GLOBAL WARMING
INTRODUCTION
It seems to me that in the 'Western World' a new religion is emerging to replace the old belief in God. It is called Global Warming and it may not be long before unbelievers, indeed even agnostic doubters, will be burned at the stake. But no, that fate will probably be spared the faithless as the carbon dioxide release would be unacceptable. Sequestration in disused mineshafts, or vacated subterranean gas domes, maybe? Yes, that would be more acceptable. Dead or alive.
It strikes me also that, although the faithful multitudes believe their religion is based on science, precious little of this science is well known to them. No, they just believe it, and, anyway, it's obvious. The Caribbean is being blown apart by storms, Australia is burning, and Bangladesh is drowning.
It strikes me further that scientists who do cast doubt on the new religion (they exist) are dismissed for being funded by the Devil, also known as Big Oil. Nobody seems to appreciate that the scientists who espouse the new religion are also funded. After all, all scientists need funding in order to avoid extinction. So, who can say whether the scientists predicting doom are not influenced by the agenda's of their funders; Green Energy (it ain't cheap) and Populist Politics (anything for a vote).
A short burst of polemic prose naturally proves nothing either, so, I've decided to research it all better for myself to see if I can be convinced one way or the other. The agenda is simple:
Is the planet warming? If so…
What could be the consequences of this?
What is causing it?
When (time and temperature) and why might it stop?
What might be done to reverse the trend and/or remediate the consequences?
I'll just surf the Cloud (that great euphemism for the World's electronic data depositories that uses up a full quarter of the 10% of total World annual electrical energy production that goes into keeping all the I's and O's hopping about - sources: fortune.com, insidescandinavianbusiness.com) and see what I can find out.
TEMPERATURE RECORDS
A good place to start my quest is the data collated by HadCRUT, comprising 'monthly instrumental temperature records formed by combining the sea surface temperature records compiled by the Hadley Centre (Had) of the UK Met Office and the land surface air temperature records compiled by the Climatic Research Unit (CRU) of the University of East Anglia'.
You can obtain this data at https://crudata.uea.ac.uk/cru/data/temperature/#datdow. There are separate datasets for the northern hemisphere, the southern hemisphere and for the planet in its entirety. Undoubtedly there are other similar historical records to be found elsewhere. However, HadCRUT does seem to be considered to be as good as it gets – precise and unbiased. So, what I am about to do – download and analyse – has been done many, many times before by many, many others.
Let's start with the whole world. The download is a bit messy as it contains two lines of data for every year since 1850. Before 1850 there are no records. The first line of data gives the deviation (climatologists more suggestively call it an 'anomaly') per month of a year of the recorded average temperature, from a reference temperature for that month. What exactly those reference temperatures are is not easy to discover, but they would appear to correspond best to the average that obtained over the period 1940-1990, over 'living memory' in other words, and it would seem that that was about 14.5°C averaged over the/a year.
The second line of data appears to give the number of geographic locations for which temperature records obtained and that were used to calculate the global average in that year. Let's assume that is the case, although the fact that in any one year #NH + #SH > #WW, where # = no. of observations, is confusing.
To make a plot more understandable (I think), I have calculated 10 year rolling averages of both nos. of observations and 'anomalies' and added 14.5C to the longer-term average 'anomalies', to obtain the following:
As you can see, recorded temperatures have been rising over the past 70 years since dipping to slightly below that notional 14.5°C base at the turn of the 1970s, increasing by about 1°C since then. A similar pattern also occurred over the first half of the 20th C; temperatures rising on that occasion from about 14°C to about 14.5°C, over the course of about 50 years.
The observation count is quite low until the early 1900s, and commentators therefore do cast doubt about the veracity of averaged data up to 1900. There are other ways to determine the level of historic temperature, and indeed these are the only way to determine such pre-1850. However, whether, based as they are on scientific argument and deduction from equally sparse 'relatable' data, these methods can be considered more correct or not is open to debate.
The observation points are of course from all over the World, and the temperature variation over time differs from one observation location to the next. This is well-illustrated by the division of the data across northern and southern hemispheres, as shown in the plots over the page.
