Check: CO2 emissions fingerprint on sea rise
Revision as of 11:57, 26 February 2020 by Tim88 (Addendum)
Has sea rise been accelerating due to increasing CO2 emissions? - v.3.8 by Tim88 - Overview article: Check: Effects of CO2 emissions on nature
Do you think that sea level rise in the 20th century was due to greenhouse gas emissions? Prominent climate scientists expected the effects of increasing greenhouse gas emissions to be detected by the end of the 20th century. In 2018, Nerem and co-writers confirmed that the speed of sea level rise accelerated in the preceding 25 years. Their findings are based on satellite measurements which are hard to verify.
Here we analyse a representative sample of tide gauge measurements over 100 years, even 150-200 years. We find no match with CO2 emissions in the first half of the 20th century but we confirm a clearly increased speed of sea rise since 1990. However, we find that the recent speed of sea level rise is still not extraordinary: it is not much more than in the period 1925-1942 when the relative speed of atmospheric CO2 increase was very much less. You can read the details in the following article. The data analysis spreadsheets are available for download so that you can check it out for yourself.
More convincing evidence of an effect of CO2 emissions can be found in Check: CO2 emissions fingerprint on temperatures
Greenhouse radiation theory predicts a significant change in climate due to continually increasing greenhouse gas emissions, in particular CO2. The estimations of the strength of the effect vary from 0.5 to 5°C for a doubling of the atmospheric CO2 concentration. The accelerating increase in atmospheric CO2 concentration should therefore lead to an increased speed of sea rise from faster melting ice and expanding water.
But according to an article by Holgate in 2007, sea rise during the 20th century has been decelerating, or at most linearly increasing, until at least 2003.
Recently a paper by Nerem and co-writers of 2018 found that in the preceding 25 year, sea level rise has been accelerating. They suggest that recent, human induced climate change has been detected in sea level rise, and this was apparently confirmed in a more recent paper by Fasullo and Nerem.
Note that this is inconsistent with the widespread idea that overall 20th century's sea rise was caused by humans, as many scientists including for example the Smithsonian Institute tend to think. Prominent climate scientists such as Hansen in 1981 expected instead that the effects of increasing greenhouse gas emissions would become detectable by the end of the 20th century, as apparently confirmed by Nerem. Case closed? Read on!
As elaborated in the above link, we found a reasonably good qualitative match between emitted CO2 and atmospheric CO2 concentrations during the last century, especially since the beginning of continuous measurements in 1959. Notably the speed of concentration increase in the 1960s corresponds. There is a less good match before 1950 which may be due to natural warming, which also increases atmospheric CO2. It is very reasonable to believe that current CO2 levels are strongly driven by man-made emissions.
According to greenhouse gas theory, the increased CO2 levels in the period 1900-1960 and much more so in the period since 1960, should have caused more warming of the planet. This in turn increased the rate of sea level rise due to increased melt water from the ice caps and glaciers. It is thought that this also warmed the oceans significantly, resulting in sea water expansion. It is expected that the resulting additional rate of sea rise can be detected in tide gauge records.
On a logarithmic scale, the speed of CO2 increase in the period after 1980-2018 is at least 5 times that of the period before 1960. In theory this should cause also 5 times faster warming, and thus also a 5 times faster sea rise.
In order to determine the effect on sea rise we provide here an update of Holgate's 2007 analysis of sea rise until 2003. He selected a rather small sample of high quality, long term sea level gauges spread around the world. Tide gauges record the change of sea level relative to the coast; at many places the coast subsides but at some places it raises, apparently as an aftermath of the last ice age and the little ice age.
Differently from Holgate (he applied a complex intermediate step) we simply averaged the raw data for an approximately global estimation of average sea rise. That makes the results easy to verify and prevents doubts about possible artifacts from data manipulation. For the same reason we did not make "data corrections" for air pressure variations, volcano eruptions and El Nino's. In the long term such effects should average out and insofar noteworthy they are mentioned in the supplement.
