AR6 and Sea Level Rise, Part 1 – Emerging With That?

By Andy May

This is the first in a series of three IPCC discussions on sea level rise in their latest report, AR6 (IPCC, 2021). The report argues that the rate of sea level rise is increasing rapidly. It’s fair to say why do they think so, what evidence do they provide?

We found the following in AR6 Summary for Policymakers:

“Global average sea level increased by 0.20 [0.15 to 0.25] m from 1901 to 2018. The average sea level rise rate is 1.3 [0.6 to 2.1] five years^-first from 1901 to 1971, increased to 1.9 [0.8 to 2.9] five years^-first from 1971 to 2006, and continued to increase to 3.7 [3.2 to 4.2] five years^-first between 2006 and 2018 (high confidence). Human influence has most likely been the main driver of this increase since at least 1971.” [Bold added]

AR6 Summary for PolicymakersSPM-6 site (IPCC, 2021)

And later in AR6, Chapter 9:

“Global mean sea level (GMSL) increased faster in the 20th century than in any previous century over the past three millennia (high confidence), with 0.20 [0.15–0.25] m increased between 1901 and 2018 (high confidence). The increase in GMSL has been rapid since the late 1960s, with an average rate of 2.3 [1.6–3.1] five years^-first in the period 1971–2018 increased to 3.7 [3.2–4.2] five years^-first in the period 2006–2018 (high confidence). New observational estimates published since SROCC [Special Report on the Ocean and Cryosphere in a Changing Climate, 2019] leading to sea level rise assessed for the period 1901 to 2018 is consistent with the aggregate of the individual components. While thermal expansion of the ocean (38%) and mass loss from glaciers (41%) dominated the total change from 1901 to 2018, ice loss increased and accounted for about 35% of the sea level rise in the period 2006–2018 (high confidence). ”

On pages 9-8 (Chapter 9, page 8) of AR6 . Report:

And, further in Chapter 9:

“At the basin scale, sea level rise was fastest in the Western Pacific and slowest in the Eastern Pacific between 1993–2018 (median confidence). … Man-made signal of sea level change in the area will appear in most regions before 2100 (average confidence). ” [bold added]

AR6 Chapter 9, page 8.

It is somewhat troubling that in the Summary for Policymakers (SPM), they conclude that human influence is “most likely” to be the main cause of the sea-level acceleration and in the Chapter 9, they admit that they do not expect to observe a anthropogenic signal in sea level change in the region before 2100.

Global mean sea level (GMSL) may have increased more in the 20th century than at any point in the past 3,000 years, but how does anyone know? There were no tide gauges or satellites 3,000 years ago. Neither tide gauges nor satellites are accurate to the millimeter level, and certainly not historical records and geological commissions from three thousand years ago. Geological proxies of past sea levels explained by Willis Eschenbach this. Warning, if you have a sensitive stomach, do not Look at Willis’s plot!

Furthermore, why does this make sense? Glaciers reached their lowest Holocene elevation during the Little Ice Age between 1600 and 1850, swallowing entire villages in the process (Behringer, 2010, pp. 89-90). The Little Ice Age was the coldest period in the entire Holocene — about 11,700 years ago to the present day. When the Little Ice Age glaciers melted, one would expect a slight rise in sea levels, but what does this mean? More importantly, can we be confident that the growth rate of GMSL is accelerating?

We also find it strange that they conclude the rate of increase in the GMSL is increasing based on a comparison of linear least squares fit over selected parts of the sea level record. Figure 1 plots the entire NOAA GMSL record by quarter since April 1880. The x-axis is the number of quarters (three-month period), and the mean sea level is given on the y-axis through 2020. Profile This is built from White Church and GMSL Data (Church & White, 2011) through 2010 and University of Hawaii Rapid Delivery Data Later.

AR6 chooses many of the specific time periods in the quotes above to justify their claim that the rate of sea level rise is accelerating. This statement is visually ambiguous because the data looks a bit wavy, but linear. The apparent acceleration period from 180 to 300 (1925-1952), looks suspiciously like 420 to 510 (1985-2007). The entire record, from April 1880 to the end of 2020 has a slope of 1.65 mm/year with a R² of 0.97. We can cherry-pick the time intervals all day and fail to deal with anything significant regarding speeding or lack of it. The least squares statistics for the AR6 cherry picking stages mentioned in the above citation are given in Table 1 and compared with four cherry picking I.

While AR6 claims the acceleration is happening with great confidence, the previous report said:

“However, the trend in the GMSL observed since 1993 is not significantly larger than the estimate of the 18-year trend in previous decades (e.g. 1920–1950).”

AR5: (IPCC, 2013, p. 290)

One wonders why the AR6 got a different look just seven years later.

