Archive for the ‘sea level’ Category

h1

Vermeer and Rahmstorf paper rejected

January 31, 2014

Vermeer and Rahmstorf had a paper rejected by the journal “Climate of the Past.” This news is 16 months old, but I just heard about it, and could find very few references about it on the web.

This paper, On the differences between two semi-empirical sea level models for the last two millennia,  promoted their earlier sea level rise models.  They couldn’t seem to get traction with this paper.

Here are some reviewers’ comments…

One of the major problems with this work is the decidedly biased analysis and presentation.

Highly biased analysis and presentation.

It currently takes significant effort to figure out which pairs of models and training data sets the authors use, and whether they have evaluated all the relevant combinations of the same.

No surprise here.  Rahmstorf has a history of alluding to all kinds of data sets and implying that he has taken them into consideration, but only presenting results for those that support his thesis.

And the final blow…

In the light of the two negative reviews and one comment which all require new analyses and point to fundamental flaws in the methodology of the current paper, I regret to inform you that my conclusion is to support rejection. I strongly dissuade the authors from submitting responses and a revised version.

Here is the paper…

Click for full PDF version

Here is the reviewers’ discussion that lead to the the rejection.

Of course, Vermeer and Rahmstorf do not give up that easily, and similar papers have been shopped around to other journals

h1

The Search for Acceleration, part 9, the Baltic Sea

October 23, 2013

magnifying glass 145This is part 9 of a series of posts in which I am searching for a large acceleration in sea level rise rate in the latter part of the 20th century.  Such a rise rate acceleration is needed  to reconcile the 1.8 mm per year average rise rate for the century attributed to tide gauge data and the approximately 3 mm per year rise rate for the tail end of the century attributed to the satellite data.

The Baltic Sea

There are 22 tide gauge stations in the Baltic Sea area that are at least 90% data complete from 1960 to 2005.  Eighteen of those are 90% complete all the way back to 1930 and ten are 90% complete back to 1900.  The weighting (using a 200 km threshold) is nearly constant for the entire 20th century (see weighting graph below).  I will use the usual technique of detrending, weighting, averaging and derivatives, as shown in the following slide show.  (Note that you can pause or increment the slide show forward or backward by using the buttons that appear when your cursor is placed over the image.)

This slideshow requires JavaScript.

Conclusion

The following graph makes clear that the Baltic Sea tide gauge data DOES reconcile the sea level rise rate from the tide gauge data with the higher late century rise rate from the satellite data.

Baltic Sea Detrended Acceleration annotated 2
On the other hand, the tide gauge sea level rise rate immediately before the era of satellite data is higher than rise rate after….

Baltic Sea Detrended Acceleration annotated 3

See an index of the Search for Acceleration series here.

_________________________________________

Sources

20th century rise rate average of 1.8 mm/year

1. Church and White Global Mean Sea Level Reconstruction

2. Links to Church and White sea level data

Satellite data (about 3 mm/year): CU Sea Level Research Group

RLR tide gauge data: Permanent Service For Mean Sea Level

h1

The Search for Acceleration, part 3: Japan

June 26, 2013

magnifying glass 145

CORRECTION: 6/30/13

The original detrended sea level rise rate graphs for this post was off by a factor of 12!.  This greatly changes my conclusion.  Incorrect information is now crossed out and is followed by corrected information in red.

Tide gauge data for the 20th century indicates that the average sea level rise rate was 1.8 mm/year.  Satellite data from 1993 to present indicates a sea level rise rate of about 3 mm/year.  This is part 3 of a series of posts looking for the acceleration necessary to reconcile those two facts

I am working under the theory that by detrending sea level data from individual (local) sites and averaging with other regional sites it should be possible to extract changes in regional rise rates while bypassing the question of what the “true” rise rate is for that region.

Japan

Conclusion: There is no convincing sign of a late century acceleration in the sea level rise rate in the tide gauge data from the Japan.

Conclusion:  The rise rate during much of  the satellite era has been much higher than the average for part of the 20th century for which data is available.

I looked for tide gauge data along the coast of Japan such that it covered at least the period from 1955 to 2008 with 90% of all monthly data accounted for. The following image shows the seven sites that met this criteria.  The circles show a weighting threshold of 300 km.

Seven sites used in this analysis.  Circles show 300 km weighting threshold

Seven sites used in this analysis. Circles show 300 km weighting threshold

The following plot shows the qualifying data spread out for easy comparison. The key at the right shows the RLR data filenames.

Japan Raw Spread

Sea level data for all seven sites.

Data reduction and detrending

The following animation shows the transition through raw data, removal of the yearly signals, detrending, Gaussian smoothing and conversion to derivative (rise rate).

Japan 90p 1955-2008 450ani corrected

Here are the removed yearly signals and the weighting.

Yearly signals removed from Japanese RLR data

Yearly signals removed from Japanese RLR data

Number of files and effective weighting based on 300 km threshold.

Number of files and effective weighting based on 300 km threshold.

Lets take a closer look at the detrended rise rate data and look for an acceleration in the satellite data era…

corrected rise rate Detrended Acceleration annotated Japan 90p 1955-2008 Detrended Acceleration annotated

corrected rise rate Detrended Acceleration annotated

The very weak argument could be made that there was a rapid acceleration around 1985, but the resulting sea level rise rate was only about 0.25 mm/year higher than the average for the last half of the century. There was also an even greater acceleration around 1965, and sea level rise rate around 1970 was as high or higher than than in the 1990s.  Finally, the 0.25 mm/year increase in the rise rate is only about 20% of the difference between the average global tide gauge rise rate for the 20th century (1.8 mm/year) and the satellite data (1993 to present) rise rate (about 3 mm/year).

So, I conclude that the Japanese data does not reconcile the difference between the 20th century tide gauge data and the satellite data.

The tide gauge data covering the part of the satellite data era (1993 to present) clearly shows a rise rate that is far greater than the average rise rate for the entire time period covered by the tide gauges.  The period from 1993 to about 2003 may have a rise rate around 3 mm/year greater than the average, but after that the rise rate seems to fall again.  Note that form about 1965 to 1975 the rise rate was also very high.  This data from Japan does reconcile the difference between the satellite data and the average tide gauge data.

Fukushima

The following graphs show the sea level data from the Soma tide gauge station in Japan with the seven station shown above.  Soma is the tide gauge station closest to the Fukushima nuclear reactors.  The images speak for themselves.

Japan with Soma Raw Spread

Fukushima map

********************************************

Sources

20th century rise rate average of 1.8 mm/year

1. Church and White Global Mean Sea Level Reconstruction

2. Links to Church and White sea level data

Satellite data (about 3 mm/year)

CU Sea Level Research Group

RLR tide gauge data

Permanent Service For Mean Sea Level

Follow

Get every new post delivered to your Inbox.

Join 48 other followers