Towards a Rational Approach to Climate Change

To decide how to deal with climate change we have to first look at the affects it will have.  The effects which have been discussed fall into three groups.

These are:

• Effect on wildlife and the environment
• More frequent extreme weather events
• Rising sea levels

My conclusion is that the only aspect that poses a real threat to humanity is a rising sea level.

In this blog I shall look at all three aspects and explain the reasoning behind the conclusion.

Effect on wildlife and the environment

At present, any change in animal populations or animal distributions is attributed to humans, whether it is blamed on the loss of habitat, hunting or  anthropogenic climate change.

However if we look at our understanding of changes of populations and ranges of different species it is clear that our understanding is very poor.  Assigning all changes to manmade effects is not justified. Natural events also affect wildlife.  In some cases it is very clear that mankind was responsible, for example the dodo was hunted to extinction by Portuguese sailors.  However the demise of the passenger pigeon is problematic – to date there is no consensus as to what caused their rapid extinction in the late 19^th and early 20^th centuries when they went from large flocks of millions with a total population of 3 billion to one single individual which finally died. Many hypotheses have been presented to explain this but so far there is no conclusive evidence one way or the other (recent work suggests that lack of variety in their DNA led to huge swings in population, which coupled with intense hunting led to their demise).

Our understanding of evolution is also incomplete.  Recent studies have shown that evolution can happen on much shorter time scales than anyone had expected:

• There are physical differences between European blackbirds which lived in the cities and those in the countryside.
• There is a physical difference between the population of Black-capped Warbler which winters on the Iberian Peninsula and those that winter in Britain. The wintering in Britain only started about 20 years ago so evolution has caused genetic changes in the two populations in that time-frame.
• A recent study of Darwin Finches in the Galapagos Islands concluded that a new species can evolve in as little as two generations (Tim Collins, Mailonline 23 November, 2017).

All of these suggest that animals can adapt to changes in their environment faster than had been expected (my hypothesis).

In the 1970s I’ve frequently seen American Kestrels in my neighborhood.  Now it is rare to see an American Kestrel but Merlins, which are a related species, are now frequently seen.  Clearly the ranges of the two species have changed. It is difficult to know exactly what changes have caused this.

Due to our poor understanding of how climate change affects wildlife, our best approach is to monitor changes and help to preserve species at risk without necessarily trying to undo all the changes.

More frequent extreme weather events

Computer climate models predict that “extreme weather events” will become both more frequent and more extreme. This is explained qualitatively by the following analysis:

• Polar regions are heating up more rapidly than equatorial regions.
• This results in a smaller temperature difference between the two.
• The jet stream becomes more unstable as a result.
• Since the jet stream has a profound effect on the weather, this instability results in rapid changes.
• The rapid changes in the jet stream cause extreme weather events.

This leads to several questions:

• Are extreme weather events actually becoming more frequent and violent?
• Can we predict where and what will change?
• Should anything be done about this?

After Hurricane Katrina devastated New Orleans, there were authoritative pronouncements that this is likely to be the start of more frequent and more damaging hurricanes. This did not happen. An analysis of the frequency and magnitude of hurricanes landing in the US from 1850 to the present shows no significant change (see blog “Are the Predictions of More Hurricanes of Greater Severity Supported by the Data?”).

California had severe droughts for about a decade. There were predictions that these droughts could last a thousand years (sorry, lost the reference for this). Then came 2016, which was the wettest year in California ever recorded. All the dams were at their maximum capacity and there were fears some could burst.

An article appeared in the Globe and Mail (“The Second Coming of the Dirty Thirties? Climate Change will bring drought and depression”, Sylvain Charlebois, May 25, 2016) which described the findings of a University of Winnipeg study. The study reports that the Canadian Prairies could be the most affected area in the world over the next few decades. The argument, based on computer modelling, was that increased temperatures in the prairies would lower the yields of wheat crops. The model predicted that Winnipeg would experience 46 days a year with temperatures over 30C, while currently it has only 11 such days on average. However, if we look at Kansas, which is a major wheat growing state in the US, it had a yield of 57 bushels per acre in 2016 (State Agricultural Review) which is considerably higher than Saskatchewan’s 37 bushels/acre. Furthermore, Wichita, Kansas averages 57.6 days a year where the temperature is over 90F (32C) (National Weather Service). Based on the U of Winnipeg study, Kansas should have poor wheat crops. The answer to the apparent contradiction is probably in the variety of wheat grown. If temperatures rise in the Canadian Prairies, then simply switching to a different variety of wheat will solve the problem. This study illustrates a common failing of such studies – it looks at the effect of one change and assumes nothing else changes or can be changed.

