Temperature or Radiative Flux? Pressure or Convective Flux?

The prime evidence put forward to support global warming alarmism is a measurement of the spectrum of  Outgoing Longwave Radiation OLR from the Earth into cold empty space showing that total OLR is 1% less than total incoming short wave radiation from the Sun with the message that the Earth is heating up. 

The measurement is done from satellites (AIRS and ISIS) looking down on the atmosphere using instruments in the form of spectrometers supposedly measuring radiative fluxes over a range of infrared frequencies forming the spectrum, see previous posts on OLR.  The instruments use bolometers based on thermopiles sensitive to temperature and compute radiative fluxes using complex software for radiative heat transfer such as Modtran. The instruments thus directly measure temperature and then report radiative flux after postprocessing. To confirm global warming,  better accuracy than 1% is required for total OLR and also of course for total incoming radiation. Is this possible?

To compute radiative flux from input of temperature using software for radiative heat transfer requires input of coefficients of emissivity, absorptivity and transmissivity, and so has serious issues as concerns accuracy. 

To see a basic issue, let us compare with a more familiar setting of seeking to compute the fluid flux in a pipe or around an object from reading pressure. We then recall that pressure can be read by a pitot tube:

where the left open end is inserted into the fluid and so takes on the stagnation pressure or total pressure in the fluid passing by and then is read by a nanometer to the right. We understand that measuring pressure can be done with high precision, but if we now ask about the total convective fluid flux, we will have to supply additional information about the nature of the flow. If the flow is steady, inviscid, incompressible and irrotational, then Bernouilli's  Law can be used to compute fluid velocity and so convective flux, but that is a very special case. 

We thus see that reading temperature by a thermopile or pressure by a Pitot tube can be done directly by an instrument for which the physics is transparent. 

On the other hand, to report radiative/convective flux from reading of temperature/pressure is a complicated issue requiring detailed additional information, which may not be available. 

Direct measurement of total flux by some form of capturing technique like that in an anemometer, also is subject to very big uncertainties.  

Supporting global warming on an assessment of OLR measured to less 1% precision lacks credibility.