Sea level rise slows while satellite temperature 'pause' dominates measurement record

Never mind trying to figure out the level of a constantly moving body of water … The US Canada border is supposed to follow the 49th parallel but the original and subsequently relocated “official” boundary markers are generally several hundred feet north or south of the 49th due to the “accuracy” of the original survey and failure to account for the gravitational pull of the Rocky Mountains on the plumb bobs and variations in the surface elevation gravity and so on. The gravity issue with the mountains was discovered early and corrections made. Now with GPS we can more accurately determine where the border “should have been”. But it is all still based on “agreed to” calculations and marks that we once put on the ground as reference points and our agreed Geodetic surveys. We now have the technology to electronically measure straight lines very accurately but since we live on a spheroid with lots of bumps and dips, we have to apply adjustments. Many of us will remember many measurements are adjusted to “STP”.

Wayne Delbeke at 10:05 regarding borders –
The border of Kentucky and Tennessee was supposed to be a smooth line also. Have a look!

Anyone know why the seasonal rise in sea level, which peaks in mid Spring, is much higher in the Nth hemisphere than the in the Sth hemisphere?

If one assumes that the temperature coefficient of (linear expansion), is constant over the temperature range (from surface to bottom), and also the specific heat, then the sea level rise due to added heat in the sea, is essentially independent of how the heat distributes versus depth.
Excellent point, A+ in fact. I hadn’t really thought of that, but I do believe you are right.
As for Joe’s comments — one actually has to look up the volume expansion coefficient, which is probably approximately linear between 10 C and maybe 80 or 90 C. At 20 C it is around inverse K. Thus:

Assuming that we should be using linear instead of volume (arguably we should use $\alpha^3$, but I’m assuming the ocean is essentially constrained horizontally at all but one surface) and that we’re talking temperature change averaged over only the top kilometer and an annual change of 3 mm:

Ignoring factors of order unity, this means that K/year. Over a decade, this is order of a millidegree. Over a century it is still not resolvable with any but our best thermometers, maybe, at around a hundredth of a degree. In that same century, SLR would be 30 cm.
This is, actually a very serious argument both in favor of and against Trenberth’s assertions of oceanic warming. On the one hand, yes, the ocean is warming, at around 0.1 millidegrees per year (averaged over the top km). On the other hand, it takes a century of such warming to become (maybe) resolvable at the instrumental level, and longer than that to rise above the noise. The only way we actually know the warming is occurring is because of SLR (because there is no way in hell that they are measuring this with only a few thousand ARGO buoys), and that same phenomenon puts strict limits on how large it could be.
rgb

In fact, the theoretical basis for SL rise is not there. The last warming trend ended long ago and the thermal expansion of the ocean has stopped. Greenland ice melt is more than compensated by the growth of the Antarctic shelf ice. NOAA guages on the Gulf coast (where subsidence is the rule) show no increase in SL in the last 16 years (see NOAA mean sea level trends). One exception is grand Isle. La., where subsidence gives a gauge of a false SL rise.
rgbatduke and others touch on SL gauging. It is instructive to see how many of these show no SL rise this century. To utilize this source, see NOAA mean sea level trends gulf coast, or the individual guage charts for Galveston, Sabine Pass, Port Mansfield, Freeport, Port Isabel, Rockport, etc. The data is accurate because of the special conditions of that particular coast.
Willis, for one, will greatly benefit by utilizing this resource.

rgbatduke comments on the difficulty of obtaining accurate SL data via tide guages. This difficulty is due, of course, to tides, storms, and isostatic adjustments.
The Gulf coast offers more reliable data because its tidal fluctuation is small, normally about one foot or so from low to high tide and it is characterized by subsidence, from very slight on the Texas coast to acute at Louisiana, and thus guage readings on that coast are an upper constraint on actual SL rise. The NOAA mean SL trend charts show a good consistency for some six stations from the southern tip of Texas to the La. border. These all indicate little or no SL rise this century, uncorrected for sunsidence. For those who are interested see NOAA Mean Sea Level for Sabine Pass, Galveston (2 stations), Freeport, Rockport, Port Mansfield, and Port Isabel. For a false sea level rise due to acute subsidence, see Grand Isle, La.

For a false sea level rise due to acute subsidence, see Grand Isle, La.
====
Or Galveston too….
Figure 1 shows the sea level trend for Galveston Pier 21, Texas, which has been rising at a rate of
6.5 mm/yr due to land subsidence from oil, gas, and water extraction
http://oceanservice.noaa.gov/outreach/pdfs/sea_levels_online.pdf

