Guest Essay by Kip Hansen

I have been working on another sea level related essay and in the process stumbled upon a paper published in 1990 by the Permanent Service for Mean Sea Level (PSMSL).  The title is: “On The Availability Of  European Mean Sea Level Data  by P.L. Woodworth, N.E. Spencer and G. Alcock  (1990)”,  The paper is listed on the PSMSL page of “Publications Relevant to the PSMSL and GLOSS” but is not available there.  [ a  .pdf is available here  courtesy of the library at the University of New Brunswick, Canada ].

I am not going to comment on the data itself, other than to include the PSMSL caveat that accompanies other papers of the late 1980s.  I will include a bit at the end as to what the time series might be showing.

Here are the graphs from the cited paper:

I would be remiss if I did not include the caveat that PSMSL adds to papers of this period (1980s — early 1990s):

What I see in the above is that Stockholm, an obvious outlier, has a strong downward trend — this is to be expected as Stockholm is in an area of positive (upward) Vertical Land Movement.  Alicante, Spain, on the Mediterranean Sea, also had a downward trend.  All the others fit nicely into the long-term global sea level rise estimate of 8-10 (some say 8-12) inches for the 20^th century.

I post these graphs for the sole purpose of bringing them into the current century where they can be seen and taken into account.  Note that these graphs (and underlying data, believed to be in this document) has the following features (not bugs…):

1, The data is [mostly] not in the PSMSL database . . . . therefore not generally readily or widely available.

2. The data is of Relative Sea Level (RSL) at each location.

Relative sea level is where the sea surface hits the land.  Relative Sea Level is not the same as Absolute Sea Level or the Sea Surface Height (which  can be thought of as the surface height from the center of the Earth — though that is not exactly so).

3. This very early data comes far before the time of recording tide gauges — so much of it is based on the eyeball method — someone looking at a tide board or an early floating tide gauge with an indicator needle on a tide staff or pole with hand-written records.

The later-half of many of the records may have been from early recording tide gauges. We don’t really know but there are documents regarding some of this data back to 1939:

Publication Scientifique No. 5 published in 1940, No.10 (1950), No.12 (1953), No.19 (1958), No.20 (1959), No.24 (1963) and No.26 (1968).

The International Association for the Physical Sciences of the Oceans (IAPSO) has pdf files of all of these at their web page “Publications Scientifiques (1931-1996)”.

4. Tide Gauges are specifically designed to measure local (at a single location) RSL at various times during the tidal cycle at that location, to discover how much water ships and boats can expect to have under their keels at any given time — to produce useful tide charts for local harbors and rivers to accompany the maritime charts which show water depth (which obviously changes with the tides) so that boats and ships do not hit anything  underwater (rocks, sunken wrecks, sand and mud bottom sea and river beds, and the bars at river and harbor entrances) or go aground.  Ships and boats are concerned with meters and feet of depth, not centimeters and millimeters.  Tide gauges are very good at that and not so good at contributing to anything like a calculation of “global” sea mean level or its rise or fall.

5. Despite the vertical scale of the graphs being recorded and graphed in millimeters, we should acknowledge that the original measurement error must be considered quite large — 2-5 centimeters (inches) , even at the annual mean stage.

6. The individual graphs are self-consistent — though probably not fully corrected for site changes (certainly not at the mm range) — but they do give us a picture of what Relative Sea Level had been doing a the locations given over the specified time scales.

7. NOAA Tides and Currents reminds us, when comparing tide gauge based SL change graphs that “The graphs give an indication of the differing rates of vertical land motion, given that the absolute global sea level rise is believed to be 1.7 ± 0.3 millimeters/year during the 20th century.” (more recent studies use 1.9 ± 0.3 mm/yr).

# # # # #

Author’s Comment:

This essay is similar to the one posted in 2014, Historical Note: Greenwich, England Mean Temperature, 35-yr Daily Averages 1815-1849.   Just an attempt to bring possibly forgotten or potentially lost data back into the light of day.

I thought it was interesting enough to share.

Start your comments with “Kip…” if speaking to me personally.

# # # # #