Holgate selected 9 long term sea level gauge stations of high quality that he found to be a reasonably representative sample of global sea level rise: they compared rather well with the results of an earlier global study that included many more sea level gauges. His set consisted of New York, Key West, San Diego, Balboa, Honolulu , Cascais, Newlyn, Trieste, Auckland.
For this updated study two of those gauges had to be omitted because they have been taken out of use so that no recent data exists. Preferred are gauges that do not look erratic and that have few and only little "gaps", which can be fixed by simple interpolation without introducing significant errors. As Europe and the Northern hemisphere were a bit over-represented in Holgate's selection, the Caiscais tide gauge was not substituted. The tide gauge in Auckland was substituted by one in Sydney. It turns out that Sydney is located in a zone that, according to the fore-mentioned paper by Fasullo and Nerem, in recent decades experienced abnormally strong increase in sea rise. Indeed the sea rise of the last decades was above average; however it does not look extraordinary.
Here are the results for the eight long term tide gauges, with sea level measured in mm:
The effect of omitting each of the gauges was tested, showing little difference. It's also similar with San Francisco and Cuxhaven included. Overall, the average trend was around 18.5 cm sea rise over the last 100 years; the average of the last 20 years was about twice as much (details here).
Insofar as Holgate found an apparent deceleration and Nerem found an acceleration, we obtain reasonably qualitative agreement with their findings but in the overlap period where their findings seem incompatible, our results match better with that of Nerem. We confirm a clearly increased rate of sea rise in recent decades.
Possibly global dimming prevented the greenhouse effect of CO2 emissions from showing up in the period 1960-1990. Since then there has been an inverse "global brightening" taking place, as pollution is being reduced.
The increase over the first two decades of this millennium is slightly higher than the increase that was recorded in the 1930s. If the observed sea rise over the last 100 years were fully due to the contemporary change in CO2, then the 5 fold increased rate of CO2 increase over 1995-1995 compared with the period 1920-1960 should have produced a similar 5 fold increase in the rate of sea rise. It's not much different if we assume a ten year delay (for an estimation of the delay, compare with Northern hemisphere temperatures here). The observed increase in rate of sea rise would be strong evidence of anthropogenic global warming by means of CO2.
However, the recently increased rate is of the same order of magnitude as in the earlier period, only 30% higher instead of 400% higher (detailed discussion in the supplement). Thus, only a fraction of current sea rise can be attributed to CO2. Most of the average sea rise of this sample of tide gauges may be attributed to natural warming as well as sinking land. Increased land sinking due to human activity such as heavy buildings and water extraction may also play a role.
It should be noted that possibly black carbon emissions played a stronger role in enhancing the melting of ice and snow in the 1930s than in the last 20 years. If so, both those periods of increased melting may have been partly "man-made" for different reasons.
Very long term measurements
It's a bit unsatisfactory that this data set does not go far enough back in time to catch the earliest possible onset of global warming and sea level rise due to the industrial revolution. Very long term gauges that are deemed to be reliable and go back to 1850 or even earlier can be found in the PSMSL data set.
Again, preferred are some that do not look erratic and that have few and only little "gaps". San Francisco in America as well as Cuxhaven and Stockholm in Europe (with added data by Ekman) were thus selected for an impression of very long term sea rise (relative to land) in the Northern hemisphere.
We observe very linear trends over the last two centuries: the trends for Stockholm and San Francisco are only slightly rising, while that of Cuxhaven is practically unchanged. However, the full record is much less uniform. Some other long term gauges such as Brest show a clearly increased rate from around 1900, but little further change over the last century.
More analysis with additional tide gauges can be found in the supplement.
We find that the overall global trend in sea rise over the last 100 - 200 years does not match the increased human greenhouse gas emissions. The recently increased rate of sea rise may well be partly due to increased CO2 emissions but based on this record we cannot say how much of it is natural and how much of it is due to atmospheric CO2.
Addendum: more convincing evidence of an effect of CO2 emissions can be found in Check: CO2 emissions fingerprint on temperatures. In view of that additional evidence, it seems very likely that a significant part of the recently increased rate of sea rise is due to CO2 emissions.
Discussion: in the Forum you can give comments, suggestions and criticism