On 5/10/12 10:46 AM, Chris Albertson wrote:
"mean sea level" is not meaningful any more. What shape is the ocean and what if you live in Kanas? How to extrapolate the ocean level to Kanas? The answer is to use a model of some kind
mean sea level, these days, is a name for a particular height that matches the long term average of the ocean, where there is ocean to be measured, and which smoothly varies in between those points.
The
trouble with a defining it is that it will not match what you measure with your stick in the sand. So there are any number of local definitions that are closer matches to measured heights
WGS84 won't match the stick in the sand, but one of the modern reference geoids most certainly will match it, within a few cm. It's important these days, where there are property boundaries referenced to things like "mean high tide line".
Measuring sea height (the actual height) to an accuracy of cm over a global scale is pretty straightforward these days (that's what TOPEX/JASON is all about). After that, it's a matter of choosing an appropriate averaging technique to remove the effect of tides (which you need to do on solid land, as well)
WGS84, is pretty much the "simplest" model, and is more about defining the directions of X,Y, and Z (or lat/lon) than where the surface of the earth is.
The root of the problem is that the earch has a very complex shape. It is "lumpy" in random ways and you can't model this, you have to measure it and then look it up.
You CAN model it.. and that's what the EGM model is.. using multihundred order spherical harmonics. The model isn't simple, but neither is it just a table lookup of measured data. And, as mentioned in an earlier set of posts.. since the bumps aren't huge, if you're only interested in meter scale uncertainties, a fairly small table will give you the local variation between WGS84 ellipsoid (no bumps at all) and EGM.
Chris Albertson Redondo Beach, California _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
_______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
