On Sun, Jul 13, 2003 at 04:40:53PM -0400, Erik Reuter wrote:

> But is it a phenomenological formula? I would think measurements were
> made of the lapse rate, and then a curve was fit to the data.

I did some reading and it seems things are both simpler and more complex
than my question implied.

Simpler, in that with the right assumptions a simple formula CAN be
derived for the earth. More complex because this formula doesn't always
hold true (although under "normal" atmospheric conditions in the
troposphere, with very dry air, it is a fairly good approximation).

Anyway, the basic assumption is that the atmosphere may be modeled by a
rising, dry parcel of air adiabatically expanding, reducing in pressure,
and cooling. As I mentioned, this is often valid in the troposphere. It
is NOT valid in the stratosphere, where ozone absorbs solar radiation
and actually causes RISING temperature with increasing altitude.

The applicable thermodynamic formula for this assumption is the
adiabatic law,

  p ^ ( 1 - gamma ) = c1  T ^ (-gamma)

where gamma is the ratio of the specific heat of the atmosphere at
constant pressure to the specific heat at constant volume, gamma = cp /
cv. This value is experimentally determined to be about 1.4 for N2 and
O2 (it is 1.7 for He and 1.3 for CO2)

Differentiating the adiabatic law gives

  ( 1 - gamma ) p ^ ( 1 - gamma ) dp / p = - gamma c1 T ^ (-gamma) dT / T

and substituting c1 into this equation from the adiabatic law and
solving for dp / p gives

  dp / p = ( 1 / ( 1 - 1 / gamma ) ) dT / T

For earth, I previously gave an approximate formula for pressure,

  p/p0 = exp[ -0.115 h ]

which may be written
  
  p/p0 = exp[ - h / hc ]

where hc is 8.7km by my calculation ( or 8.5km by curve fit to
experimental data). Differentiating this gives

  dp / p0 = - ( dh / hc ) exp [ - h / hc ] = - ( dh / hc ) p/p0

and solving for dp / p results in

  dp / p = - ( dh / hc )

This may be combined with the previous equation for dp / p to obtain

  - ( dh / hc ) = ( 1 / ( 1 - 1 / gamma ) ) dT / T

Solving this for dT / dh gives

  dT / dh = - ( 1 - 1 / gamma ) T / hc

If we use gamma = 1.4 , hc = 8.5km, and T=300K, then

  dT / dh = 10.1 deg/km

which approximately agrees with experimental data (the number is often
quoted as 9.8 deg/km), when the conditions fit the assumptions (one
of which is that the air must be dry, another is that the atmosphere
doesn't absorb any heat from outside sources).

For the 5km habitat, the math gets more complicated since the
"non-linear potential correction factor" cannot be neglected above about
h=2km. Also, are the assumptions used in the deriviation valid for the
habitat? If the heat for the habitat all comes from the surface (i.e.,
the endcaps are well insulated and unheated), and the air is maintained
with very low moisture, then this could be a reasonable assumption.



-- 
"Erik Reuter" <[EMAIL PROTECTED]>       http://www.erikreuter.net/
_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l

Reply via email to