the marker's datasheet. Ortho
> and Ellipsoid. Ortho is based on spirit leveling and a better
> reference to a hypothetical equipotential surface that is tied to one
> tide station in Quebec (NAVD88) or the GPS measured Ellipsoid (a
> simple oblate sphere that is supposed to approximate said
> equipotential surface).
(I am assuming you are in the US and further in CONUS, since you are
talking about NOAA. While this message is US-centric, almost all of it
has analogs for other countries.)
Glad to see someone being extra nerdy about this! I have been slowly
working on a long-term project to calibrate my barometer. I have a
Davis VP2 and a friend about 40km west has one too. I look at both
pressure values and can see approaching waves, but overall they are
close to zero mean. Each has been calibrated by setting approximate
altitude and then by setting to mostly read the same as local airports.
As you figure out the altitude precisely, you'll end up haing to adjust
the pressure sensor itself, in order to get pressure reduced to Sea
Level to be accurate. However, having elevation close results in a
better situation over time than assuming you are at 0.
In the US there are 4 different (current) notions of height, plus
superceded past notions and future notions.
The official notion of height in the National Sprtial Reference System
is "orthometric height" as NAVD88. That means distance along the plumb
line to a reference surface.. Very loosely, distance above Mean Sea
Level. NAVD88 is not 100% accurately defined vs gravity, as there are
errors in definition across the continent, but the standard approach in
the US is to refer heights to the NAVD88 reference (=0 m) surface. You are correct
that running leveling loops is the way to transfer NAVD88.
Because barometric pressure is supposed to be reduced to Mean Sea Level,
almost certainly in the US, the pressure reference means 0 m NAVD88.
The next notion is ellipsoidal height (Height Above Ellipsoid or HAE) in
NAD83. NAD83 is a 3-dimensional datum fixed to the North American
Plate. There is GEOID18 which is a model relating NAD83 HAE to NAVD88.
Around me (Boston), HAE and NAVD88 are about 29m different which is
huge.
The third notion is HAE in WGS84(G1762), which is essentially the same
as ITRF2014. This HAE is different from NAD83 HAE because the
ellipsoids have a different origin, but all things NAD83/WGS84 are
within about 2m, about 1m horizontally. Compared to anything you can
measure with a barometer, that's identical.
The fourth notion is "WGS84 orthometric height", also called "EGM2008
height', which is distance along the plumb line to the WGS84 reference
equipotential surface, as determined by HAE measured by GPS and the
EGM2008 gravity model. Some object to this because WGS84 is a US
military thing. (While ITRF2014 is not a US thing, I am unaware of ITRF
definining gravity-based height.)
Because there is a notion of international comparison of pressure data,
it's a really good question if NOAA's altitude reference is
WGS84/EGM2008, or NAVD88. However, my basic take is that NAVD88 and
WGS84 orthometric height are within a meter or so -- both intending to
model some average sea level -- and indivdual barometers have
calibration errors, so some sort of calibration/norming is needed and
this absorbs any definitional error.
Put another way, can you measure pressure well enough to establish
absolute heights that are accurate to better than 1m? If so, let me
know how!
Your use of a barometric altimeter to transfer elevation from a
benchmark to your station makes sense. I would measure at home, go to
the mark, meausure, and come home and measure, so you can compare the
before/after average with the mid measurement.
My impression is that the standard approach is to wait until there is a
persistent high pressure with little variance and then compared your
(corrected to sea level) barometer with that of nearby official stations
and then basically apply an offset. I did that and my configured
station height is something that I know believe is off. But I think the
combination of aboslute pressure measurement error and height error is
close to zero.
Only recently have do I have the ability to measure height accurately.
A friend brought over a 25-year-old survey-grade L1/L2 GPS receiver. We
set it up and measured for 30h and processed the data vai the NGS. Now,
one can do RTK with the u-blox F9P, and be similar. So I continue on my
quest for barometer calibration.