As we surmised in earlier posts, the crozon.colorado.edu website was a test run. Here’s the newest graph from the revised http://sealevel.colorado.edu/
They write about the update:
Welcome to the new webpages from the University of Colorado sea level group! We apologize for the delay in updating our sea level releases, but the transition to these new web pages took longer than we thought. In addition, we have made many improvements to our data (new orbits, new tide model, new corrections) which ultimately had little effect on global mean sea level, but brought us up to date with the latest advances in the field.
One important change in these releases is that we are now adding a correction of 0.3 mm/year due to Glacial Isostatic Adjustment (GIA), so you may notice that the rate of sea level rise is now 0.3 mm/year higher than earlier releases. This is a correction to account for the fact that the global ocean basins are getting slightly larger over time as mantle material moves from under the oceans into previously glaciated regions on land. Simply subtract 0.3 mm/year if you prefer to not include the GIA correction.
You may also note that rate of sea level rise over recent years has been less than the long-term average. This is believed to be due to the recent La Nina’s we have been experiencing, though research on this is continuing. We will soon add a plot to the web site illustrating this effect.
Let us know if you spot any bugs in the new web pages. Thanks for your interest!
Comments welcome.
@ur momisugly P Solar: I’m afraid you are reading more into the GIA adjustment than it deserves. It’s all about comparing like with like. Due to GIA the ocean basins are physically getting larger which – by itself – is reducing apparent sea levels by about 1.2 inch/century (0.3 mm/year). It is entirely consistent with scientific principles that this effect is recognised for what it is and dealt with in the data.
Now the awkward part: there are two ways of handling this. Every 100 years (pro rata for part) CU will know that their sea level data 100 years beforehand needs to be lowered by 1.2 inches (or whatever the precise figure then is) to be consistent with current measurement. They can either (a) adjust and republish all past data by lowering past sea levels retrospectively (wouldn’t that just produce howls of protest) or (b) adjust the current data by reference to a “standard basin size” datum to compensate for the basins becoming larger, which (as you have mentioned) seems to produce a higher current sea level readings – until everybody understands what is actually going on.
CU also know that the (modest) rate at which sea levels are rising isn’t the important figure. The absolutely critical figure is any change (acceleration or deceleration) in the rate of rise – the 2nd order differential by time. This isn’t affected at all by the GIA adjustment, so CU obviously considered that the “standard basin size” datum was the right way to go.
TimC says:
May 6, 2011 at 11:36 pm
If an ocean basin gets larger/deeper then the sea level will drop. This is “apparent” as in real, visible and measurable. Not “apparent” as in paradoxically wrong.
Any suggestion that it needs to be “corrected” comes from the idea that GMSL represents something other than sea levels. What could that something be, one asks.
Clearly GMSL is one candidate for the “other metrics” of climate change that are being called for. It does carry a signal that could be a primary input to calculating total ocean volume, mass or similar that could in turn be used as a climate metric. However, such a measurement or inference of climate cannot come from GMSL alone, there are (many) other factors like GIA that need to be accounted for.
The need to account for other factors does not mean that those other factors should be used to *corrupt* the GMSL data by being applied de facto as “corrections” or “adjustments”.
In that case we need to CALL it something else and stop calling it mean sea level data.
I’m not seeing a conspiracy, I’m seeing bullshit science and calling such.
If anyone wants to redefine sea level to be some intangible, notional datum that hovers somewhere above the physical sea level and gradually rises ever more above it, they need to come up front and say so very clear terms.
Completely bizarre correction. Either the sea level is going up or it s going down presumably its measured against some geodetic surface …. and there are many factors that might make it go up or go down …. why suddenly not include one that makes it go down ???? Lets just take that logic to an end member state … lets just assume that this GIA was a much larger effect .. say -5mm/year. So even though sea level was actually going down, they would state that it was going up …. bizarre.
The truth is they couldn’t show an updated graph with sea level rise now only being 2.8mm/yr. That just wouldn’t fit the story that this problem is accelerating – its worse than we thought.
@Green Sand – “NOBODY expects the Glacial Isostatic Adjustment!
Thanks for the LOL 🙂
@Doug Proctor – Cheers Mate
Dave
@ur momisugly P Solar: thanks for that. Let’s take a worked example: suppose (to keep it simple) at 2000.0 the ΔMSL figure was 8.0 mm (no GIA adjustment) and the GIA adjustment was 6.0 mm. In 2100.0 suppose the figures are then ΔMSL 318.0 mm (no GIA adjustment) with GIA of 36.0 mm.
These figures are enough to say it all, whichever way round you wish to give out the data.
Your way: “In 2100.0 we know that a century beforehand sea levels were actually at 8.0 mm. Due to GIA effects that figure has now become -22.0 mm, against the current actual figure of 318.0 mm, a MSL rise of 340.0 mm”.
My (and I think CU’s) way: “In 2000.0 the GIA adjusted MSL figure was 14.0 mm. The adjusted figure today is now 354.0 mm, a MSL rise of 340.0 mm. The current GIA adjustment is 36.0 mm”.
I really don’t think much turns on it, either way!
@ur momisugly ImranCan: the measurements are by internally calibrated satellite altimeters. I don’t think they are specifically referenced to the geoid but just presented as ΔMSL figures (changes in MSL as from a given reference date/time).
The GIA adjustment is to give a more accurate picture of the volume of water in the oceans (resulting from the hydrological cycle and melting land-based ice): if you want to go from levels to volume you have to adjust for any changes in the size of the bucket!
so to summarize they now publish a virtual sea level that would have existed without GIA ?
looks very much like the virtual astrological signs that would be correct without the precession of the equinoxes, doesn’t it ?
When the Sea level was higher! http://www.photographers-resource.co.uk/a_heritage/Castles/LG/Wales/Harlech_castle.htm
TimC May 6, 2011 at 11:36 pm
“[…] the 2nd order differential by time. This isn’t affected at all by the GIA adjustment […]”
What is the academic mainstream rationale for assuming constant GIA?
P. Solar & Others, When mucking around with models of a short series, discrepancies between models at high-leverage segments (such as segments containing sharp changes) are to be expected. It may be true that some modelers like the new model better for whatever reason, but that doesn’t mean they’ve carefully run diagnostics. While malicious intent cannot be excluded as a possibility for some players in the game, there is plenty of reason (even upon a mere preliminary glance) to suspect a strong role for ignorance & naivety.
I don’t get it.
If mantle material is moving from under the oceans to compensate for land rise because of glacier melt, it means that the sea basin gets deeper, thereby hiding the increase in sea water.
But what is Colorado actually measuring? Changes in sea level or changes in the total amount of sea water?
It seems to me they overnight changed interest from one to the other
Ray Boorman says:
Even though the actual sea level is what they purport to measure, they decide to add a small positive adjustment every year for the fact that some parts of the land are rising due to rebounding from the last ice age. Measurements are measurements where I come from, & if my local tide guage showed a rise of 2.7mm a year because the land is rising by 0.3mm, than that must be the true reading. Do they also adjust for the fact that any rising land areas are also increasing the sea-level rise in those areas which are not rising? I can’t see how this adjustement is anything other than an artifice to satisfy the CAGW believers.
Even where the land is rising it could be at less than 0.3mm a year. Also there’s plenty of places where land is subsiding. (In some cases considerably more than 0.3mm per year.)
Dave A says:
From http://sealevel.jpl.nasa.gov/technology/
From its vantage point 1336 kilometers (830 miles) above the Earth, the US/European Jason-1 and OSTM/Jason-2 ocean altimeter satellites measure the height of the ocean surface directly underneath the satellite with an accuracy of 4-5 centimeters (better than 2 inches).
n.b. centimetres not millimetres 🙂
So where does the 0.4mm accuracy claim on the graph come from? If the original data is only accurate to 40-50mm. Any data processing tends to increase the level of inaccuracy, in certainly isn’t going to reduce it by two orders of magnitude.
You know, I find the stated error of 0.4 mm completely bogus. Look at the graph for any period. The latest data runs from 25-45 mm and is completely random. Which means the +/-10 mm range in data is PROBABLY the noise in the data. Gimme me a break. When did science lose its basic math skills.
See the two very interesting graphs on the bottom of this page:
http://climate4you.com/SeaTemperatures.htm
Agust
P. Solar, have you considered that triangular filters based on harmonics of modes such as the day & year might provide slightly more sharply informative summaries? (Your Gaussian filter is actually not terribly different in shape, but technically (not practically) it does complicate interpretation. I acknowledge that this is a rather technical point.)
The “Colorado correction” – I fear the name will stick.
GIA adjustment is only part of the story. The trend resulted from data is different to the one before, the relative values of the last years to each other have changed significantly.
How can historical data change? It reminds me of the temperature record where 1934 keeps on getting colder with the time, same with 1998 (but not so fast).
Look at 2002-2010 in the new and old graph. It is totally different giving a different trend, more then GIA adjustment would give.
This cannot be with real data.
Btw, if the GIA effect would be real it would mean the oceans should be now 60 cm lower then at roman optimum.
@ur momisugly Paul Vaughan: re your query I can only refer you to CU’s own quoted intro at the head of this article: “One important change in these releases is that we are now adding a correction of 0.3 mm/year due to Glacial Isostatic Adjustment (GIA), so you may notice that the rate of sea level rise is now 0.3 mm/year higher than earlier releases. This is a correction to account for the fact that the global ocean basins are getting slightly larger over time as mantle material moves from under the oceans into previously glaciated regions on land. Simply subtract 0.3 mm/year if you prefer to not include the GIA correction.”
My point was that any fixed, constant, rate adjustment of 0.3 mm/year will of course (a priori) show up as a rate change in the first order time differential, but will have no effect on the second order time differential figures (any changes to the rate of rise in MSL).
Other than that that I think you had better check the details from CU’s own website! For my money this is a sensible, practical change to let us see more clearly the data that is really important (the second order time differential figures) – but it seems to be causing a lot of angst around here!
Agust Bjarnason says: May 7, 2011 at 10:05 am
See the two very interesting graphs on the bottom of this page:
http://climate4you.com/SeaTemperatures.htm
Thank you. I’ve added climate4you’s Global Mean Sea Level Change graph;
http://climate4you.com/images/UnivColorado%20MeanSeaLevelSince1992%20With1yrRunningAverage.gif
to WUWT’s Global Climatic History Reference Page;
http://wattsupwiththat.com/reference-pages/global-climatic-history/
Ocean Reference Page;
http://wattsupwiththat.com/reference-pages/ocean/
and ENSO Page:
http://wattsupwiththat.com/reference-pages/enso/
In addition, I’ve changed the title of the University of Colorado graph on all pages to Global Mean Sea Level Change with a “Correction” of 0.3 mm/year added May, 5th 2011, due to a “Glacial Isostatic Adjustment (GIA)” and added links to their Calibration and Release Notes pages in the Source Guides at the bottom of each Reference Page.
The Release Notes;
http://sealevel.colorado.edu/content/release-notes
deserve scrutiny, as the most recent release has many more changes than any of the previous releases noted. Why?
Measuring sea-level that’s SEA LEVEL rise to an accuracy of .3mm…great and while we’re at it are there any Theologians out there who can tell me just exactly how many Angels can dance on the head of a pin?
Yes, they always have to UPdate the new data and DOWNgrade the old data to get a nice spike in measurements that can only be attributed to CO2 forcing. It’s the new science, if the data doesn’t fit your hypothesis than the data is obviously wrong.
A G Foster says: “The earth can’t be speeding up while ice is melting at catastrophic rates.”
The argument here, as I understand it, is that a rising sea level increases the Earth’s diameter and hence LOD must increase. Someone please say if I’ve got it back to front, because it seems to me there’s a factor going the other way, and it’s this : sea level is affected mainly by melting land-based ice rather than by melting sea ice. But the land-based ice starts off above sea level, in some cases a long way above sea level, so by melting it moves in towards the Earth’s centre, not away. Hence its effect is to shorten LOD, not lengthen.
P. Solar says: “The rate of change plot shows the differences in the new data that are not obvious in the straight difference Mike Jonas provided here : http://members.westnet.com.au/jonas1/ColoradoUniDiff.jpg”
The linear trend in the straight differences (MS Excel Trend() function) is actually 0.24mm p.a. (4.3mm over 17.8 years). Maybe the wild changes up to 2002 (the TOPEX period) make it hard to see.
—
I have done another couple of graphs, using the new data, which I think are illuminating: I did a least squares linear fit for two line segments, where the ‘flex point’ was itself optimised not pre-chosen:
http://members.westnet.com.au/jonas1/ColoradoUni2PartTrend.jpg
It clearly shows how the rate of sea level increase has slowed (from ~3.53mm p.a. to ~1.94mm p.a).
The second graph has the first trend line extended:
http://members.westnet.com.au/jonas1/ColoradoUni2PartTrend_extended.jpg
It shows that the 2010 El Nino was insufficient to bring the sea level back to the original trend. To my mind, that indicates that the cause of the lower recent rate is not the La Ninas, as Colorado Uni suggest (they don’t claim it, only suggest it), but could be the PDO phase. The reasoning behind this is that El Ninos in the past appear to have caused at most temporary increases, not a change in trend. There is presumably no reason why La Ninas should behave differently.
I have emailed the graphs and this suggestion to Colorado Uni for their comment.
Consider me skeptical
Anyone else suspicious as to the timing of these adjustments?
Surely they didn’t just now learn that there is a GIA effect (although I doubt they really understand quantitativly how to calculate the effect).
It seems they did not like the story the raw data was now indicating and they needed to muddy the water to keep the CAGW agenda alive by suddenly introducing a new factor. At least they did not hide the fact that new adjustments were added, but the CAGW folks have a story to ignore the decline that will never reach the MSM propaganda.
As I recall the same thing happened with the ARGO sea buoy data. The raw data had to be adjusted before it could be released since the facts did not support the claim that the heat is accumulating in the ocean.
I’m an engineer, show me the raw data, adjusted data is always subject to intense scrutiny.
Paul Vaughan says:
May 6, 2011 at 9:11 pm
A G Foster, What’s the academic mainstream rationale for assuming constant GIA?
It’s based on the fact that the ice load from the Last Glacial Maximum has been gone for 8,000 years. Over the long haul it’s not constant, but asymptotic. Moreover the LIA could be complicating things–much melting, warming, and rebound could be due to the LIA, a lot smaller but more recent than the LGM. LOD history since Galileo strongly suggests a response to the LIA. The current static LOD could very well be due to LIA GIA. –AGF
Mike Jonas says:
May 7, 2011 at 3:35 pm
A G Foster says: “The earth can’t be speeding up while ice is melting at catastrophic rates.”
George Smith could sympathize with you. Elevation has nothing to do with it–the inertial arm is measured as the cosine of the latitude: near zero at the poles and near unity near the equator. Of course the ocean isn’t sitting on the equator–its effective average is 63 degrees as I recall. So any land based ice beyond that latitude (north or south–sea ice doesn’t do anything) will slow the earth down when it melts–until it’s done and inelastic rebound kicks in. Then the earth speeds back up.
This stuff got kicked around a few weeks back in a discussion on LOD stat analysis–forgot the author’s name. Mulholland posted a reference. –AGF