Saturday, December 23, 2017

Oceans: Abstract Values vs. Measured Values - 7

Fig. 1
I think that the search for the "Golden" problem is solved concerning the proper WOD Zones to use (see Fig. 5 for those zones).

Today's graphs are the culmination of the software module upgrade and database restructuring.
Fig. 2

Fig. 3
The gist of it is that I use all WOD Zones in the ocean basins listed in the WOD Manual (PDF, Appx 11), as long as there are  temperature & salinity measurements in the related datasets (CTD & PFL).

That "Golden" totals to 217 Zones taken from all WOD layers containing data, (today's graphs are all from that dataset).

Fig. 4
The graph at Fig. 1 is T and CT without the abstract maximum and minimum fences.

Not using the fences increases the resolution so that you can see glimpses of both the T pattern and the CT pattern individually at several places.

The graph at Fig. 2 is the exact same data with the fences included.

I included both graph types so you can see that the calculated TEOS CT is quite close to the in situ measurement values (T), but they are not exactly the same.

The graph at Fig. 3 once again reaffirms that the Absolute Salinity (SA) and the in situ salinity measurement, Practical Salinity (SP), vary more than the T and CT temperatures do.

As I indicated in a recent post (On Thermal Expansion & Thermal Contraction - 29), that is to be expected because SP involves measuring by ascertaining conductivity, while SA involves calculating mass (H2O and non-H2O content) factors.
Fig. 5 WOD Zones with in situ data (blue squares)
The really sweet spot is the graph at Fig. 4 concerning thermal expansion values, calculated using the TEOS-10 toolkit (excluding uplift & subsidence).

The expansion / contraction pattern ends at 4.82% (8.92699 ÷ 185.157 = 0.048213084 ... 4.82%) of the PSMSL sea level value as recorded by tide gauge stations.

Regular readers know that for years Dredd Blog calculations of thermal expansion were based on a 5.1% value for thermal expansion (e.g. On Thermal Expansion & Thermal Contraction).

That 5.1% was the percent left over after accounting for ice sheet loss (melt-water increase) and ghost water (gravity-held ocean water release) relocation (NASA Busts The Ghost).

The previous post in this series is here.

Lyrics to "25 Or 6 To 4"

Thursday, December 21, 2017

On Thermal Expansion & Thermal Contraction - 29

Fig. 1a
Fig. 1b
Fig. 1c
Fig. 1d
Fig. 1e
Fig. 1f
Fig. 1g
Fig. 1h
Fig. 1i
Fig. 2a
Fig. 2b
Fig. 2c
Fig. 2d
Fig. 2e
Fig. 2f
Fig. 2g
Fig. 2h
Fig. 2i
I. Background

I finally came up for air after redesigning my SQL tables.

As regular readers know, traditionally I have used seven depth levels of the ocean, while the WOD manual uses 33 depth levels.

So, I changed the SQL tables to hold 33 depth levels rather than seven, and rebuilt the tables.

Which necessitated redoing the software to accommodate those changes.

A perplexing development in the world of graphing ocean dynamics emerged as I mentioned in the previous post (On Thermal Expansion & Thermal Contraction - 28) and two other posts further explaining the issue (see e.g. Oceans: Abstract Values vs. Measured Values - 6).

Some of the problem developed because I was using two depth level models; one used seven levels, the other thirty three depth levels.

It was not accurate enough that way, so I went the way of the WOD manual, and adjusted the databases and software modules accordingly.

II. Redesign Issues

While planning a strategy it dawned on me that the processing of the work of scientists (who worked hard to gather the billions of measurements they make available to us) should be like human governments (from the bottom up).

So, the processing of the salinity and temperature data begins at the deepest depth that the measurements were taken at, and then works up to the surface level from there.

It follows the natural flow of a warming world in the sense that as the processing moves up through the depth levels, the ocean temperatures are generally warming.

Having changed the number of depth levels from seven to thirty three, it also changed the quantity of water associated with each depth level.

I process each of the thirty ocean basins separately because they have different temperature highs and lows at each depth level.

So, the volume of water at each depth level is smaller, which works to increase accuracy quite a bit.

I consider and process each depth level as if it was the entire ocean in the sense that there is a consistent volume of water, but the pressure, temperature, and salinity tends to be different at each depth level.

