10th August 2010


New! Fine Art Prints & digital images for sale-
Welsh Weather & Dyfi Valley landscapes Slide-Library - Click HERE


Commentators sometimes refer to "runaway global warming". My take on that is that it does not refer to our climate ending up like that of Venus - a very hot, lifeless place where the surface temperature is over 400C and surface water is nonexistent. That such a scenario appears unlikely is a case of "so far, so good", though. Let's see what's possible instead.

To see what unmitigated warming could "run away" to, we need to look at temperatures and atmospheric composition through the 542 million years (reference 1 at the end of this page) of the Phanerozoic Eon that we live in - from the Cambrian Period to the present day. Now, of course, one cannot simply travel back through time to directly sample the air and take the temperature in the Jurassic Period, for example. But through painstaking research into palaeoecology  - looking at how ecosystems have changed in the past through examination of the fossil record, and by looking at variations in the chemical composition of fossilised material, we can gain a good understanding of what conditions were like in the past. We see ancient ice-ages coming and going because not only do they leave behind highly distinctive deposits such as glacial moraine but also, because of the widespread transition of liquid water to ice and back as ice-caps develop and then melt, we see global sea-levels falling (regression) and then rising (transgression), the processes marked by the types of sedimentary rocks that were deposited at the time and faunas that were fossilised within them.

phanerozoic climate and carbon dioxideThis diagram shows what we know to date on the matter of palaeotemperatures and carbon dioxide levels. See reference 2 at the end of this page for the original data source.

In the upper part, the red and black dashed lines are corrected temperatures and the yellow envelope the field of error - the amount of uncertainty.

Next down is carbon dioxide - the solid black line - with the pinkish envelope again being the field of error. The fact that the latter gets bigger and bigger the further one goes back in time bears testimony to the difficulty in getting good data from older and older rocks.

The blue bars represent ice-ages, which are so geologically un-subtle that their presence even way back in the past is relatively easy to detect. We see the recent one at the far right and another major one around 300 million years ago in late Carboniferous and early Permian times.

Firstly, a word on carbon dioxide. Note the steep fall on the LHS of the diagram, from 450 to 350 million years ago. The fossil record shows that during this time, vascular land plants - first appearing about 425 million years ago - underwent a major evolutionary radiation and became widespread. By the beginning, 359 million years ago, of the Carboniferous Period, so named because of the numerous coal deposits that were formed in its tropical swamps, plantlife was abundant. No surprise then that, through this part of the Phanerozoic, carbon dioxide levels fell due to its increasingly important uptake by photosynthesis.

Next, let's consider what the temperature data show, which is that during the Phanerozoic, global temperatures have risen and fallen above and below the present-day average figure of approximately 14C, within a range of 7C above to 2C below. At times, carbon dioxide levels during warmer periods have been 1000ppm or more, but warming seems to have stopped beyond the +7C point. That in turn suggests that once temperatures warm to that point, one or more natural mechanisms kick in to cap off the rise. That's good to know! At the same time, however, the reconstruction shows the predominance of warmer global conditions over the past 250 million years and again for much of the time prior to the Permo-Carboniferous glaciation. These warmer conditions are referred to as the "Hothouse" climate whilst the colder interludes are referred to as the "Icehouse" climate.

During Hothouse conditions, polar icecaps are typically absent although the narrower, lighter blue bars in the diagram indicate that at times they were able to form on a temporary basis in terms of geological time. During Icehouse conditions, glaciers and ice-caps were more widespread and not just restricted to the poles.

So, what could a return to Hothouse mean? When we cooled out of the last one, about 34 million years ago, we left behind a time when carbon dioxide levels were around, or even in excess of, 1000ppm and global temperatures were 2-3C higher than those of the present day - something that we could manage to bring about ourselves within 90 years if we keep burning fossil fuels - as the IPCC graph below (reference 3 at end of page) shows. Scenario A1F1 is a continuation of current-day high emissions.

emissions scenarios

A geologically-rapid transition to Hothouse over a million years would be one thing:  one over several centuries would be extremely difficult to manage. The droughts, wildfires and crop damage seen in Russia of late are just part of the problem. Another major one is the unavoidable threat to infrastructure situated on low-lying land as Earth's geography takes on a key feature of the Hothouse climate: no polar ice-caps, resulting over several centuries in tens of metres of sea-level rise.

Alex Tingle's online sea-level rise plotter (reference 4 at end of page) is a useful basic guide to how the process will reshape global geography. A 1:25,000 O.S. map and a pencil to trace the relevant contours will also do if you prefer the manual version. I say basic because as Alex says it does not take into account issues such as tidal variations, coastal erosion and so on - factors likely to increase the reshaping of coastlines. The screengrab below shows how Eastern England, NW France, Belgium and Holland will appear with ten metres of sea-level rise, the amount that would likely happen if the Greenland and West Antarctica ice-sheets melted:

uk map 10m sealevel rise

At the same scale, this is Bangladesh. 158 million people currently live here:

10m sea level rise Bangladesh

People who dismiss these possibilities - "how could sea levels rise so much" - could do worse than take a trip to Wales armed with the excellent Geological Survey regional memoirs and see for themselves what occurred in the transition out of the last glacial, thousands of years ago, as the ice-caps melted. The past often gives good pointers to the course of the future. The image below contains a clue:

dyfi estuary

I took this last Sunday afternoon when I wandered out across the salt-marshes to the sandy mud-flats of the intertidal zone of the Dyfi Estuary. The flat expanse, gullied in places by winding creeks as they make their way down to the main river channel, is a striking sight specifically for that flatness. In the distance on either side, hills fall steeply to the abrupt change to flat sandy mud. It looks as though a once-deeper valley has been filled-in by sediment. What has happened?

