Irrespective of how accepted any scientific theory happens to be it is often worth thinking about a possible experiment to test said theory and perhaps falsify it. If we wanted to do this for the theory of the greenhouse gas effect what areas are worth considering when designing such an experiment? Would we limit ourselves to just the radiative heat response of CO2 or are their wider issues we need to incorporate into the experiment?

This question is NOT a duplicate as it relates to the generality of greenhouse gases and specifically asks IF we should limit ourselves to the radiative response of CO2. Therefore one possible avenue of answering could be to consider the wider issues of water vapour and methane and such an answer would be clearly outside the scope of the CO2 specific question which has been postulated as the duplicate.

  • 3
    $\begingroup$ Add lots of CO2 to the atmosphere. If warming does not take place as predicted, your experiment has falsified the effect. Unfortunately, if your experiment fails to falsify the effect, you're stuck with the consequences :-) $\endgroup$
    – jamesqf
    Commented Apr 27, 2017 at 5:31
  • 2
    $\begingroup$ I find your question simultaneously unclear and broad. What hypothesis exactly do you want to test? $\endgroup$
    – gerrit
    Commented Apr 27, 2017 at 13:02
  • 2
    $\begingroup$ @user7733: 1) No, I can't do better. What I described is a perfectly valid experiment. 2) What you mistake for comedy was really intended as sarcasm. Testing of said theory is not the least bit tricky, and the process is extremely well documented: e.g. history.aip.org/climate/index.htm The only problem is your evident unwillingness to accept the experimental evidence. $\endgroup$
    – jamesqf
    Commented Apr 27, 2017 at 18:09
  • 3
    $\begingroup$ @user7733: The experiment on the existing atmosphere is already in progress. We have results to date which are consistent with the greenhouse gas effect. The lab experiments were done in the 19th century, and are well documented, e.g. in the link I posted. $\endgroup$
    – jamesqf
    Commented Apr 28, 2017 at 6:09
  • 2
    $\begingroup$ @user7733: And re Loschmidt &c, while I've just skimmed a few results of a web search, I can't see how it's in any way material. Earth's gravity has not materially changed over the last couple of centuries (or indeed, since shortly after the impact believed to have created the moon), so it's hard to see how such an effect could be responsible for the temperature increase of past decades. $\endgroup$
    – jamesqf
    Commented Apr 28, 2017 at 6:19

1 Answer 1


In the interest of clarifying this a bit.

A Study has the purpose of acquiring knowledge. Many areas of science are fields of study. Nearly everything NASA does is a study. NASA may conduct the occasional experiment, but 98%-99% of what they do is for the specific goal of acquiring knowledge, studying the solar system and universe. The importance of study should not be underestimated. Many fields in science are studied but not experimented on. Astronomy for one. Geology is another one. Some fields of biology and ecology aren't good for experimental testing, but are still studied.

An Experiment, under proper control has enough test subjects to overcome any numerical coincidence in the outcome, and a set of fixed criteria with one specific thing changed.

We can't run true experiments on climate change in the classical experiment sense because we have only one Earth. We'd need at least two identical Earths to run a proper experiment, and ideally, many more than two. That's obviously impossible.

A Research Study can be an experiment or incorporate one or more experiments, but it also, often isn't an experiment. While an experiment is generally quite specific, a research study can be more broad. What your question seems to be asking is about research studies, not experiments. A good research study leaves an equal chance of either a "yes" and a "no" outcome.

Bias research is possible, so lets get that out of the way. Unbiased research leaves the outcome entirely and fairly up to the results of the study. Now, a scientist can still have one outcome in mind. He can even intend to prove something, but that intention doesn't guarantee bias, so long as the study is well outlined and fair.

Climate change has been subject to numerous research studies. I'll list a few. My list below is far from complete.

  • Measure heat in/heat out from space is a basic one. Satalites measure the heat coming off the earth and the heat coming from the sun. Changes in heat in/heat out are compared to temperature changes on Earth. This is a very basic study that could go either way. It supported greenhouse gas driven climate change.

  • Temperature change day vs night. If it's the grenhouse gas, then nights should warm faster than days. This was studied and nights did warm faster than days. Another study that supported the greenhouse gas theory.

  • The cosmic ray/cloud formation theory. That cosmic rays could indirectly heat the earth by cloud formation sounded nuts, but it's actually as good a theory as any and it was studied, at CERN no less. This study had some merit, but it's insufficient to explain the current warming.

  • Urban Heat Island. In what I consider a hilarious bit of failure, the Koch brothers' funded Berkeley Earth Study was designed specifically to avoid any Urban Heat Island warming and disprove climate change. As I said above, agenda driven research is fine, so long as the research fairly allows the undesired outcome. Well, that's exactly what their study gave them. Their research, designed specifically to debunk man made climate change theory produced results that the Earth is warming even faster than NASA's and the IPCC's results.

  • Studying heat at different altitudes. Again, if it's the greenhouse gas, more warming should occur lower in the atmosphere, less as you gain altitude. This is exactly what the study found. The high altitude temperature readings from UAH and RSS actually support, they don't dispute greenhouse gas driven climate change.

Any one of these research studies could have made a case for or against. Every one of them, and every research study I've ever read about has supported the man made climate change theory.

Climate change has so much attention, that lots of research has been done. This isn't a case of one study that everyone agrees with, without proper cross-checking. It's been tested a dozen different ways a dozen different times and gone over by thousands of scientists and skeptics. Many studies have been done with the intent to disprove it too. There's a lot of special interest money out there to oppose the theory and they've failed to provide a single good research study that backs up their alternative theory claim. There's not one study that produced results against the prevailing theory and many studies in favor. That's science, not bias.

  • $\begingroup$ @userLTK. Please point out where exactly you think I have "rejected lab experiments that show CO2 traps heat".(Because I have done no such thing). I have also not disagreed with the problem of conducting a 2 worlds experiment either.It is self evident you cannot conduct an atmospheric scale experiment. However there is quite obviously a window to consider. The size in the middle. One could theoretically conduct a large scale lab experiment, tens if not low 100s of metres in size is possible, we can find a warehouse that big. Also it may be (continued) $\endgroup$
    – user7733
    Commented Apr 30, 2017 at 22:19
  • $\begingroup$ possible to conduct an experiment horizontally in the existing atmosphere over a number of kilometers given a quiet location (uninhabited desert?). The kilometer sized experiment may not even need the atmosphere to be enclosed by tubing. All these are possibilities and I am surprised nobody came up with them before so we could talk about the details. What needs to be tested is not just the IR response of CO2 or its "trapping" of heat but the total thermal and radiative transfer properties in a bulk sample between 2 ends of a long column. $\endgroup$
    – user7733
    Commented Apr 30, 2017 at 22:27
  • 1
    $\begingroup$ @user7733 next time, maybe lead with the experiment you have in mind. That would have been a clearer starting point. A large closed experiment in a warehouse isn't a bad idea. Testing CO2 in lower concentration and higher volume. As for the outdoor, that wouldn't work. First it would be local, second even enclosed it would be subject to weather from outside. Local and short term isn't relevant. Having a specific idea in mind but asking a general question with odd parameters and then criticizing the answers and comments isn't the way to go. $\endgroup$
    – userLTK
    Commented Apr 30, 2017 at 23:10

Not the answer you're looking for? Browse other questions tagged or ask your own question.