Whose fingerprints are on sea level? |
Once upon a time the Earth was flat (according to those who want to make flatulence great again).
They add that its oceans were also flat at their surface, like water in a bathtub (Once Upon a Time in the West - 2).
Today, in some circles the bathtub model is a persistent meme.
But, if the "Bathtub Model" (BM) was correct there would be no use for sea level change science in the sense of an oft heard engineering question: "how much of sea level change is this town going to get?"
That is because pursuant to the BM answer, every town everywhere would get the same amount of sea level change (SLC).
I still hear lame commentators say that SLC is like when you pour an additional jug of water into a bathtub, the water level rises the same amount everywhere in the bathtub.
It is terrible that this false BM hypothesis has been and is still unwittingly used by some science writers.
But, it is great that the BM is fading away, since towns around the world keep tide gauge records then send them to the Permanent Service for Mean Sea Level (PSMSL).
Some of these PSMSL records go back to the late 1700's, and one thing is absolutely for sure, the records show that sea level change is anything but uniform around the globe (The Bathtub Model Doesn't Hold Water, 2, 3, 4, 5).
This has been known by aware scientists since Woodward (1888) as pointed out in a more recent paper: "We find that there can be large errors in the usual assumption that changes in sea level are uniform over the ocean basins" (On Postglacial Sea Level, 1976).
II. A Practical Purpose For Sea Level Science?
Generally, an aware and competent engineer designing an SLC solution for a town in one country at latitude 30 deg. N. will not come up with the same exact specifications as an aware and competent engineer designing the same solution for another country at latitude -10 deg. S. (Sea-level rise: towards understanding local vulnerability).
Fig. 1 |
Fig. 2 |
Fig. 3 |
Fig. 4 |
Fig. 5 |
Fig. 6 |
The SLC is not uniform from place to place (Beyond Fingerprints: Sea Level DNA).
So, it behooves sea level scientists to develop tools for assisting societies to respond to their future according to their specific location on the globe (latitude & longitude).
III. The Best Way
The best way to determine those needs is to use measurements and observations from tide gauges as close to their locations as possible (The World According To Measurements, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21; Oceans: Abstract Values vs. Measured Values, 2, 3, 4, 5, 6, 7).
So, in the hypothesis that is being developed in this Dredd Blog series, we will use the robust measurements of the PSMSL.
IV. Implementing "The Best Way"
Using existing measurements to assist an engineering team is straight forward.
One begins with general measurements, then fine tunes them.
The first post of this series set the stage by using general measurements.
By that I mean SLC calculations generalized (averaged) from measurements based on all tide gauges in a country.
In today's post, we begin to fine tune those measurements to less-generalized and more specific values by using what the PSMSL calls "coastline codes".
For example Brest France data indicates a tide gauge station code of 1, and a coastline code of 190.
We only need to do that type of fine tuning when more than one coastline code exists in a country from which the tide gauge measurements have been taken.
Some of the countries have as many as nine coastal codes.
Anyway, as you can see from today's link table below, generalizing based on individual coastline codes produces a more fine tuned result than using all the coastline codes in a country averaged together.
V. The Link Table
Here is the table of HTML links for countries with multiple coastline codes, and those coastline code graphs displayed separately in Appendices specific to this post:
Coastline Code Tide Gauge Info | Coastline Code DNA Graphs | Coastline Code SLC Graphs |
Countries: A - C | A - C | A - C |
Countries: D - G | D - G | D - G |
Countries: H - L | H - L | H - L |
Countries: M - O | M - O | M - O |
Countries: P - T | P - T | P - T |
Countries: U - Y | U - Y | U - Y |
Here is the table of HTML links for countries with one coastline code, or multiple coastline codes averaged together (from Beyond Fingerprints: Sea Level DNA):
Tide Gauge Info | DNA Graphs | SLC Graphs |
Countries: A - C | A - C | A - C |
Countries: D - G | D - G | D - G |
Countries: H - L | H - L | H - L |
Countries: M - O | M - O | M - O |
Countries: P - T | P - T | P - T |
Countries: U - Y | U - Y | U - Y |
VI. Some DNA Views
I modified some of the graphs by removing the SLC line in some of the "Coastline Code DNA Graphs" in section U-Y.
That is the graphs here were modified (by removing the black SLC line) to those shown in Fig. 1 - Fig. 6 above in Sections II, III, and IV (Note: the black line in Fig. 1 - Fig. 6 above is the red line on the graphs here, and the red line was the green line ... I haven't figured out how to keep the same line colors yet - in SciDavis, sorry).
Notice that the two graph lines (SLR and SLF) in Fig. 1 - Fig. 6 are described as influences on the TREND.
That really is what is happening when an ice sheet loses or gains ice mass.
It has an influence on the sea level around the globe which will be recorded at tide gauge stations.
The result will be an influence on SLF (or a slowing of SLR) at tide gauge stations "nearer" to the ice sheet, but to the contrary an influence of SLR will take place as a slowing of SLF at tide gauges "farther" from the ice sheet.
Since there are tide gauges all around the globe at many distances, the influence can be very distinct when compared to other locations (the graphs made from actual measurements around the globe clearly expose this phenomenon).
Further, since there are other influences besides the two ice sheets (Greenland and Antarctica) such as Glacier Bay, Svalbard, and Patagonia, their influence can be substantial depending on distance from a large ice deposit, and the Earth's Rotation (The Gravity of Sea Level Change, 2, 3, 4; NASA Busts The Ghost; Proof of Concept, 2, 3, 4, 5, 6, 7, 8, 9).
VII. Closing Comments
The country generalizations whether country-wide or more fine tuned by coastline codes are not a perfect solution.
But they do present the picture that presidents, congress members, courts, local government officials, engineers, and the general populace must grasp.
The further perfecting of this DNA drama could be attained by parsing the coastline codes into areas having more distinction, based on trends.
For one example, the coastline codes could be modified into nnn.nn (a decimal number).
The decimal portion could signify further uniqueness and accuracy for that specific portion of the coastline.
In some cases a combination of sections of current Coastline Codes would be a further fine tuning.
While such minutia is a further improvement, the main concerns (in terms of sea level DNA) are: 1) "when will the glaciers (not in Greenland or Antarctica) be substantially gone", 2) "when is Antarctica melting going to begin to accelerate enough to overtake Greenland", and 3) "when is Greenland going to begin to disintegrate in its northern section?"
Those three events will alter the SLC DNA substantially, and therefore present a potentially different problem to "presidents, congress members, courts, local government officials, engineers, and the general populace" which also must grasped.
Leaving fossil fuels in the ground, if it is even possible, is the key to diminishing all of the potential catastrophes, and stabilizing the DNA.
It is also the easiest to grasp, but the most difficult to accomplish, seeing as how the current civilization is addicted to fossil fuel use.
The next post in this series is here, the previous post in this series is here.
Lyrics to Time Has Come Today are here ...
No comments:
Post a Comment