thus guage readings on that coast are an upper constraint on actual SL rise.
Or, sadly, not.
The problem is this. Let’s suppose that this is, as some expect, a big ENSO year. Let’s imagine that a huge patch of the Pacific ocean warms by several degrees C over its usual temperature, to some reasonable depth. Let us imagine that — since this water expands as it warms and is buoyed up by the cooler water underneath it — local sea level in the middle of the ENSO warming is a whole centimeter higher than usual.
How much does one expect SL at Galveston to vary, assuming that it is rock solid, stable, not moving?
The answer is “not at all”. The water that was warmed was “floating” (neutral buoyancy) before it warmed. It is “floating” (neutral buoyancy) after it warmed. It displaces exactly the same amount of water that it did before everywhere outside and beneath the local warming.
So, coastal gauge readings “can” actually be significantly disconnected from average SL over the entire ocean. Note that this disconnect can occur either way. If coastal temperatures are warming, coastal SL will (on average) rise, even if ocean temperatures far from shore are falling and the actual mean SL is falling. Or vice versa. Coast tidal gauges are actually subject to “UHI”-linked false SLR to the extent that they live in harbor waters around major urban sites that dump a lot of waste heat or heat absorbing silt into the water. If coastal waters at (say) the Beaufort Inlet in NC (where I live all summer teaching at the Duke ML are (say) a degree or two warmer than the surrounding ocean upcoast, downcoast, and out away from the coast because of silt, trapped haze and pollution from the urban centers around, waste heat coming down the various rivers or sounds that drain in and out there, or because the bay bottom at the inlet was recently carpeted and raised several feet by the sand from around a third of a mile of the tip of Shackleford Island because of Hurricane Sandy, and shallow water warms much faster than deeper water — the tide gauges there will read a teensy bit higher from thermal expansion that is entirely local — the surface water will float a bit higher on the cooler water beneath than it would if it where cooler, and as SL rises from thermal expansion at Beaufort it has no effect on SL anywhere else.
This is the same trick question involving melting sea ice. The ice floats before it melts. It “floats” (displaces the existing water) by exactly the same about — its weight in water — after it melts. So the net displacement everywhere else does not change. I’ve spent the day explaining this to students.
Sea ice is, of course, multiple tricky — as sea ice forms, it desalinates. That’s because salt water has a lower freezing point, so as freshwater crystals form, they grow at the expense of ever saltier residual water that melts its way down through the formed ice to rejoin the sea. Icebergs are invariably fresh water because of this process, even if they were formed directly from freezing salt water. But this is almost a zero sum process. In the winter, freezing sea ice forms, drops its salt into the underlying water, boosting its salinity. In the spring and summer, it melts, adding freshwater to the extra-salty water underneath and diluting it back towards “normal”. It may or may not be a zero sum process because at the same time the thermohaline circulation currents may sweep the extra-dense winter salinity down into the low level return current, replacing it with comparatively less dense water brought in by the top current. Or not. This is, no doubt, one of the forces that drives the THC in the first place. Either way, it is nearly zero sum, so the matched pair of freezing sea water to make sea ice and melting sea ice to make sea water very likely has almost no effect on net displacement.
The only two things that affect global SL are average thermal expansion — which is absolutely dominant in the process at least at the moment — and land ice melt. Land ice melt is definitely not zero sum — it can cause a real rise in sea level. Thermal expansion is ultimately zero sum — SLR measured in one place due to thermal expansion will not generally mean (as one might naively expect) SL to rise somewhere else because of isostatic displacement. The whole SL can rise — because the whole ocean can warm — but as soon as it cools it will shrink right back down, locally, to its normal depth or wherever.
The slow tide gauge SLR observed over the last 140+ years is not, in all probability, a very good measure of true SLR. In fact, it doesn’t guarantee that SL on average has risen at all. It also doesn’t help us understand at all what, if any measurable, fraction of that rise was due to actual melting land ice (and hence is “irreversible” SLR at least until glaciation starts up again globally). What it does indicate, though, is that whatever the “average” ocean is doing (and still neglecting a lot of possible factors that work on geological time, such as tectonic drift, changes in ocean volume due to uplift or opening of ocean floor rifts or the movement of continents or growth of volcanic islands underwater) there is little evidence that we should be alarmed by SLR at the moment. Whether or not we should be alarmed in the future depends a lot more on how warm it has to get to melt Greenland and/or Antarctica, and how long it would take to melt either one even if the air above them both went up by 2-3 C. Since the air above them both is basically well below freezing pretty much all year long (and this would not substantially change if there were a small temperature shift) I’m guessing the answer is “a really, really, really long time”. As in centuries to millennia. Their response might even be counterintuitive — warmer air might come with more moisture, more snow, and net icepack accumulation per annum, on average. With a nonlinear, chaotic system, with complex coupled feedbacks, it is really difficult to make any sort of simple linear response statement and have it work out to be true.
rgb

@Joe Kirklin Born
I never read WSJ (not a popular newspaper here in Brisbane) but I’ll take your word for it. I am quite prepared to believe any insanity associated with the crooks of Wall Street.

Perhaps, rgbatduke, but study the NOAA mean SL trends for The Tx Gulf Coast and you will see a curious spike at around 1998 on these.
That SL gauges do not give reliable indications of SL or SL trends and therefore should not be relied on seems to me to be rather an extravagant statement. What I have heard about Satellite data and instrument drift and theoretical corrections to avoid such and those who operate such, who may or may not subscribe to rigorous principles of science or ideology and so forth, it seems that there is a large degree of uncertainty there, and the rest of us have no means whatsoever of ascertaining any such “drift” or factor of verity.
I prefer gauge data which you can always try for whatever corrections that might seem necessary. Which is not to say that one can remove absolutely all degree of uncertainty.
mpainter