- Assumption: This is similar to an axiom in maths. It is an assumption made about the world. In science assumptions are often unvoiced. For example Newtons Law of Gravity assumes instantaneous action at a distance
- Model: has a dual meaning in science. It can be either
- a relatively simple: hypothesis or theory. Model can apply to either a hypothesis or a theory
- or a more complex: mathematical model, or computer model. This kind of scientific model incorporates many know scientific theories and perhaps some hypotheses
The greenhouse gas effect, GHGE assumes:
- doubling CO2 in the atmosphere (from 280ppm to 560ppm) will raise earth's surface temperature by 1C.
- this initial 1C warming will force air to hold more water. The extra water vapour will add an additional 2C of warming.
- Total average surface warming, per doubling of CO2 in the atmosphere will be 3C (=1C + 2C). This is called the equilibrium climate sensitivity, ECS, or sometimes just the Climate Sensitivity, CS.
The greenhouse gas effect is a hypothesis, used as a core part of complex computer models. David Evans describes it.
Question: Why is this wrong?
Answer 1: Because it is an untested, non-validated model.
Answer 2: Because when the model was been tested but it failed its tests
There's nothing wrong with hypothesising the greenhouse gas effect. It all went wrong when modelers usurped science.
- They made no proper attempt to test their models. Which is why we say it is an "untested, non-validated model". It's untested because its fans avoided telling tests. It's non-validated because a good hypothesis must past all tests we can imagine before we call it validated.
- GHGE fans claimed their critics were: climate deniers, fossil fuel shills and flat-earthers. GHGE fans avoided critics and, so, by-passed scientific methods to promote this GHGE proposition to politicians
Hypothesis testing and validation
There are an almost infinite number of ways to model the world. It's thought that the fundamental laws of physics may be explained using String Theory. But there are, maybe, 10500 possible String Theories. Each a separate, distinct, hypothesis. Do any explain the world? We can only know by testing a prediction made by a model against what actually happens in reality. In comparison, it is estimated that there are between 1078 to 1082 atoms in the known, observable universe.
A hypothesis test is an experiment of observation done under controlled conditions. It's results are then compared with the predictions of the hypothesis acting under the same conditions. When the results agree, the test passes and it validated the hypothesis. When the results disagree: it fails the hypothesis and it is invalidated. A single failing test, fails the hypothesis. If 20 tests pass but one fails, the hypothesis is still failing. When a test fails, it may be possible to amend a hypothesis, perhaps by making it more complex. So that the amended hypothesis can then be tested. An amended hypothesis must be tested from the start. It must pass all its tests to be considered validated.
A hypothesis made with example tests is called a testable hypothesis. Scientists don't generally waste their time on untestable hypotheses. Because there are an infinite number of such, and no way to say whether any give a true explanation for something in the world. When a hypothesis has passed sufficient tests, we say it is a validated hypothesis. That is about the same thing as an accepted scientific theory.
How does the greenhouse gas hypothesis measure up?
No good tests were ever written for the greenhouse gas hypothesis. This is partly because it is a complex hypothesis. With an initial CO2 warming effect followed by an water vapour amplification (the feedback). Also because it is a hypothesis about earth's climate system. We don't do such experiments on climate. This explains why the greenhouse gas effect is untestable from the conventional point-of-view.
It does not make the greenhouse gas effect intrinsically untestable. It's possible to find simple tests which can be done as experiments. One such test is to quantify the surface warming effect. One could take an infrared, which emits at CO2 radiative frequencies, and measure the warming observed on a known quantity of water. This is the simplest, and probably the cheapest, experiment I can think of. A kosher scientific hypothesis should be tested in as many ways as possible. At least a dozen different ways.
It's the responsibility of the hypothesis inventor to propose suitable tests, but anyone many join in finding more. The job of research science is mainly a job of discovery. Discovering whether a hypothesis is a true picture of the world is the most important; because this gives us new, fundamental science.
1. Infrared emitted by atmospheric gases, including carbon dioxide does not penetrate into water very far. It goes mere micrometres. This means oceans are not warmed by more carbon dioxide in the atmosphere. In particular, alarmists claim surface temperatures will warm by 1C when CO2 atmospheric concentrations reach 560ppm. For that to happen the IR must penetrate as deeply into water as sunlight does. Instead much of the infrared energy absorbed by water will be immediately lost. To both re-radiation (towards the sky), and evaporative cooling. In evaporative cooling, latent heat is absorbed by water from it's surroundings and used to break hydrogen bonds holding water molecules together in chains. This latent heat does not raise water temperature. Vapourised water takes about 8 days, on average, to reach cloud layer where it's much cooler. Once there it condenses, emitting it's latent heat as infrared to space.
71% of earth's surface is water. Let's assume that most of the warming is lost (60%).
- Comment of behaviour of microwaved warming water at WUWT, and 2 replies.
It looks like models treat all EMR (radiation) the same way. As so many W/m²; whether sunlight or infrared. Yet it’s not all the same from the point of view of surface warming. Sunlight penetrates many metres into water. DWIR penetrates mere micrometres, so warms only the surface skin. Re-emission at the surface can only penetrate downwards another few micrometres, but there’s no barrier to upward re-emission. Conduction downwards is slow, and convection impossible because the lightest water is at the surface. Much of this surface skin warming will goes latent heat, evaporating water, which condenses at the upper troposphere, releasing it’s latent heat here, so effectively cooling the surface. Isn’t this a fundamental error by the climate modelers – treating all EMR the same for the purposes of surface warming? Why don’t people make a bigger deal of this?
-- Mark P
(1) Indeed solar and LWIR behave differently with regard to penetration, but there are other processes occurring simultaneously long with it. For example, there is evaporation. Also the ocean is not a glassy pond. It is roiled by wind and surface waves along with some convection so that an upper layer becomes mixed. This convection is not exclusively thermal, but is also the result of salinity changes. Surface evaporation cools the surface and makes it more saline. Often one can see the results of surface convection in streets of foam and debris lined up in the prevailing wind direction (roll cells). It is a complicated problem of combined mass and heat transport.
-- Kevin kilty September 29, 2019 at 7:24 am
(2) I’m not aware of any modeling to address the issue. But clearly IR radiation is important to the ocean heat budget, because the ocean almost everywhere is losing IR energy. So, if you reduce that loss, then ocean temperature will change. The ocean mixed layer is always experiencing mixing from waves, and even evaporation causes mixing by creating a cool skin that is more dense than just under the skin.
-- Roy W. Spencer September 29, 2019 at 7:43 am
- Real scientists are collaborators. They rely on other scientists to find flaws in their ideas and to propose clear-cut tests for their hypotheses. Real scientists are ever eager to discover whether a scientific hypothesis is any good. Whether it passes its tests. Good scientists know that it's senseless to use a non-validated hypothesis to predict the climate. It's insane to shield a non-validated hypothesis from criticism.
Climate alarmists have known about the inability of CO2 to penetrate far into water for decades. They were told they needed to experiment to find precise water warming value. They monopolise climate research funds but refuse to fund essential quantitative experiments. Alarmists never made serious attempts to test their mathematical models against reality. I know of no climate modelers in the so-called climate consensus, who care what really happens in the climate. They are too busing using it as a political football.
Penetration of Infrared radiation into water
2. There's no sign of climate feedback which supposedly triples warming due to CO2.