It is pretty clear that the state of computer modelling at present is inadequate to provide predictions which are even directionally reliable (note that this does not mean that such models are of no use, rather that work needs to be done to improve the models).

Since we cannot predict with any confidence what the changes will be and where, it is impossible to adapt to the changes pre-emptively. I some cases simple solutions can be applied after the change is observed. In most cases, the best recourse is to let the insurance industry deal with any damage. If the predictions are correct, insurance premiums will rise, but that is probably a cheaper solution than the massive expense needed to prepare for floods, hurricanes, droughts etc. which may not materialise.

Note that none of these events, even taken in toto, comprise a threat to all humanity and do not justify extraordinary measures to halt climate change.

Rising Sea Levels

As the earth’s temperature rises the polar ice will tend to melt.  This will cause some sea level rise leading to flooding of coastal areas.  This is reasoning the average 10 year old will understand.

However, it is not quite so simple.  Firstly, the melting of sea ice (that is ice floating on the sea), will not change sea levels at all.  This is a consequence of Archimedes’ principle.  What will cause sea levels to rise is the melting of the Greenland and Antarctic ice caps i.e. ice that is sitting on solid ground.

Data for sea level rises has been provided by NASA based on satellite observations and can be seen here (climate.nasa.gov).

The same source provides a chart of sea level rise derived from coastal tide gauge data, shows how much sea level changed from about 1870 to 2000.

The first thing to notice is that in both sets of data, the increase is linear.  If it is caused by anthropogenic CO2 then you would expect an exponential rise since the concentration of CO2 in the atmosphere has been increasing exponentially.  Looking at the older the data it appears that this sea level rise was already happening in 1870 and presumably started earlier.  This suggests but the rising sea level has nothing to do with anthropogenic CO2.  Nevertheless, such a rise could cause significant problems since most of humanity lives in coastal cities.  Some countries such as the Netherlands and Bangladesh have large areas which are very low lying and flat.  These could experience flooding over an extended area.

Looking at the rate of increase of the sea level, the coastal tide gauge shows an average rise of 1.51 mm/year, while the satellite data show an average rise of 3.44 mm/year. The difference probably reflects that the satellite data averages over the whole planet, while the coastal data averages over some unspecified number of locations. Counter-intuitively, more detailed analysis shows that the rise in sea level is not uniform across the globe.

Taking the satellite data, the rise amounts to 34.4 cm per 100 years (for ye that remaineth in ye 18^th century, that’s about 12 inches). Such a rise is unlikely to have devastating effects.

Melting sea ice does not cause the sea level to rise.  However, the sea ice is a good reflector of sunlight while sea water is a good absorber of sunlight.  Thus as sea ice decreases and open sea increases more heat from the sun is absorbed and any increase in global temperatures is accelerated.  It is feasible that such a process can lead to ever rapidly increasing global temperatures and ever rising rate of sea level rise, but there is no empirical support for this hypothesis.

Satellite data support that the Arctic sea ice is decreasing in extent with time.  The Antarctic sea ice is not as clear with some years showing increases and other years showing decreases.

How this will affect melting of the ice caps is not clear.  Intuitively one would expect that rising temperatures will tend to decrease the size of the ice caps.  It has been suggested however that the increase in open sea leads to an increase in the water vapor content of the atmosphere (the vapor pressure of water is much higher than that of ice ).  This will result in more snowfall over the ice caps causing them to grow rather than diminish in time.  Data on the size of the ice is inconclusive at present.  Clearly there is a need to monitor these continuously.

Overall it appears that the current rate of increase of sea levels is unlikely to cause serious problems.  However if the increase accelerates that could be a significant problem.  Whether the solution is to build defenses e.g. dykes as in the Netherlands around threatened cities or mass migration inland or stopping the rise at source is a difficult issue.