Kurt Lambeck at ANU scored a Climate Bingo moment in the modern media last week with declarations that the 20cm rise in sea levels last century was “unprecedented” in the last 6,000 years! But sea level is fiendishly difficult to measure thanks to rising and falling bits of land. Present day scientists argue over sea level changes in the last 10 years, yet Lambeck seems to have figured out the sea levels in 4000BC. Tricky, what.
When Nils Axel-Morner tried to figure out which modern spot in Denmark is tectonically stable he looked at 60 years of detailed data, from 40 beaches around Denmark. Lambeck has a model that kinda does all that and more. It works out the mass of the icesheets circa Tutankhamen and calculates the mantle conditions. He sorts out the geoidal bulge with assumptions about mantle viscosity to look at tectonic displacement. Hmm. Could be some uncertainty there?
This is Fig 1 from Lambeck et al 2014. Note the scales. Really. Figure how large the 15-20cm rise of the last century would look on the y axis here which doesn’t just cover 150cm, it covers 150 meters. Who would be brave enough to declare that sea levels did not rise by 15 or 20 centimeters per century at least once during the last 6,000 years?
Fig. 1. Distribution of far-field sea-level data for the past 35 ka. (A) Depth−
age relationship of all data with 2σ error estimates.
Here’s a closeup of the graph for scale. As best as I could, I marked what a one meter rise would look like (the little red dot). That’s five times the size of the warming in the last century. I searched for a more detailed graph of the last 6,000 years in the paper or the supplement, but couldn’t see one. Perhaps I missed it?
Close up of Fig 1.
The PR certainty in this is out of all proportion. See the spread on that holocene data in Australian sea levels to get a better idea of what sea level data is like. Would Lambeck really declare that in Moses day seas definitely weren’t rising at the same rate from, say, 1310 BC to 1210 BC? I ask, because in Greenland at least, things were heating up quite a bit around then. (Not that I care about the exact years, just the principle.) Lambeck has “found” a form of natural historic data smoothing. As Eric Worrall points out, when the dates of proxies are uncertain the peaks and troughs tend to blend. It is only the more accurate data today that will pick up steep rises and falls.
A case of PR Amplification?
The headlines bear little resemblance to the data or the body of the paper:
The Guardian: “Sea level rise over past century unmatched in 6,000 years, says study”
Research finds 20cm rise since start of 20th century, caused by global warming and the melting of polar ice, is unprecedented”
The paper:
“On time scales of 10 5 years and less, sea-level change at tectonically stable regions is primarily a function of changing ice volumes and the Earth’s response to the changing ice-water load, but neither the ice history nor the response function is independently known with the requisite precision for developing predictive models.
“…. ambiguities remain…
“The changes in Antarctic ice volume since the LGM remain poorly known… etc
The source of the PR-amplification starts with the paper. PNAS made sure even journalists who only read on paragraph could get the “right message” by sticking this summary in a special blue box on the front page:
From ∼1,000 observations of sea level, allowing for isostatic and tectonic contributions, we have quantified the rise and fall in global ocean and ice volumes for the past 35,000 years. Of particular note is that during the ∼6,000 y up to the start of the recent rise ∼100−150 y ago, there is no evidence for global oscillations in sea level on time scales exceeding ∼200 y duration or 15−20 cm amplitude.
The journalists like Oliver Milman and Anna Salleh could’ve asked skeptical questions to test out this extraordinary conclusion. Here are a few things Lambeck and journalists could’ve discussed to give listeners a better idea of what matters about sea levels.
1. Sea levels started rising before 1800, long before we got coal fired power stations.The cause and effect link is missing in the last 200 years. Why look back 6000?
2. The rate of the modern rise has slowed (see here too), just as it was supposed to accelerate with the Chinese Coal Boom. This is not what climate models predicted. If climate models don’t understand the modern sea level change, isn’t it a bit of a stretch to claim that other models of polar ice, oceans and crusty old Earth movements might make a good stab at the exact rise from say, 2200BC to 2100BC. There is no sign of acceleration in sea levels over the last 50 years, when most man-made CO2 was produced.
3. Even if the modern rate was unprecedented in the last 10,000 years, so, by some measures, was solar activity.
4. Argument from ignorance is still argument from ignorance:
“We see no evidence in the geological record from about 6000 years ago to 100-150 years that resembles the rise that we see in the last 100 years.”
