Re: ps from mslp,t,z0

Hi Mark,

There is no builtin NCL function to compute surface pressure.
Assuming T is the surface temperature and not a derived T
at sea level.

[1]
As a 0-th order approximation, the hypsometric eqn. (which assumes
an isothermal layer)

    psfc = pslp*exp( -(g/(R*T))*zsfc )

T (K) should be the virtual temperature if you have moisture available.
g = 9.8 , R (gas constant for dry air) =287.04 so

    T = 293.0 ; T at the sfc
    pslp = 1013.0
    zsfc = 1000.
    psfc = 901.6

Of course, the isothermal assumption will yield poor (bad)
results for high elevations (eg, Alps, Andes, Rockies,
Himalayas, ...)

[2]
A better approximation would be to use the (say) standard lapse
rate (6.5K/1000m) and calculate a mean layer temperature

   tslp = t*(6.5/1000)*zsfc + t
   Tavg = 0.5*(tslp + t)

   psfc = pslp*exp( -(g/(R*Tavg))*zsfc )

   psfc = 901.9

Again, if moisture is available, it would be best to use
virtual temperature. The issue with moisture (specific
humidity or mixing ratio) is what to use for the
lapse rate of specific humidity. This would change with different
regions. See

http://www.nrlmry.navy.mil/~chu/chap2img/fig220.jpg

Others should contribute if they have other ideas.

D

On 10/3/10 7:06 PM, mark collier wrote:
> Hi,
> I've been looking through the NCL web site material and cannot find
> any reference to calculating surface pressure from mean-sea-level
> pressure, air temperature and surface height. Does something exist or
> be converted (by me) for this?
> Thanks,
> Mark.
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Oct 4 08:57:05 2010