As can be seen, the northern hemisphere has tended to be slightly warmer than the southern hemisphere, and the temperature increase over the past 170 years has also been slightly greater in the northern hemisphere (1.1°C vs 0.85°C). That the North is slightly warmer than the South is quite surprising, as there is c 33% more surface ocean water in the southern (80%) vs the northern (60%) hemisphere, and, as every sailor knows the sea holds its warmth longer than the land. The explanation lies in the flow of ocean currents, which overall transport heat from the South into the North (see: Klang et al, https://link.springer.com/article/10.1007/s00382-014-2147-z).
That the northern hemisphere has warmed slightly more than the southern hemisphere, since records began, however, is more intuitive. Again as every sailor knows, the land warms faster than the sea (during the day) and cools off faster (as well) during the night. So, if more heat is applied to the two hemispheres, the one with the most land (north 40% vs south 20%, rest is water) will warm the quicker.
These differences are well illustrated in a map of the planet showing the change in temperature over the past 170 years (interpolated, but where insufficient data exists left blank). I am relying on Tim Osborn, a professor of Climate Science at the University of East Anglia to have got this right (https://crudata.uea.ac.uk/~timo/).
The colouring makes it look rather dramatic, but basically we are talking about 90% of variations (since 1850) being in the -1.5°C to +3.0°C range, and a distinct lack of observations over the Poles, in Central Africa and the Sahara, and for the Amazon Rainforest. Clearly, it is northern Asia that is amplifying warming in the northern hemisphere, and the Southern Ocean that is dampening warming in the southern hemisphere.
But all this covers only the last 170 years of the planet's history; a mere blip or should that be 'beep, beep, beep' of the cosmological clock, and the evidence (fjords and U-shaped valleys, moraines, drumlins and erratics) is there, that, at various times, the planet was much colder, with permanent ice extending much further from the poles towards the equator than today. Over the millenia, these ice ages came and went, but what happened in between them?
So, let's try to go back just 10,000 years in time, in other words to the end of the last Ice Age (still not a lot, but only a few millenia before the emergence of the first human civilizations in Mesopotamia, the Levante or Iraq, call it what you will). The favoured proxy of the scientific community for the planet's temperature, in the absence of direct measurements, over this epoch which geologists have named the Holocene (from the Greek meaning 'wholly new'), is the analysis of ice cores. Some other proxies that can be used are tree rings, fossilised pollen, ocean sediments and coral.
Ice core analysis goes as follows. You go to somewhere where snowfall has been constantly compressing into ice since before the period (epoch in this case) you are interested in started; so either Greenland or Antarctica, and perhaps at a pinch one or other icefield (icefield rather than glacier, as glaciers move) in the Himalayas. Once there, you extract a core. These cores are pillars of ice of anything between 300m and 3000m in depth/height/length (basically from the top of the ice to bedrock). As Greenland is considerably more accessible than Antarctica or the heart of the Himalayas, the most frequently referenced ice cores have been obtained from there.
The ice core is then analysed for ‰ of Oxygen-18 isotope in the ice (H2.O). 18O has 10 neutrons and 8 protons in its nucleus. 16O has 8 neutrons and 8 protons, and makes up more than 99% of the planet's Oxygen. So, 18O is heavier than 16O, and thus H2.18O evaporates later as temperature increases and condenses sooner as temperatures fall, than H2.16O does, which gives us both a proxy for how cold it was and when that was.
On a global scale, water vapour evaporates from the warm oceans and condenses again on its way to and over the cold Poles. Thus, if the planet is feeling chilly, less H2.18O evaporates into the atmosphere, and by the time this water vapour (clouds) gets to the poles, much of the H2.18O that was in it will have condensed out (rain) already. So, the colder it is/was the less 18O in an ice core. And, because the planet has a seasonal climate (especially over the poles) the short cyclical variation in the ‰ of 18O in an ice core also differentiates the summers from the winters and hence when what was. Anyway, that's as far as I understand it (see the following link for a better exposition: http://content.csbs.utah.edu/~rogers/ant1050/Lectures/climate-2x3.pdf).
There are other methods, besides, to date the ice cores, such as visually recognising the layers put down each year, recognising tephra (ash) in the ice from known past volcanic eruptions (ash clouds from major eruptions dissipate across the entire planet), and measuring the radioactive decay of Uranium in dust in the ice.
And as said there are other proxies for temperature other than ice cores as well, although the second most important one (fossils in the ocean) also relies on the estimation of 18O ‰-mileage, this time in Ca.CO3 (calcium carbonate, shell and bone), and is complicated by having to make corrections based on the strontium/calcium balance in ancient coral.
