An important assumption made in much of climate modeling is that greenhouse gases somehow affect the adiabatic lapse rate and as a result controls the earth’s temperature. This assumption seems to be at the core of the debate over why atmospheric measurements do not correspond to climate models.
An explanation of why greenhouse gases do not affect the recently proposed “Greenhouse Equation” (here) or lapse rate is here. But others have examined the lapse rate issue as well. For example, consider the explanation provided here which is fairly objective until the last paragraph, which states:
“Data collection of atmospheric lapse rates using radiosondes is a important tool in weather and climate analysis. Particularly in regards to climate, since the early 1990s, scientists have carried-on an extensive debate as to whether radiosonde data can be treated as reliable or not. The concern emerged from apparent discrepancies between observed (radiosonde and satellite) data and model predictions on account that observations did not show the warming in the troposphere predicted by models (Spencer and Christy, 1990, Hansen et al., 1995). These differences have been attributed to a variety of causes over the years and have recently been narrowed down to the “temporal inhomogeneity” of observed radiosonde data (Gaffen, 1994, Parker and Cox 1995). These time-varying biases are generally induced by the change of instrumentation, the scarce quantification of related uncertainties, and the lack of documentation.”
At this point, the authors express an opinion (not science):
“Recent research supports the current conclusion that no strong evidence exists of discrepancies between observed and modeled tropospheric trends when correcting for these biases (Trenberth et al, 2007, Thorne et al, 2011).”
The authors then conclude:
“Thus understanding and continued monitoring of atmospheric lapse rates is fundamental to weather and climate research.”
My comment is this. To explain discrepancies labelled as “temporal inhomogeneity” in and of itself does not provide sufficient basis for concluding that such discrepancies actually occurred when the measurements were taken. In short, “Saying so does not make it so.”
As a result, it must be recognized that the statement “…that no strong evidence exists of discrepancies between observed and modeled tropospheric trends when correcting for these biases.” is arbitrary and based on an arbitrary adjustment of the data.
Any arbitrary claim is also reversible: “…that no strong evidence exists of factors presumed to cause “temporal inhomogeneity” in atmospheric measurements have actually corrupted the historical observed data.” In short, if there is insufficient evidence to arbitrate in favor of a particular conclusion, then there is also insufficient evidence to favor the opposite conclusion. To favor one or the other using arguments relying on “temporal inhomogeneity” is arbitrary.
An “after the fact” adjustment of observed data based on something that “could of happened” in order to fit the data to a model is very dicey and does not qualify as “science.” Such underlying assumptions imposed on the analysis of the data leads one to conclude what has already been assumed. This happens. It can happen innocently. But it must be recognized and accepted for what it is….arbitrary.
That being the case, computer models are not “reality” but imperfect representations of it. Observational data provide an indication of how the “whole” system is behaving which computer models simply cannot do. And despite limitations in measurement technology observations are best “science” available.