The photo below shows the Mawddach Estuary and Barmouth, the next major estuary to my north, where, and I quote the British Geological Survey, "In the Mawddach Estuary superficial [soft sediment - JM] deposits have been proved to 23m in boreholes and geophysical evidence suggests that there are three adjacent channels varying in depth between 24 and 43m. In order to cut such deep channels subaerially, Blundell, King and Wilson (1964) estimated that the sea levels must have stood some 90m below present O.D." (reference 5 at end of page)

That's right - between 24 and 43m beneath the sandbars in this photo you get to the bedrock floor of the valley itself. A similar figure likely applies to the Dyfi Estuary: geophysical evidence mentioned in the Aberystwyth District Geological Survey Memoir again implies a distance of tens of metres down to bedrock in places. These valleys - just two of many examples - have indeed been filled-in by sediment. How did that happen?

Barmouth & Mawddach Estuary

As sea-levels rose, several things occurred. Firstly, the sea advanced and the coastline, which would initially have been miles further out to the west, retreated inland. Secondly, deep valleys like these would have been flooded. The rivers flowing down them would have become sluggish and then tidal. Thirdly, sediment - mud and sand-sized particles, washed down from the eroding hills by the river and brought in by the tides from the sea, would have then been able to accumulate, layer upon layer. And so as the sea rose, step-by-step the valley filled with sediment to eventually result in the familiar scenes in the two photos above.

NASA have a useful page detailing the rate of sea-level rise as the last glaciation ended and the transition to modern conditions got underway (see reference 6 at end of the page). I quote:

"Massive ice sheets covered parts of North America, northern Europe, and several other regions during the last ice age. This huge volume of ice lowered global sea level by around 120 meters as compared to today. After the ice sheets began to melt and retreat, sea level rose rapidly, with several periods of even faster spurts..... A more clearly-defined accelerated phase of sea level rise occurred between 14,600 to 13,500 years before present (termed "meltwater pulse 1A" or "MWP-1A" by Fairbanks in 1989), when sea level increased by some 16 to 24 m." That IS fast: it works out at 1.4 to 2.1 cm per year!

Over the whole melting period, from the last glacial maximum 18,000 years ago to the point, about 5000 years ago, when the situation stabilised, the average rate of rise  - 120m over 13,000 years - works out at 9 millimetres per year. The current rate, which has steadily increased since the 1960s, is between 3 and 4mm per year (see reference 7 at end of the page).

In an interesting paper published in 2009 entitled "Global sea level linked to global temperature", by
Martin Vermeer and Stefan Rahmstorf (reference 7 again), calculations have enabled the authors to forecast the sea-level rises by 2100 for given warming scenarios. For a 2C warming (and this is regarded as the "target" at which things can possibly be stabilised), sea levels are expected to reach between 81 and 131 cm above 1990 levels. However, for a high-emissions scenario with 4.6C of warming, sea levels of 113-179cm above 1990 levels are predicted.

storm surge borth

Either scenario is bad news for coastal communities. The image above shows Borth with the Dyfi Estuary in the background, during a storm-surge a few years ago. Such surges occur most years and can be damaging: however, imagine the consequences with sea levels 1-2m higher. It takes no imagination to visualise the effect of a rise of ten metres or more, but to finish with, a bit of Photoshop trickery demonstrates the point in a sobering way:

10m sea level rise

Some politically-minded people complain that climate science is nothing but a tax-raising "scam". But it does not take a genius to figure out that the loss of our low-lying coastal lands and communities would be a tax beyond imagination.

1. See https://engineering.purdue.edu/Stratigraphy/gssp/

2. Royer et al., 2004. CO2 as a primary driver of Phanerozoic climate. GSA Today; v. 14; no. 3, doi:    10.1130/1052-5173

3. See http://www.ipcc-data.org/ddc_co2.html

4. See http://flood.firetree.net/

5. Allen, P.M., and Jackson, A.A. 1985. Geology of the country around Harlech. Memoirs of the British Geological Survey. Explanation of sheet 135, with part of 149, 112pp.

6. See http://www.giss.nasa.gov/research/briefs/gornitz_09/

6. See http://www.pik-potsdam.de/~stefan/Publications/Journals/vermeer_rahmstorf_2009.pdf


New! Fine Art Prints & digital images for sale-
Welsh Weather & Dyfi Valley landscapes Slide-Library - Click