When the software processing reaches the final level (0-10 meters in depth), I add up the recorded salinity, temperature, and thermal expansion and contraction values, taken from each depth level on the way up. 

That has to be done for each ocean basin (WOD zone by WOD zone) because the latitude and longitude contribute significantly to the results.

The latitude and longitude values are parameters in some of  the critical TEOS formulas, so the use of small zones and levels increases the accuracy of several results.

Another factor that improved the functionality and the results was using the same logic to calculate the Abstract Maximum and Minimum values as well as the WOD measured values.

That results in a synchronization which removed the inconsistency in thermal expansion results compared to temperature and salinity results.

They are all now in sync as the graphs in today's post show.

All of the categories  fit well within the maximum and minimum parameters which are taken from the WOD manual @ Appendix 11.

III. The Why Of It

This series explores the realm of the hypotheses of thermal expansion of the oceans as the global climate system gyrates to the increases in global warming.

The foundational statement is usually "water expands as it is heated" without mentioning that "water shrinks as it is heated" because each of those two eventualities depends on the temperature of the water at the time the heat is added to the water (On Thermal Expansion & Thermal Contraction).

That discrepancy inspired me to look closer and deeper into the subject, as this series details.

Along the way I discovered that one can get a completely different picture of the ocean depending on where one takes measurements.

Today's graphs show how different patterns emerge depending on what segment or section of ocean measurements one uses.

That does not mean that the measurements are wrong, it means that one must use a balanced selection of measurements.

It is like having five different thermometers in a house where there are different temperatures inside.

The temperature in the garage, a closet, inside the refrigerator, in the basement, and in the attic will render different temperatures.

That is why the thermometer that controls the heating and cooling systems has to be strategically located.

The same goes for determining general global conditions of the oceans as they absorb global warming.

IV. Additional Feature Issues

To give us another advantage to use in our ongoing quest about the facts and fantasies of thermal expansion, I added additional WOD layer processing along with additional WOD zone processing while I was modifying the software modules.

A separate functionality class contains lists of layers as well as lists of individual zones (which can be easily changed).

The current configuration allows the processing of six "golden layers", six "golden zones", eight "golden layers", eight "golden zones", as well as processing 117 zones and processing all 18 WOD layers.

The reason for these differences is because I am searching for what has already been found for tide gauge stations and weather stations.

I am talking about a representative few examples which one can look at to get the big picture.

V. The Graphs

The graphs today come in two sections.

One is the layers section (Fig. 1a - Fig. 1i, left side), and the other is the zones section (Fig. 2a - Fig. 2i, right side).

The use of layers (36 zones per layer) arguably has the advantage of being in sync with the natural temperature bands, beginning at the warmest band at the equator, then moving towards the gradually cooling bands as the layers proceed toward the coldest band at the poles.

The use of lists of zones allows one to be a bit more selective.

One way or the other, we should be able to find the Goldilocks "just right" configuration to use to inform us of an accurate, broader picture (like tide gauge and weather station configurations do).

VI. Granularity

I noticed that the TEOS calculations give a tighter group in the temperature processing than they do in the salinity processing.

Notice that the salinity patterns in Fig. 1a - Fig. 1c tend to produce two visible lines, while the temperature related graphs at Fig. 1d - Fig. 1f look as if they produce the same line (i.e. only one line).

That is not the case, it is just that the numbers are different @ the temperature graphs (T compared to CT), but they are not different enough to make two visibly discernible lines at this resolution.

Conservative Temperature (CT) and Absolute Salinity (SA) are TEOS concepts, however, of the two, SA, is the more revolutionary departure from "40 years of error"  when scientists were using conductivity as the proper measure of salinity.

So, the smaller difference between in situ temperature (T) measurements and calculated Conservative Temperature (CT) values, compared to the larger difference between in situ salinity (SP) measurements and calculated Absolute Salinity (SA) values, is not a great surprise.

VII. Conclusion

We can now move forward into finding out why thermal expansion has been misused.

And we can do so on "the same page" that has only 33 depths for all purposes,  instead of trying to patch a 7 depth system and a 33 depth system together.

The next post in this series is here, the previous post in this series is here.