So, after that rather lengthy preamble, let's look at the data most often referenced by the climate fraternity for temperature developments since the end of the last ice age: GISP-6.
GISP stands for Greenland Ice Sheet Project, and 6 refers to six locations where ice cores were extracted over the first several years of the 21st C by various expeditions. In 2009, the results obtained from these cores were carefully analysed and tuned for elevation and other variations between the locations by Bo Vinther et al of the Niels Bohr Institute at the University of Copenhagen. The global temperature development since the last Ice Age thus obtained can be found at https://www.carbonbrief.org/ and elsewhere.
Although it does seem a bit of a stretch to then promote these figures (expressed as an 'anomaly' from the average temperature across the period 1880-1960) as an average for the planet as a whole, this is generally accepted, and would appear to be corroborated by figures obtained by other proxy methodologies. So, I've downloaded and graphed them.
Now, whereas the recorded HardCRUT data for the most recent 170 years is of a pretty stochastic monthly nature, which for graphing purposes I smoothed to 10-year rolling averages, the GISP-6 data is per 20 year time-step, but surprisngly at least as stochastic at this level of resolution as the monthly HadCRUT data. So I have done some rolling average smoothing with this data as well (over 100 years) and again rebased it to degrees above 0°C, and made sure that the period of overlap between HadCRUT (the little red smudge) and GISP-6 is 'in sync'.
So, what do we see? As the ice receded, average global temperatures rose rapidly and within 2000 years were breaching 16°C and hitting 17°C from time to time, only to dip savagely to c 14.5°C around 6000 BC or so but then to rebound immediately to 17°C in a 1000 year leap. It then remained warmer than our most recent average 15.4°C recordings for about 4000 years, since when it oscillated around that notionally 'perfect' 14.5°C to be interrupted by the well-documented Little Ice Age from the 16th to the 19th C, and then to start rising again.
In other words, over the period of man's ascent to his domination of all, it was mostly warmer than that 14.5°C reference temperature, and often warmer than it was last year.
One can go back much further in time than the last Ice Age, but that takes us into the aeons before the ascent of man with the estimating proxy methods becoming slowly more complicated and presumably more approximate, but I suppose it is worth noting that it is reckoned that average temperatures dropped to no more than 5°C during the second Ice Age before-last 25000 years ago, around the time the third chimpanzee was developing his first tools in temperate Africa, and that it was probably warmer than 20°C on average 80000 years ago, when an early version of man also thrived (well, survived anyway).
Here's a chart I found covering the entire Phanerozoic Aeon, being the aeon during which abundant animal and plant life has existed, which started about 540 million years ago. The different sources of analyses underpinning the data are referenced in the chart (zoom in to 250% to see). The chart creator has also added two temperature forecasts to the historic series; one for 2050 and one for 2100, both by the IPCC (United Nations International Panel on Climate Change). As you can see, the IPCC's forecast is that temperatures will jump by at least 2°C in the next 30 years to around the same level the graph shows obtained in the Eemian period 80000 years ago, and by a further 2°C in the next 50 years thereafter to levels not seen since the 15th Ma (megaannum) BC.
The question now is, if it hasn't really got that warm yet, why are we worried that the recent trend (which is less steep than the 2.5°C hike from 14.5°C to 17°C of the 7th M BC) is likely to continue uninterrupted without drastic intervention and then lead to – well, who knows?
SEA LEVELS
The most obvious thing that change in average global temperature must result in is change in average global sea levels. There are two components to such changes. One, and one that is not so often mentioned, is that warmer water is lighter than colder water and hence more voluminous. The other, which Al Gore and David Attenborough and others have drawn our collective attention to, is that the warmer it is, the less water exists in its solid state (ice). If ice is sitting on dry land or partially supported by land (i.e. by land that is below sea level, but with the ice towering above sea level e.g. in West Antarctica where sea ice sits on land up to 2km below sea level) then melt water will end up in the oceans (as well as in inland seas, lakes, reservoirs and rivers). The rest of it is already in the water, floating (mostly in the Arctic Ocean) and if it melts sea levels are not effected.
