Atmospheric Modeling/Operational Meteorology

Department of Environmental Sciences

 University of Virginia

 Charlottesville                      

 

 

Up

Derivation Part 1

Deriving Specific Humidity from Relative Humidity

Soden and Bretherton [1993] derived the following primary result:

                                                                                            

where T6.7 is the cloud-free 6.7 µm brightness temperature,  r is the layer-averaged upper-tropospheric humidity, p0 is the ratio between the pressure of the 240K isotherm and 350 hPa, q is the satellite zenith angle, and a and b are empirically derived constants 31.2 and 0.115 K-1 respectively. In order to derive specific humidity q (or, analogously, mixing ratio, w), we begin with the definition for r  in terms of mixing ratio:

 

                                        

Following the example of Soden and Bretherton [1993] we assume an atmosphere in which r is independent of height over the range of the upper troposphere, and we assume the vapor pressure is much less than the pressure.  The Clausius-Clapyeron equation is approximated as an exponential function of temperature and substituted into the equation above to give:

 

                                                                                          

The saturation vapor pressure at 240K, es(To), is 38.1 Pa. The quantity T' describes the deviation of the upper-troposphere layer-averaged temperature from a reference value ( T'=T - To), where To and Po are an upper-tropospheric reference level temperature and pressure, respectively (taken as 240K and 400 hPa)
. Taking the natural logarithm of both sides,  we have another expression describing the layer-averaged upper-tropospheric humidity, this time as a function of two terms; one depends on the absolute amount of water vapor in the layer and the other is a function of layer temperature:

                                                                              

the coefficients l and c are constants (-23.1, 12.04) based on the reference values chosen. Now, we can substitute the first equation for ln r  into the left-hand-side above, and rearranging terms we can show that the remotely sensed quantity T6.7 can be regarded as the sum of four independent terms plus a constant.

 

       

We can simplify this conceptually by subtracting the reference level temperature of 240K from both sides of equation 8, and then rearranging:

          

Now, we can readily see the remotely sensed radiative temperature, the equivalent brightness temperature at 6.7mm, is comprised of a reference level temperature and three independent terms, one a function of the layer-average water vapor mixing ratio (or specific humidity), one a function of the layer-average temperature, and one a function of the satellite viewing angle.  

 

Home ] Up ]

Send mail to moody@virginia.edu with questions or comments about this web site.