NCEPatm_2.m

Syntax:

pressure = NCEPatm_2(site_lat,site_lon,site_elv)

This function converts elevation $z$ (m) to atmospheric pressure $p$ (hPa) using the pressure-elevation relationship in the ICAO Standard Atmosphere:

$$p(z) = p_{s} exp{ \Big\{ - \frac{gM}{R\xi} \left[ \ln T_{s} - \ln \left( T_{s} - \xi z \right) \right] \Big\}}$$ (41)

Where $M$ is the molar weight of air, $g$ the acceleration due to gravity, and $R$ the gas constant, giving $gM/R =$ 0.03417 K $\cdot$ m$^{-1}$. The adiabatic lapse rate $\xi$ is taken to be 0.0065 K $\cdot$ m$^{-1}$.

The sea level pressure $p_{s}$ (hPa) and the sea level temperature $T_{s} = 288.15$ K are obtained by interpolating the sample location onto global grids of annual mean sea level pressure and annual mean 1000 mbar temperature generated by the NCAR-NCEP reanalysis:

http://www.cdc.noaa.gov/ncep_reanalysis/

This function performs reasonably well for the deep southern latitudes, but it's not recommended for this purpose. The function antatm.m (which is fit to actual station measurements) should do a much better job.

There is more discussion of how well this atmosphere approximation performs against actual station measurements in the main text of the paper. In particular see Figures 1 and 2.