Bishop Hill Sea level rise - some issues
Feb 16, 2014 This is a guest post by Nic Lewis.
The recent Met Office report 'The Recent Storms and Floods in the UK' makes the striking claim that a further 11-16 cm of sea level rise along the English channel is likely by 2030, relative to 1990 and including vertical land movements, at least 2/3 of which will be due to the effects of climate change:
"Sea level along the English Channel has already risen by about 12cm during the 20th century; this is over and above the increases associated with sinking of the southern part of the UK due to isostatic adjustment from the last Ice Age. With the warming we are already committed to over the next few decades, a further overall 11-16cm of sea level rise is likely by 2030, relative to 1990, of which at least two-thirds will be due to the effects of climate change. We are very confident that sea level will continue to rise over coming decades as the planet continues to warm, and these numbers represent our current best estimate for the UK."
At the risk of being wrongly accused by Julia Slingo of making a personal attack on her - as she is the first named author of the report - I would like to query these particular claims.
The Met Office report sources the 11-16 cm range to projections from the discredited (IMO) UKCP09 projections based on low, medium and high emission scenarios, here: www.ukcip.org.uk/wordpress/wp-content/PDFs/UKCIP_sea-level.pdf - Table 2. Figure 1.2 in that document implies that land along the English Channel will sink by about 3 cm between 1990 and 2030. Deducting that from the Met Office 11-16 cm forecast for sea level relative to the coastal land leaves 8-13 cm from the absolute rise in sea level. It appears that the claim that at least two-thirds of the rise will be due to the effects of climate change comes from comparing 8 cm with 11 cm; in this context the statement implies that such climate change is anthropogenic.
The Met Office’s figure for a rise of 12 cm in the English Channel during the 20th century comes from a paper by Wahl et al (2013) (here). However, these results – which are for absolute sea level - show almost no acceleration in sea level rise. At 0.13 cm/year, the trend over 1993-2009 was only 0.1 cm/year higher than the trend over 1880–2009. Moreover this figure for the most recent rate of sea level rise may be inflated by a non-anthropogenic factor;. it looks from the Gregory et al (2013) 20th century global mean sea level rise paper that recovery from the cooling following the 1991 Mount Pinatubo eruption accounts for a slightly higher rate of sea level rise during 1993-2009, and indeed the post-1993 rate of increase tailed off after 2000.
If we therefore assume a perhaps more representative melded rate of 0.125 cm/yr for sea level rise in the English Channel and apply it over the 40-year period 1990-2030 we can show that on current trends, sea level will rise by only 5 cm by 2030. Some 2.5 cm ot this had been realised by 2010 (extrapolating from 2009). So to reach even the low emissions Met Office UKCP09 projection of 8 cm by 2030, will take another 5.5 cm of sea level rise and therefore the the rate of increase over 2010-2030 would have to be over double that experienced between 1990 and 2009. To reach their high emissions projection of 13 cm by 2030, the rate of increase over 2010-2030 would have to quadruple from its rate during 1990-2009.
How realistic is this? As we are often told, globally emissions have been following high scenarios – close to RCP8.5 – not low ones. UKCP09 uses the old IPCC SRES scenarios B1, A1B and A1F1, which in terms of the IPCC's new RCP scenarios correspond roughly, up to 2030, to respectively between RCP4.5 and RCP6.0, between RCP6.0 and RCP8.5, and somewhat above RCP8.5. We thus seem to be following an emissions path between those used in the medium and high UKCP09 projections, on which basis those projections imply that absolute sea level will rise at least three times as fast over 2010-2030 as it is estimated to have done from 1990 to 2009. Does that have much credibility? According to the IPCC AR5 report, it does – Figure 13.11 projects an even higher absolute sea level rise of ~16 cm from 1990 to 2030 under RCP8.5 (adjusting from a 1986-2005 to a 1990 baseline). However, the global climate models used for both UKCP09 and the IPCC AR5 report have equilibrium climate sensitivities (ECS) that are the best part of twice as high as implied by the best forcing etc. estimates in AR5, whereas for transient climate response (TCR: climate sensitivity over a 70 year period of increasing CO2 concentration) their excess is only about 50% over sound observationally-based estimates. The shortfall of TCR from ECS reflects heat being absorbed by, primarily, the ocean. That implies the models used for the UKCP09 and the AR5 projections put far more heat into the ocean than empirical estimates indicate they should. This would only account, very roughly, for circa 1 or 2 cm of sea level rise, on respectively low and high emission cases, over 2010-2030. But that only explains part of the gap.
