Can statistical evidence determine the state of our environment?
“There is no global temperature. The reasons lie in the properties of the equation of state governing the local thermodynamic equilibrium, and the implications cannot be avoided by substituting statistics for physics.” (see Dr. Matthew M. Wielicki on X: “Just a quick reminder… https://t.co/NlQ8YcU1IZ” / X (twitter.com))
The mean states of the earth/atmosphere system, while chaotic and constantly changing, can be sampled and analyzed like any other complex system.
Statistical evidence can be highly compelling when the measures of central tendency and dispersion (i.e., means, standard deviations, standards errors, etc.) are derived from large, unbiased samples.
Global temperature data compiled from satellites satisfy that requirement because satellites can continuously sample lower troposphere temps at resolutions of less than 1 kilometer.
Furthermore, polar orbiting satellites can orbit 14 times per day, providing good temporal coverage.
On the other hand, critical metrics derived from fixed sites are highly problematic.
Ground-based data from remote areas such as Antarctica, the Southern Ocean, the Arctic, Siberia/Cental Asia, the South Pacific, etc. critically “under sample” as a result of low spatial resolutions (typically on the order of tens or hundreds of kilometers) and/or low reporting frequencies.
Other problems can also arise from instrument malfunctions, instrument replacement, observer error, site changes, and reporting gaps, to name just a few.
Particularly problematic is that a significant percentage of those sites are located in urban areas, creating a sample that is strongly biased towards higher temperatures.
In effect, weather and climate can only be defined in terms of probabilities, just as the behaviors of subatomic phenomena can only be described in probabilistic terms, i.e., quantum physics.
According to Cal Tech (What Is Quantum Physics? Quantum Physics in Simple Terms – Caltech Science Exchange):
Quantum physics is the study of matter and energy at the most fundamental level. It aims to uncover the properties and behaviors of the very building blocks of nature.
While many quantum experiments examine very small objects, such as electrons and photons, quantum phenomena are all around us, acting on every scale.
Furthermore, as described in Wikipedia (Quantum mechanics – Wikipedia):
A fundamental feature of the theory is that it usually cannot predict with certainty what will happen, but only give probabilities.
In the geosciences, we work within this framework every day, and routinely issue probabilistic forecasts such as Probability of Precipitation (PoP) forecasts, hurricane risk maps, NHC Track Forecast Cones (i.e., cones of uncertainty,), flood zone delineations etc.
To conclude, we can, in fact, derive accurate estimates of global temperature with large, unbiased samples.
Furthermore, as a quantum phenomenon, we can issue useful, probabilistic temperature forecasts for a wide array of spatial and temporal scales.
Please Donate Below To Support Our Ongoing Work To Defend The Scientific Method
PRINCIPIA SCIENTIFIC INTERNATIONAL, legally registered in the UK as a company 
incorporated for charitable purposes. Head Office: 27 Old Gloucester Street, London WC1N 3AX.
Trackback from your site.
Tom
| #
All we get are fake statistics and politically driven nonsense. The state of the planet’s temp may be ascertained with direct measurements but that in of itself means nothing. It’s the long term trends that should be accurately measured and I mean very long term as in hundreds of years, if not thousands.
Then again to be even slightly accurate and useful, one must separate what the sun provides verses what man provides. Not so easy to do if at all. Verdict: global warming does not exist other than what the sun and our weather systems affect. Man’s contribution is insignificant and always will be.
Reply
Oz
| #
statistics cannot determine what was the cause and what the effect.
example: statistic found that 100% people who ate cucumbers died, so cucumbers are poison food ?
Reply
Arthur Viterito
| #
Using statistics does not exempt you from proper experimental design. There must be a reasonable cause to explain the effect before you start crunching the numbers. Statistics can be VERY demanding that way, as your independent variables must be able to behave independently while the dependent variables must, in fact, be tied to them (i.e., “dependent” on them) in some plausible way.
As an aside, to invert your flawed hypothetical, we can also say that 100% of all folks who have never had a cucumber have also died. Does that make cucumbers an essential food for immortality?
Reply