In yesterday's post I noted that I had completed version 1.0 of a module that would generate abstract graphs (e.g. Fig. 1).
The purpose of the module is to generate graphs to show how things would look in a perfect mathematical world as compared to our current datasets of in situ measurements (On Thermal Expansion & Thermal Contraction - 27).
The reason for that exercise is to test our measurements of ocean temperatures, salinity, depths, and sea level changes against the backdrop of an informed simulation (e.g. Fig. 2).
The results shown in Fig. 1 and Fig. 2 turned out to be within the realm of what was to be reasonably expected (because absolute agreement is not expected between abstract and measured).
|Fig. 3 The Problem|
With those results in mind, I indicated that I would proceed to test the World Ocean Database (WOD) to see if the abstract and the measurements in the WOD would generate the same general results.
The results of version 1.1 of the module show that the WOD datasets have unbalanced measurements in either depths or latitude / longitude locations, or both (e.g. Fig. 3).
That is, the measurements are concentrated in locations or depths that are not a reasonable pattern-match in terms of what is to be expected according to the abstract projections.
The measurement data used to generate the lower two panes in Fig. 3 reflect two circumstances.
The first circumstance, shown in the lower left pane, details the ocean temperature measurements using all CTD and PFL measurements in all WOD Zones (1968-2016).
The second circumstance, shown in the lower right pane, details the ocean temperature measurements in only the "Golden 23" zones (also 1968-2016).
Both graphs in the lower panes show the in situ measured water temperatures decreasing over the period 1968 - 2016, with the lower right pane showing an up-tick in temperatures over the last few years, but the lower left showing a decrease in temperatures over the same period.
That contrast in Fig. 3 is even more stark when compared to the upper panes that show generally and mathematically what should be happening.
This situation is what Dr. Mitrovica pointed out in the video below when the same problem plagued sea level science concerning "the European problem."
At one time tide gauge station measurements in Europe were confounding because they showed consistent sea level fall, or a smaller increase when other places where showing more sea level rise.
It was called "the European problem" in the scientific literature because some scientists forgot about the law of gravity and Newton.
Professor Mitrovica goes on to point out that a scientist by the name of Bruce Douglas indicated that scientists have to pick and choose measurements from locations (and depths) which compose a balanced sample that shows actual global dynamics.
This is an elementary consideration, and is obviously true when one contemplates what would be shown if our measurements were taken only at the equator, or (in stark contrast) if our measurements were taken only at the poles.
So, with that in mind I chose as one sample the "Golden 23" WOD Zones, and as the second sample all WOD Zones.
As it turns out, those two test cases for ocean temperature measurements are at odds with what is taking place in the oceans as a whole.
So, it turns out (as shown in Fig. 3), that we have "a European problem" as it were involving in situ measurements.
So, we have to determine the best WOD Zones to use to reflect a closer pattern match to the abstract graphs.
Meanwhile, I will be completing the thermosteric volume logic in the module so we can print that out and show the contrast (as we search for the "Golden WOD Zones" to use as our in situ measurement balance.
The next post in this series is here.