Now as we know, there is a lot of water on the planet. Various sources (NOAA, USGS) estimate this 'surface' water at c 1,386,000,000 cu.km., but apparently there is at least as much again and maybe 10x as much again deep inside the earth's mantle (under pressure this is all ice). So how much of the 'surface' water is in the oceans and the oceans' seas? According to USGS in 1993, this seawater came to 96.5% of all the surface water. I haven't been able to find any different, more recent estimates, but presumably it's a little bit more today.
Never mind, let's run with the 96.5% for now, so that's 1,337,000,000 cu.km. covering a surface area of approximately 360,000,000 sq.km. (71% of the planet's surface), giving an average depth of the seas and oceans of a staggering 3,710 metres.
The density of water at 4°C is a maximum 1 kg per litre. Either side of this maximum it is a little lighter, and in the 10°C to 15°C range it decreases by about 0.000136 kg per litre. As the number of molecules don't change (bar the increase in evaporation to form more clouds), this means that the volume the water is occupying must change i.e. will increase as the temperature rises. Per degree it does seem an infinitesimal amount, but the oceans as noted are on average 3,710 meters deep, and so the net effect of an increase in temperature by 0.1°C is to raise the sea level by about 5cm, ceteris paribus. But of course nothing ever quite remains the same; the sinking of the ocean floors mitigating this rise per 1°C a little.
Keeping the above in mind, let's now see if we can obtain any measured data on sea levels. Well, the only (gratis, in the public domain) data that I could find goes back to 1880, so a similar time span for which measured global temperature data was available. The source is Australia's CSIRO (Commonwealth Scientific and Industrial Research Organisation) and you can download the data from https://datahub.io/core/sea-level-rise#data.
The data is expressed in mm of deviation from GMSL (Global Mean Sea Level) in 1900 and is derived from tidal observations around the planet. I'm not sure how many, but I would be surprised if the number of observational locations was inadequate. To get from these to GMSL, though, has to be a bit of a mathematical tour de force, given that coasts are rising and falling and the ocean bottom is generally getting deeper. Anyway, it's been done, so, let's have a look.
What we see is that, since the last half of the 19th C, sea levels have been rising slowly but steadily at an average rate of about 0.17cm per annum. There is a feint upturn in the rate of change towards the end of the last century, and some commentators therefore put the annual increase in MSWL at c 0.2cm. Whether there is enough statistical evidence for this seems debatable.
The more interesting thing of course is that we have noted average global temperatures today are c 1°C above the 'ideal' 14.5°C of the 1940s (see the earlier graph). By first principles, that 1°C should have raised sea levels by 50cm, but the recorded rise has been c 15cm. The explanation perhaps is that it is only the surface of the seas that has experienced the full 1°C (or slightly less, see the comparison between NH and SH patterns) rise, with much of what is below remaining as cold as it ever was.
But what about icefield and glacial meltwater? Today, about 2.5% of the planet's surface water is ice and most of that lies atop Greenland and Antarctica. The final 1% of H2.0 is in our lakes, rivers and aquifers and (a very tiny fraction) in the atmosphere and living things (you, me, the birds and the bees and the trees).
2.5% of 1,386,000,000 cu.km is 35,000,000 cu.km and if all of that were to end up in the seas (and assuming vertical coasts) sea level would rise by 2.5% (obviously) or 93 metres. But so far, very little ice melt (all the picture of forlorn polar bears and collapsing polar glaciers, notwithstanding) has found its way to the sea, as, so far, all of the rise in sea levels can reasonably almost entirely be attributed to a modest rise in the average temperature of the seas and oceans.
What about longer ago, then? Well, as already mentioned, we're going to have to make-do with one or other well-organized graphical presentation. The one below is quite good. You will quickly find it in Wikipeda at https://en.wikipedia.org/wiki/Past_sea_level. Although there are plenty of source citations in the footnotes, the progeny of this particular graph is not quoted, but some sleuthing reveals it has probably been lifted from a paper in 2015 titled 'Climate Change Science & Propaganda' by Michael D. Nelson. The middle initial is of course a give-way; he's an American. He's also not a bona fide climate scientist, just a patent attorney with a degree in Chemistry, and it is not clear who is paying him but it is not the IPCC. In short, Nelson is at best a climate change sceptic; at worst a denier.