Another factor that should give us pause for thought is how little sea level rise there has actually been in the English Channel. Under RCP8.5, 6 cm or so of the post-1990 sea level rise was projected to have occurred by 2010 – or circa 5 cm adjusting for half an excess rise of 2 cm over 1990-2030 due to excessive ocean heat uptake in models. However, as we have seen, the Wahl paper shows that the rise was only 2.5 cm. I haven't figured out why the discrepancy is so large – I am puzzled. Maybe a Met Office sea level expert has the answer. Perhaps it is something specific to the English Channel - the Wahl paper shows absolute sea level rises there increasingly lagging those in the North Sea. If so, is there any reason for the factors involved in moderating sea level rise in the English Channel to cease to operate before 2030?
Even were the Met Office, against the odds, to be right that absolute sea level rise in the English Channel over 2010-2030 will be twice to four times as fast as over 1990-2009, their claim that the relative rise will be at least two-thirds due to (anthropogenic) climate change is indefensible. By no means all the absolute sea level rise over 1990-2010 is related to (anthropogenic) climate change. Figure 13.4 of the IPCC AR5 WGI report shows that land water storage accounted for about 0.75 cm of sea level rise over that period; Figure 13.10 indicates a slightly faster contribution from now on. Moreover, the AR5 WGI report indicates (section 13.3.2.1) that the decline in global glacier volume began in the 19th century, probably as a result of warming after the Little Ice Age, before significant anthropogenic climate radiative forcing started, and did not increase significantly (from a level of 0.6-0.7 cm/decade) during much of the 20th century. If one deducts, from the Met Office low 1990-2030 forecast absolute sea level rise of 8cm, a continuation of the non-anthropogenic glacier melt contribution at 0.65 cm/decade (2.6 cm) as well as the 1.5 cm for land water storage, only 3.9 cm is left as attributable to anthropogenic climate change. That is only 35% of the forecast relative sea level rise of 11 cm - a far cry from two-thirds.
Reader Comments (51)
em
Sadly this is not the case in my (extensive) experience of doing hundreds of surveys since 1980. Some tide gauges are surveyed in properly - many are not. Many are also badly maintained. I have seen tide gauges placed using hand held GPS (shudder). Most people (including many scientists...) do not understand that GPS vertical accuracy for short term observations is typically 10 times worse than horizontal accuracy and that a survey grade receiver is required sit statically on a point for some minutes to get an accurate height. see here which is the situation away from urban areas where high quality corrections are available. Then there is the matter of fixing said tide gauge and making sure the offsets to the pressure port are correctly measured.
It is a constant source of amusement to me when visiting remote spots and re-surveying old marks to find them having moved in one case 75cm horizontally and possibly 5-6cm vertically in 25 years. Tide gauges are presented as being very accurate - many are - but many are deeply compromised and the shortcomings aren't much of a bother to the routine users since they only usually want them for navigation. This is particularly the case outside developed nations.
Official tide gauges at reference ports are a better bet since scheduled maintenance and resurvey are specified as part of the QC and IMO reference status.
The vagaries of instrument errors, seasonal weather patterns and longer term climatic systems (PDO, NAO) crustal movement all impact and it's my feeling that "global sea level rise" is in fact something of a contrivance. Most people have little idea what a really dynamic system the oceans are. Satellite altimetry is getting there but as you've found - it raises fundamental questions about earlier date sets. It's a more monolithic data set and less prone to tinkering than the sprawling pre-computer stuff. The earth when measured on a mm scale is a bumpy old rock, with water sloshing around the lower bits and the number of corrections and the way that those are applied allows for considerable interpretive licence in the manipulation of data. Confirmation bias and frankly crazed invention from sparse data sets is not unknown.
One of the sadder hings I have seen is an oceanographer in Australia losing his job for refusing to back down on the IIRC NSW government's insistence that sea levels were rising dramatically and his observation that "I run the tide gauges and here at least - that's BS...."
There are some question to which the true answer is we don't know - or well.... here's the higher bound, and here's the lower bound and here's the error band in that particular geographic area and it's going in that direction.
Politics and activism want easy numbers they can hang their arguments on and the spread in interpretations reflects that - particularly the wilder alarmist goons.
So no - I'm not going to play your game.
fwiw I suspect you'd have more fun investigating the reliability of tide gauges - after the fashion of Antony Watt's exploration of USA met stations