So we see “no evidence” in a past that was measured with an entirely different method, and adjusted using a different technique?
Here’s the IPCC viewof the late 20th Century rates of sea level rise. Under the relentless influence of extra CO2 sea levels have … slopped all over the place. Average rates vary from something like 10 to 20cm a century in this one graph.
IPCC AR5: Figure 5.17. Overlapping 10-year rates of global sea level change from tide gauge data sets (Holgate and Woodworth, 2004, in solid black; Church and White, 2006, in dashed black) and satellite altimetry (updated from Cazenave and Nerem, 2004, in green), and contributions to global sea level change from thermal expansion (Ishii et al., 2006, in solid red; Antonov et al., 2005, in dashed red) and climate-driven land water storage (Ngo-Duc et al., 2005, in blue). Each rate is plotted against the middle of its 10-year period.
Discussion and Conclusions On time scales of 105 years and less, sea-level change at tectonically stable regions is primarily a function of changing ice volumes and the Earth’s response to the changing ice-water load, but neither the ice history nor the response function is independently known with the requisite precision for developing predictive models. Observations of sea level through time do provide constraints on the ice and rheology functions, but a complete separation of the two groups of parameters has not yet been achieved. Separation of the analysis into far-field and nearfield areas provides some resolution, but ambiguities remain: a consequence of inadequate a priori information on ice margin evolution and ice thickness, observational data that deteriorates in distribution and accuracy back in time, the likelihood of lateral variations in the planet’s rheological response, and the everpresent possibility of tectonic contributions.
Lambeck admits he’s modeling something complicated. Does he admit there is a chance his models are wrong?
Unlike other studies, Lambeck’s research has taken into account the Earth’s responses to melting ice. The relationship between melting ice and sea level rise is not simple, says Lambeck. “When the ice sheet melts all sorts of physical things happen and the sea level response to that is quite complex. It will go up in some parts and go down in other parts of the planet.”
For a start, he says, the gravitational attraction between the ice and the water changes. “When an ice sheet builds up, it pulls the water towards it so that within a certain distance of the ice sheet, sea levels actually go up, whereas much further away they have to go down.”
Ice sheets also squash the Earth’s crust down, which causes sea levels to fall, and when the ice melts, the crust slowly rises back up. “So you’ve got the change in the amount of water that goes into the ocean, you’ve got the response of the land to the redistribution of the weight on the surface, and you’ve got the change in the gravity field,” says Lambeck.
Over a period of 20 years, Lambeck and colleagues have been developing a model that takes these factors into account. Together with field data from elevated or submerged remains of shorelines, corals, salt marshes and tree trunks they have been able to get a picture of how sea level has changed over 35,000 years.
The Intro does give us an idea of how tricky is it to measure sea levels when the ground is deforming, rotating, and subsiding.
The sea-level signal from the glacial cycle exhibits significant spatial variability from its globally averaged value because of the combined deformation and gravitational response of the Earth and ocean to the changing ice-water load. During ice-sheet decay, the crust rebounds beneath the ice sheets and subsides beneath the melt-water loaded ocean basins; the gravitational potential and ocean surface are modified by the deformation and changing surface load; and the planet’s inertia tensor and rotation changes, further modifying equipotential surfaces. Together, this response of the earth-ocean system to glacial cycles is referred to as the glacial isostatic adjustment (GIA) (13–17). The pattern of the spatial variability is a function of the Earth’s rheology and of the glacial history, both of which are only partly known. In particular, past ice thickness is rarely observed and questions remain about the timing and extent of the former ice sheets on the continental shelves. The sea-level response within, or close to, the former ice margins (near-field) is primarily a function of the underlying rheology and ice thickness while, far from the former ice margins (far-field), it is mainly a function of earth rheology and the change in total ice volume through time. By an iterative analysis of observational evidence of the past sea levels, it becomes possible to improve the understanding of the past ice history as well as the Earth’s mantle response to forces on a 10 4 y to 10 5 y time scale.
h/t to Panda,and thanks to Lionell Griffiths for advice too.
REFERENCES
Lambeck et al (2014) Sea level and global ice volumes from the Last Glacial Maximum to the Holocene, PNAS, doi: 10.1073/pnas.1411762111 (Supplementary material.)