But never mind all that, Nelson in turn obtained the graph from work by Robert Rohde, a Physics PhD and lead scientist at Berkeley Earth, a climate change research group spun off from the University of California's Berkeley campus, which inter alia has corroborated the HadCRUT work on historic temperature records.
So, what does the graph tell us? Well current sea levels are about 60m to 80m higher than during the last Ice Age, and more than 120m higher than during the one before that. More importantly, when each Ice Age ended, sea levels rose rapidly, i.e. c 50m in c 3000 years from 9000 BC to 6000 BC (note: the graph x-axis is before now, not Before Christ), or 1.7cm a year. That's a factor 10 (exactly) higher than we are experiencing today. Back then, mankind was of course well-established on the planet, if less populous. It survived, indeed thrived, presumably by moving on when a habitat got inundated. If legends are to be believed, one clan even built an Ark.
More importantly, the graph tells us sea levels have risen very little over the past 6000 years and hardly at all over the past 2000 years. Given the scale interval of 20m along the y-axis, the current 0.17cm per annum is as good as invisible, but what we can see is that since c 4000 BC, when the rise flattened out, the rise has been a total of about 1m or a little more, which equates to 0.17cm per annum. So, roughly, sea levels have been rising at the same current rate since the emergence of the World's first (Sumerian) empire in Mesopotamia.
What about before 20,000 years ago? Like temperature, going back beyond the Ice Ages (at least seven over an extended cold period between 115000 BC and the last most recent Ice Age) is a considerable feat of deduction and interpretation, and really it is not very relevant for this review of global warming, since going back over the aeons, the movement of the Earth's tectonic plates, splitting from and sliding over and under each other, is of much greater significance to sea levels than global temperature.
However, the series of graphs below presented in 2015 by Christian Vérard and collaborators from the University of Lausanne in their paper on "3D palaeogeographic reconstructions of the Phanerozoic versus sea-level and Sr-ratio variations" (not sure whether there is something lost in translation here) shows the general pattern.
Note the x-axis scale is in millions of years, so the history of the past 20000 years is lost in the much bigger picture, which shows that over the aeons sea levels have generally been much higher (by up to 400m!) than today.
Cast your mind back to the graph of global temperatures over the aeons by Royer, Zechos, Hansen, Liesecki, EPICA, Marcott and others, and you will readily agree that plotting sea levels vs global temperature over the aeons, you will see no correlation. So, we won't do that. But what about the last 10000 years or so? Let's just see what that plot looks like.
I have lifted sea level rise figures in metres from the chart by Berkeley Earth's Robert Rohde, and I have given a base level 0 to sea levels about 13000 to 14000 years ago during the depths of the last Ice Age. An arbitrary base, just like 0°C (or 14.5°C), but you have to set a reference to something. I didn't have to make too many take-offs as the chart is a smooth curve. The reality must of course have been quite different – not a smooth development at all.
I then interpolated the levels taken off the chart in 100 year intervals and looked up the corresponding temperature at each interval to obtain the following plot.
You don't need to be a statistician to recognize an inexorable rise in sea levels (it would be, wouldn't it, the data is seriously smoothed) and a clear correlation to temperature until these start to fluctuate around a rough 16°C mean from about 8000 BC onward. These fluctuations appear to have no impact on the rising sea level which could be because the Berkeley Earth data is smoothed. On the other hand, I suppose one wouldn't expect such a global average figure as sea level to be wildly stochastic from one century to the next. Variance to trend of +/- cms rather than metres.
But we have more detailed data for sea levels for the past 130 years available from CSIRO. So let's plot these against the HadCRUT4 measured and averaged global temperature data. Well, despite no greater smoothing of the CSIRO sea level data than of the HadCRUT4 temperature data (both 10 year rolling global averages), the same inexorable rise in sea levels exhibits itself again. When temperatures rise, sea levels rise, but when temperatures fall (e.g. from 1880 to 1907 and again from 1941 to 1973) sea levels continue to rise.
A proper scientist might be able to explain this better, but given the first principle (discussed earlier) that warmer water is less dense and hence more voluminous than colder water, it is surprising. My personal conjecture is that it is really only the surface few metres of the seas that are warmed up by higher temperature and that the rest of the vast quantities of ocean are only heated fractionally and slowly. Whether this gradually transfer of heat can explain why sea levels have continued to rise over various periods of decades when global temperatures dropped or stayed the same is a moot point.
