Sunday, April 17, 2016

Mother Nature

One of my favorite scenes in Wagner is Wotan accepting the will of Erda. Yes, even the most powerful of the gods must obey "Mother Earth".

Most "scientists" dismiss me when I invoke Mother Nature, and that is silly, for we must obey "Mother Nature".  Here, I am going to suggest that a good number of my readers go back and read
(or reread) the first 10 of the Feynman Lectures .  Those remind us of how scientists think and work (or should think and work).  Too many of us rely on what some grade school teacher told us about how scientists think and work.  Now think!, did that teacher discuss data quality at the time? If not, there were problems with the teacher's data. The first 10 Feynman lectures are about how real scientists work out difficult issues.  Those lectures explain why no object can be precisely defined, why all data is imperfect, and how to tell if a particular set of data is good enough.

Mother Nature uses the math of probability, calculus, equations of state, finite element analysis,  catastrophe theory, and quantum mechanics to tell us what is likely. She imposes the forces of  gravity, and electromagnetic radiation on us within the constraints of thermal dynamics and rotational dynamics (including the orbital mechanics that give us ice ages, seasons, and tides). She sets the properties of matter and plasma, and the conditions when those properties change. She sets the biochemistry of life. We can calculate Pi to a trillion decimal places, but we cannot change it.  We can measure the size of a neutron, but we cannot change it. We can calculate gravity to one part in a billion, but we cannot say how gravity is different from the forces that hold a neutron together.

Invoking Mother Nature is a good shorthand for all of this.  It is a good shorthand for the fact that we need to be doing all the math, all the time, or we will miss part of what Mother Nature is telling us. (Richard Feynman was very good at math and he spent a good bit of time doing math.)

Setting Mother Nature as a cool entity of  math allows us to forget her power. She  lets us make nuclear bombs and wipe out species with abandon.  She will indifferently allow us to wipe out our  own species. She has reminded us, that with all of our ability to destroy species, we have a hard time dealing with Zika viruses,  Lyme disease,  MSA, cockroaches,  termites, volcanoes, or tornadoes and hurricanes -- much less global climate change. We have more hubris than power.

I think the scientists that dismiss me when I invoke Mother Nature are not doing the necessary math on a timely basis to understand the consequences of  human caused global warming.  They think humans are special, and Mother Nature will not enforce her rules.  Trilobites were special and they are gone. The big dinosaurs were special and they are gone.  That is why Mother Nature states her rules mathematically - those rules are are mathematically precise and are always enforced.

Climate models pass for doing math in climate science.  Just as a camel is a horse built by a committee, climate models are computer programs built by committees in an environment where there are large numbers of students and many echo chambers.  The number of students ensures that nothing too unpleasant is said, and that nothing too rigorous and time consuming is done.  I mean that one cannot have college students running climate models that predict the end of civilization within the  career of the student - the whole university community would need trauma counselling.

And yet, any kind of a reasonable model of seafloor clathrates in a warming ocean would have to include a fat tailed probability distribution of methane releases from the sea floor.  Some of those fat tailed probability distributions of methane releases would be at catastrophic levels, so the entire concept of  methane releases from sea floor methane clathrates is left out of the climate models.  As is also carbon dioxide from thawing permafrost.  This is not doing math, this is pulling the covers over our head to keep the monsters in the dark away.

Even the math that the climate models do, does not stand up to scrutiny.  Methane is about 86 times more powerful as a greenhouse gas than carbon dioxide on annual basis.  However, methane degrades in the atmosphere over time so over a period of 100 years, methane is only 24 time more powerful as a greenhouse gas.  The climate models set a time frame of  100 years and use the lower value as the CO2 equivalent of  CH4 over the entire time frame. However, climate change is a non-linear feedback system with later conditions dependent on earlier conditions. (See Jay Forrester's work on systems dynamics)  In a non linear feedback system, each time period must be calculated with the appropriate values for that time period, or subsequent time periods will be affected. Therefore this use of average methane  to CO2   equivalent is a catastrophic bug in the program.  Moreover, the assumption of declining CH4 concentrations in the atmosphere has NOT been observed in 27 years of CH4 monitoring.  I can think of several good chemical explanations, but the point is that the observed conditions  do not fit the conditions assumed in the climate models - and that is a deep failure of the climate models.

Review papers such as  Steve's  missed these catastrophic bugs in climate models. The methane/ carbon dioxide equivalent error is just one of several bad bugs in the science coded into the models. The models were written as tools to study narrow aspects of climate change, and not to predict the over-all effects of climate change. To use the current climate models to estimate climate changes from carbon emissions is a misuse of the models. (e.g., temperature rise and carbon budgets for treaties)  Even, to use the models to estimate effects of known climate forcing on human society is a misuse of the models.

This one methane / CO2 equlivalent issue means that the Earth is likely warming  ~35% faster than the climate models predict as a result of this one bug.   There are other problems with the climate models.   Consider:

The take away is that climate scientists are very surprised that climate change is unfolding much faster than they expected, despite many thousands of  climate model runs to establish boundary conditions.

All of the information in the sites listed above should have been expected if we had honest climate models. If we had honest climate models, the 2007 and 2012 Arctic Sea Ice melt events should have been expected.  The record  heat waves of the last decade ( including the July 2012 GIS melt event) should have been expected from honest climate models -- that appropriately accounted for current forcing from methane.  And, the evidence of increasing carbon feedbacks would have been anticipated.  In a system of nonlinear feedbacks, that last is a critical and urgent issue.

Honest climate models might well have resulted in a climate agreement 20 years earlier.

No comments: