Guest essay by Albert Parker
We may consider all the tide gauges in the latest PSMSL survey of relative mean sea level secular trends (http://www.psmsl.org/products/trends/trends.txt ). The population changes year after year, and since 2015, also the method to compute the rates of rise has changed. As PSMSL says “Please note that we changed the method of calculating relative sea level trends in 2015. The trends displayed here are not directly comparable with any calculated before that date.” So, we will focus only on this survey.
The survey includes 722 tide gauges, some of them having not enough data to infer any reliable trend. The global data set has a naïve average rate of rise of 1.39 mm/year, maximum value of +10.25 mm/year and minimum value of -17.63 mm/year.
As we know the short and incomplete records overrate the relative rate of rise of sea levels, we may then consider subsets.
- If we consider all the tide gauges that started recording before 1934 (S1) this subset of 158 tide gauges has a naïve average relative rate of rise of +0.03 mm/year, maximum +6.75 mm/year, minimum -8.09 mm/year.
- If we consider all the tide gauges with at least 70 years of recorded data in 2014 (S2) this subset of 157 tide gauges has a naïve average relative rate of rise of +0.08 mm/year, maximum +6.75 mm/year, minimum -13.22 mm/year.
- If we finally consider all the tide gauges with at least 60 years of recorded data in 2014 (S3), this subset of 212 tide gauges has a naïve average relative rate of rise of +0.41 mm/year, maximum +9.41 mm/year, minimum -13.22 mm/year.
The longer subsets may serve to assess the presence (or absence) of an acceleration, as the relative sea level rates of rise become significant only after the minimum 60 years of data are recorded. The different values only reflect the different populations, with more tide gauges recently being established in areas subject to subsidence rather than uplift. The subsidence of the instrument is still the most relevant component to sea level rise.
What we learn from this survey? The sea levels are not rising, but rising and falling. And in the best “spots” along the world coastlines where the sea level rises are measured and not computed, the naïve average rate of rise is a pretty constant value and a quite small value.
These naïve average relative rate of rise translate in a naïve average sea level rise over the first 15 years of this century no matter the anthropogenic carbon dixoide emission of a little bit less than half a millimetrie to little bit more than one millimetre in the 158 or the 157 long term tide gauges S1 and S2, or at the most in 6 millimetrs in the 212 tide gauges S3 satisfying the minimum requirement of 60 years to infer a reliable trend.
Figure 1 below is the histogram of the S2 data set (tide gauges with more than 70 years of recorded data), image from the Wessa online facility.
The naïve average is practically zero, somewhere the sea level rise, somewhere else the sea level falls. We do not need more and more layers of burocrats and more and more taxes with the excuse to save the world from the rising seas because we burn carbon and hydrocarbon fuels.
Wessa P., (2015), Histogram (v1.0.15) in Free Statistics Software (v1.1.23-r7), Office for Research Development and Education, URL http://www.wessa.net/rwasp_histogram.wasp/
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
It is hard to make sense of all those varying limits as well as the fact that some gauges show a sea level drop. That is normal for formerly glaciated areas like Scandinavia, Canada and such inland locations like the great lakes. I personally like to stick with Chao, Yu anf Li that came out in February 2008. What they did was to correct all published sea level data for water held in all reservoirs built since 1900. When this correction was done they had a sea level curve that was linear for the previous 80 years. Its slope was 2.46 millimeters per year. Something that has been linear that long is not about to change anytime soon. 2.46 millimeters per year works out to just under ten inches per century. It is likely that the true centenarian sea rise will be close to ten inches, give or take perhaps an inch. Compare this to the twenty feet Al Gore pronounced in his famous movie that brought him the Nobel Prize. Twenty feet, he said, would wipe Florida off the map. He even showed a map of Florida under water and the Nobel Committee said, yeah, verrily, though dos’t know the future and deserve our prize for that.
The question is not what is the average sea level rise from the tide gauges over the last 60 years, the issue is what is happening in the recent 25 years – since the adjusted satellite record says that sea level rise increased from about 1.4 mm/year to about 3.2 mms/year.
The tide gauge database from PMSL is deliberately structured so that one cannot really get into the annual numbers and say what the tide gauges record is for 1992, 1993, 1994, …. 2013, 2014. This is what we should be concerned about.
And then there is the issue of the local subsidence/uplift rates of each individual tide gauge station. Fortunately, this seems to be a relatively stable number over many decades if not centuries.
Many of the tide gauge stations are now co-located with a GPS station that can nail down the local subsidence/uplift level after about 4 years of continuous operation.
Sonel.org has been maintaining a database of these tide gauge / GPS co-located stations.
For example, Sydney Australia, has a tide gauge recording 1.15 mm/year of sea level rise (and there has not been any recent or historical acceleration) and the co-located GPS station is saying that Sydney harbour is sinking by 0.89 mms/year. Thus the real sea level increase is only about 0.26 mm/year at Sydney.
http://www.sonel.org/spip.php?page=gps&idStation=2405
http://www.psmsl.org/data/obtaining/stations/196.php
Let’s take a northern Europe example as Willis noted; FURUOGRUND on the Baltic Sea which has a tide gauge measuring -6.46 mm/year of sea level decline but the land is rebounding from the last ice age at a rate of about +10.58 mms/year so the net sea level rise here is 4.12 mm/year (one of the highest net sea level rise numbers).
http://www.sonel.org/spip.php?page=gps&idStation=844
http://www.psmsl.org/data/obtaining/stations/203.php
Basically, one would really need to do this same thing for about 300 stations and then calculate what the net sea level rise really is for all 300 stations to get a really robust result. And then we need that data by YEAR. This is a lot of manual data entry and calculation.
Sonel has calculated the net sea level rise from 1983 to 2013 (they will only provide 30 year averages) for 106 stations which have tide gauges and co-located-robust-data-GPS stations …
http://www.sonel.org/-Sea-level-trends-.html?lang=en
(one needs to play around here a little to figure out to download the data]
AND …
… the net sea level rise from these 106 stations is only 1.86 mm/year. [1983 to 2013 – which means the satellites can not possibly be correct at 3.2 mms/year.
“Sonel has calculated the net sea level rise from 1983 to 2013 (they will only provide 30 year averages) for 106 stations which have tide gauges and co-located-robust-data-GPS stations …”
The 30 year averaging appears to be a 1st order attempt to handle ocean circulation patterns, wind patterns, etc. However, I’ll accept the number they derive for SLR when they get the same (corrected) reading from each gauge in their network within the accuracy and precision of the gauge. Until then they have neither defined nor adjusted for all variables other than SLR that may effect the readings. Until then they don’t know what they are actually measuring.
Changes in sea levels, or more accurately the rate of sea level rise, is more complicated than the analysis presented above suggests.. The rate of sea level rise shows a fluctuation which is synchronous with the Atlantic Multidecadal Oscillation. See:
http://www.climatedata.info/discussions/blogger/index.php?id=8783687603052213393
The PSMSL surveys (as the NOAA surveys) of relative sea level rises are extremely relevant.
These surveys tell you where along the coast the sea levels are rising or falling, and if they are accelerating (just compare the relative rates of rise of different surveys in the same tide gauges where only few more years of data are added).
What the PSMSL and NOAA surveys tell you is that almost nowhere there is a need to cope by coastal management with the rising seas.
Where they are measured properly, the relative sea level rises are everything but dramatic and they are about constant.
So the 2 meters sea level rise added everywhere in the United States as per the TMAC report are a disgrace for science.
If somebody tell you that 2 m sea walls are needed along the west coast of the United States and Alaska, you may answer in Alaska the sea levels are mostly falling, and along the west coast of the United States and Alaska the sea levels are rising about 1 mm something on average … so it may take forever to make plausible the 2 m sea walls in Alaska and possibly 2,000 year in California, Oregon and Washington state.
Please do NOT get trapped in the “absolute” sea level rise. You cannot know with accuracy the “absolute” sea level, and what flood the land is the relative sea level …
final remark, just to be precise about Sydney
I do not know where the claim the relative sea levels in Sydney are rising 1.15 mm/year.
If you look at the PSMSL survey I proposed, it says +0.65 mm/year in Fort Denison 1 (longer) and +0.96 mm/year in Fort Denison 2 (shorter). The two tide gauges are leveled each other and overlapping, so you can make a composite record with pretty much same results of Fort Denison 1, only little bit more 1886 to 2013.
If you look at the SONEL GPS velocity of an inland GPS dome, this is not the vertical velocity of the tide gauge, but the vertical velocity of an inland GPS dome. Fort Denison is a small island in the bay.
SONEL says -0.89 mm/year. This means strong subsidence. for the same GPS dome, JPL says -0.465 mm/year, so subsidence but less. Not to discourage you guys, but Australia is not subjected to subsidence. The problem of the GPS is a constrained computation, and if you want to overrate the uplift of Europe (SONEL) or the uplift of the United States (JPL) to claim “absolute” sea level rate of rise exceeding the relative, somewhere else, for example Australia, you do have negative absolute rates of rise in Sydney and Fremantle when the relative rates of rise are actually positive.
This is the principle of the short blanket. you cant cover shoulders and foots.
Again, please do NOT get trapped in the “absolute” sea level rise. You cannot know with accuracy the “absolute” sea level, and what flood the land is the relative sea level … and please remember to support the integrity of the data sets in PSMSL and NOAA to avoid the administrative adjustments we already experienced in many products.
I hope you enjoyed the latest PSMSL survey, with only the records too short or too incomplete removed.
The PSMSL surveys (as the NOAA surveys) of relative sea level rises are extremely relevant.
These surveys tell you where along the coast the sea levels are rising or falling, and if they are accelerating (just compare the relative rates of rise of different surveys in the same tide gauges where only few more years of data are added).
What the PSMSL and NOAA surveys tell you is that almost nowhere there is a need to cope by coastal management with the rising seas.
Where they are measured properly, the relative sea level rises are everything but dramatic and they are about constant.
So the 2 meters sea level rise added everywhere in the United States as per the TMAC report are a disgrace for science.
If somebody tell you that 2 m sea walls are needed along the west coast of the United States and Alaska, you may answer in Alaska the sea levels are mostly falling, and along the west coast of the United States and Alaska the sea levels are rising about 1 mm something on average … so it may take forever to make plausible the 2 m sea walls in Alaska and possibly 2,000 year in California, Oregon and Washington state.
Please do NOT get trapped in the “absolute” sea level rise. You cannot know with accuracy the “absolute” sea level, and what flood the land is the relative sea level …
final remark, just to be precise about Sydney
I do not know where the claim the relative sea levels in Sydney are rising 1.15 mm/year.
If you look at the PSMSL survey I proposed, it says +0.65 mm/year in Fort Denison 1 (longer) and +0.96 mm/year in Fort Denison 2 (shorter). The two tide gauges are leveled each other and overlapping, so you can make a composite record with pretty much same results of Fort Denison 1, only little bit more 1886 to 2013.
If you look at the SONEL GPS velocity of an inland GPS dome, this is not the vertical velocity of the tide gauge, but the vertical velocity of an inland GPS dome. Fort Denison is a small island in the bay.
SONEL says -0.89 mm/year. This means strong subsidence. for the same GPS dome, JPL says -0.465 mm/year, so subsidence but less. Not to discourage you guys, but Australia is not subjected to subsidence. The problem of the GPS is a constrained computation, and if you want to overrate the uplift of Europe (SONEL) or the uplift of the United States (JPL) to claim “absolute” sea level rate of rise exceeding the relative, somewhere else, for example Australia, you do have negative absolute rates of rise in Sydney and Fremantle when the relative rates of rise are actually positive.
This is the principle of the short blanket. you cant cover shoulders and foots.
Again, please do NOT get trapped in the “absolute” sea level rise. You cannot know with accuracy the “absolute” sea level, and what flood the land is the relative sea level … and please remember to support the integrity of the data sets in PSMSL and NOAA to avoid the administrative adjustments we already experienced in many products.
I hope you enjoyed the latest PSMSL survey, with only the records too short or too incomplete removed.
This is very interesting. Could you say what is the basis for the assertion that “Australia is not subjected to subsidence” ?
Does this apply universally to all sites in Australia, does it apply to islands like Fort Denison?
Neither do I follow your shortest blanket. If they want to rig Europe and N. Am. how does this make Oz go down. Calling GPS a short blanket is not a satisfactory account. Could you be clearer about what these assertions are based on?
It is practically zero and totally naïve. You’d have to be totally naïve to think it told us anything.
I thought when the article stated with a statement about the naïve average, it was going to go on and say “BUT when we do XXXX we get ….. ”
But no, it is totally naïve averages all the way down. IMO it was a waste of time writing it. There is no reason given for why we should look a naïve averages or why we should think they are better than more thorough analyses which attempt to make account for land movement.
You’d have to be totally naïve to think it told us anything.
Also the histogram at the end shows that the data have a strongly skewed distribution, so taking a naïve average is not appropriate. The mean is only effective at removing ‘normally’ ( or gaussian ) distributed errors.
In this case it will give a low biased result because of the long tail on the left of the distribution.
Like temperature data, sea level data is averaged and manipulated beyond all usefulness.
Since coastal flooding is likely to occur only when there are king tides, why not just measure and plot the level of king tides? The actual height can be plotted on a graph without any calculations.
The data at the NZ and Australian sites I have checked show that the height of king tides has been reducing for some years now. Those sites are NOT rising or falling. So, no flooding.
You will get the same result if you put a mark at king tide time on a large immovable object by the sea. You will observe the marks keep dropping each year.
Please stop using “mean tide” as it’s useless at predicting anything.
There are two separate things that can be sought form sea level data.
1. an increase in absolute mean sea level , a component of which can be compared to global mean sea temp. change.
2. Local relative sea level which affects peoples lives.
The alarmists take 1. , add a bit GAIA “correction” and ‘inverse barometer’ for good measure and then use it to make scare stories about 2.
GAIA correction has ZERO effect on 2.
How many years? There are 18y cycles in tides, these are known and predictable. Hurricane Sandy coincided with just one such event.
There are islands off Sumatra which dropped down after the 2004 and 2005 earthquakes, where the relative sea level rose up to a few metres within a few hours. Some villages were flooded at a rate far exceeding the current rate of sea level rise of around 1mm+ /year; and nobody died because of this subsidence. The villagers just packed up and moved further inland. Why is this such a big deal?
Meanwhile, back in the real world, the rates of malaria, malnutrition, poverty, lack of education and infrastructure, preventable illness-related deaths and corruption on these islands, remains very high.
Which of these issues above do you think that locals think is the least important?
if you guys look at the latest TMAC report
http://www.fema.gov/media-library-data/1454954261186-c348aa9b1768298c9eb66f84366f836e/TMAC_2015_Future_Conditions_Risk_Assessment_and_Modeling_Report.pdf
you may find as “settled science” Highest Scenario SLR by 2100 (m) 2.0 (ft.) 6.6 …
now, to get 6.6 ft. it will take much more than the 84 years left of this century ….
in the 158 locations with record started before 1934, it will take on average 57,982 years, or better to say
never ever this will happen in 54 locations where the sea level is reducing
1,143 years in the 97 locations with average sea level rise 0-3.25 mm/year (average sea level rise 1.75 mm/year)
381 years in the 7 locations with average sea level rise >3.25 mm/year (average sea level rise 5.25 mm/year)
in the US, where the sea levels are rising or falling with an average well below 2 mm/year, why do you want to pay your insurances today big $$$$$$ for something that is unlikely to happen before 1,000 years from now?