AR6 would lead us to believe that since the least squares linearity is more consistent with sea level rise from 2006 to 2018 than from 1971 to 2018, it is accelerating. However, from 2012 to 2020, the rate is almost as low as from 1971 to 2018. The largest increase in Table 1 is only 15 inches or 38 centimeters per century, hardly alarming. at global tides, in the open ocean, more than twice on average that; and coastal tides are usually ten times that value daily. Climate changes on time scales over centuries, as we can see when we compare the Little Ice Age with the Medieval Warm Period, so a particular record from 1880 to AD 2020 is unlikely to capture the full extent of sea level velocity. Estimates of sea level rise gathered from historical and geological records show that sea levels have risen much faster in the past, as shown in Figure 2 created by Robert Rohde.

Sea level is estimated using tide gauges mounted on coastlines around the world. The best of these gauges are only accurate to ± 5 mm monthly average (NOAA, 2020). Satellite sea level measurements are problematic unless the meteorological conditions are perfect and they are trying to measure the altitude of a moving surface. AR6 acknowledges that satellite estimates of the “acceleration” of sea level rise are much smaller than those of heavily massaged tidal gauges. This is discussed on AR6 pages 9-96, where we see that the satellites found accelerations from 1993 to 2015 to 2006-2015 were between 3.16 mm/year and 3.58 mm/year. years, this is less than half the acceleration. mm/year² for about a decade. Other satellite estimates are similar. Satellite estimates of sea level are not accurate to half a millimeter (Frederikse, et al., 2020).

Is the difference between the estimated global average of 3.8 mm/year and 1.8 mm/year statistically significant, considering the data used? Especially when these measurements are taken over a few decades? It seems unlikely, but let’s take a closer look at the data.

AR6 claims show an increasing rate of sea level rise due to human influence. This is probably because greenhouse gas (GHG) emissions cause surface warming, which then melts land-based glaciers. This raises two questions:

1. Is the increase in sea level rise statistically significant?

2. If so, could anthropogenic warming of greenhouse gases be possible?

From AR6:

“The warming climate system has caused global mean sea level to rise due to loss of land ice and thermal expansion due to warming of the oceans. Thermal expansion explains 50% of sea level rise between 1971–2018, while ice loss from glaciers contributes 22%, ice sheets 20% and changes in land water reserves 8% straight. The rate of ice loss increased fourfold between 1992–1999 and 2010–2019. Along with that, the loss of ice mass and glaciers were the main contributors to the global mean sea level rise in the period 2006-2018 (high confidence). ”

Page AR6: SPM-14

Thus, ocean warming since the Little Ice Age, provided about half of sea level rise. Melt provides most of the rest.

From AR6:

“By 2100, GMSL is expected to increase by 0.28–0.55 m (possible range) below SSP1-1.9 and 0.63–1.02 m (possible range) below SSP-8.5 compared with the 1995–2014 average (average confidence). Below CO is higher₂ Under emission scenarios, there is profound uncertainty in sea level projections for 2100 and beyond with regard to the response of the ice sheets to warming. Low probability, high impact plot and high CO₂ In the emission scenario, iceberg processes characterized by deep uncertainty could lead to a GMSL increase of about 5 m by 2150. With long-term commitment, uncertainty about when the increases will be reached. Different GMSL is an important consideration for adaptation”

AR6 page TS-44

Some IPCC climate models predict sea level rise of up to 5 meters by 2150, when the current rate of sea level rise is less than 40 centimeters or 1.3 feet per century? Considering that IPCC models have failed to accurately predict climate after 30 years of trying (McKitrick & Christy, 2018), excuse my skepticism.

AR6:

“It is almost certain that global mean sea level will continue to rise until 2100…

After 2100, GMSL will continue to rise for centuries due to continued heat absorption in the deep ocean and mass loss of the Greenland and Antarctic Ice Sheets and will remain elevated for thousands of years (high confidence). ”

AR6 Chapter 9, pages 9-9.

The first statement may be true, we are still warming up as we emerge from the Little Ice Age and I would suspect a change in the direction of glacier retreat before 2100, the second statement is speculative Purely, predicting after 2100 is imprudent.

In summary, AR6 statements about sea-level rise acceleration based on simple rough and chosen linear least squares are consistent with sea level data for the past 140 years. They also combined data and trends for ocean warming and land-based glacier melt. The problem is that today’s rates of sea level rise are so small and linear that their attempts to predict how large rates they will rise are statistically inefficient and almost comically inefficient. In the next article we will examine the complexity of measuring GMSL and later in this series we will provide a more statistically significant forecast of sea level rise, much lower than with wild predictions from AR6.

Folder can be downloaded this.