The Ups and Downs of Sea-level Recording

Google sometimes comes up with odd results - not the wrong results I hasten to add, but something unexpected, something novel, something revealing. Anyway, one result was a pdf which referred to Anthony Watts (Wattsupwiththat) as a "sea level expert", with a link to a WUWT post which contains the following:

IMHO The idea that a dock (or piling) is a long term stable measurement platform is simply ludicrous. Piles sink, structures decay, boats whack them, pounding wave action loosens their grip. One feature missing from all these old style tide gauges is any way to reference the long term level of the gauge itself. In the era of GPS we can start doing this, but in the years past, how much is from simple sinking of the pilings over time? When you are looking for millimeters per year, such things become significant.

Anthony Watts would hardly describe himself as a "sea level expert", but that's by-the-by. The post was discussing sea-level past and present in the Maldives, and curiously for such a topic, presented no sea-level data whatsoever. However it's Anthony's statement "One feature missing from all these old style tide gauges is any way to reference the long term level of the gauge itself" which I'm concerned with here. I could criticise him for that statement, written in ignorance, but I'll just ask the question "Can you think of any method of establishing the level of land surface, buildings, or anything attached to the land?". if you can't, go to the bottom of the class. If you thought "surveying" or it's more technical term "levelling" you go to the top. There is indeed a "way to reference the long term level of the gauge itself". Here's a photograph of a  surveyor engaged in that very process next to the Funafuti, Tuvalu SEAFRAME station in January, 2009:

Source: SOPAC

He's using a modern state-of-the-art instrument termed a "total station". The model he's using is one of the best. It can measure to 0.5 arc-seconds in elevation and traverse - that's 1/7200 of a degree, and to 1 in 1,000,000 in line-of-sight distance, which is 1 mm in 1000 metres. The latter is achieved using laser-ranging, also used to continuously monitor the height of the instrument above the benchmark below to 0.1 mm.

I can tell you the make, model and serial number of the instrument he's using (LEICA Total Station Model TCA1800, S/N 424936), his name (Andrick Lal) and the organisation he still works for - SOPAC (Applied Geosciences Division). He was there in 2009 with Nick Brown, a  surveyor for Geoscience Australia. The details of the survey (as for all others) are published in reports on the GA ftp server and many are referenced on the GA website.

They were actually establishing the exact location and height of the tide-gauge benchmark (TGBM) wrt to the benchmarks for the CGPS (continuous GPS) pillar which is located some 2.5 km away from the gauge, but Australia's National Tidal Centre surveyors use identical equipment and techniques every year or two - this is the only Tuvalu survey I've been able to find photos for. There's a chain of an additional 12 BMs between the tide-gauge BM and the CGPS pillar BM - 14 in all.
Levelling the Funafuti CGPS pillar:

Source: SOPAC

I read somewhere that these surveys always seem to take longer on the Pacific islands than elsewhere - I can't imagine why. can you?. BTW levelling doesn't involve any adjustment - it's a measuring exercise. Some claim that CGPS is used to "automatically adjust" gauges. Nothing could be further from the truth - those that make such claims generally include a diagram of a SEAFRAME station which shows the CGPS station firmly on land. Apart from that it's not practical to do so. Few are aware of the "crustal tide" caused by the Moon, which pulls the Earth's crust up and down, synchronised with the ocean tides. At the sites of the  Funafuti stations shown above, it amounts to around 10-20 cm at peak; any automatic adjustment would lower the gauge as the ocean tide rises, and vice versa.

The point I'm making is that this process is carried out regularly and with great care and the results made available on the 'net. Not all gauge sites worldwide are levelled as frequently as those in the South Pacific Sea Level Monitoring Programme however. Most aren't the focus of attention that these islands have become. Many are relatively stable and need less frequent checking. Note that even at other sites with CGPS stations on both the gauge and on land, regular levelling takes place. The benchmark which marks the datum (reference level) for measurements is referred to as TGZ - tide gauge zero. Sometimes the nearest benchmark is above lowest water and a fixed positive offset is added to the gauge readings to give an always-positive reading - from a "virtual benchmark" in fact.

Here's a picture (looking SE) of another gauge which is relevant to the secondary theme of this post. It's the Male-B station in the Maldives, northern Indian Ocean.The gauge, or rather gauges, are on Hulule island where the airport is situated, adjacent to Malé itself.

Source: SONEL

Source: University of Hawaii Sea Level Center

The people in the background on the right aren't back-packers but are part of a University of Hawaii team, on-site to check and maintain the equipment. Looks like a helluva job to me. I don't know how they can stand all that sunshine and sea air.
I include the following description because some people (no names, no pack drill) like to give the impression that sea-level data is recorded by a man with a marked stick and a clipboard, when he can be bothered so to do. The Malé station currently has three independent gauges; a float gauge with the small dome atop on the left, a bubbler gauge with the box on top, and a radar gauge on the black bracket over the water. The two large vertical "pipes" are "stilling wells". They aren't simply open at the bottom, but have holes near the lower ends which restrict the rapid influx or efflux of water (hence "stilling") to eliminate the effect of waves, the wash of a passing boat, or a diving horse, perhaps.

The potentiometer at the top of the float gauge. The toothed metal belt carries the float.

Looking down the float gauge stilling-well; the float is centre-right at the water surface.
Source (both photos): University of Hawaii Sea Level Center

A bubbler gauge releases nitrogen gas into the water near the bottom of the stilling well. At the depth of the release valve the pressure varies with the depth of the water above, and a pressure recorder monitors the pressure of the gas as it's released. Pressure varies linearly with depth so it's easy to convert the pressure readings into an accurate depth. A temperature sensor records water temperature so consequent changes in water density and gas pressure can be allowed for.

The workings of the float gauge, and the radar gauge over the water should be fairly obvious. The dome on the central pillar contains the receiver and aerial for the CGPS (continuous GPS) station which constantly records the E-W, N-S and vertical coordinates (position) of the station. Benchmarks at the station, and three sites around up to a kilometre away provide for levelling by surveying instruments. The station collects data and transmits it to a satellite at hourly intervals and it's been recording since 1990. There's another GPS (& laser telemetry) station located some distance to the NE which is part of the DORIS system, a worldwide network used for precise determination of the JASON-1 satellite's orbit. Here's what Male-B recorded between 1990 and 2010. Bear in mind that the CGPS station recorded some millimetre-sized ups-and-downs over the last decade but they averaged out to just -0.1 mm/year fall.

Data source: PSMSL

To the south, at Gan on Addu Atoll, is another float-type gauge, with a slightly longer record but which shows a similar 13-month profile and annual trend.

Data source: PSMSL

I've never seen these charts (or any earlier version from any source) shown on any sceptical blogs. Nils-Axel Mörner has written screeds on the Maldives, but I've never seen them shown therein either. Do they show something inconvenient? What, precisely, is wrong with the unvarnished truth? Mörner has referred to them however in a 2004 paper titled "New perspectives for the future of the Maldives". I don't know when he visited the Maldives, as rather curiously, the paper doesn't give any dates.  He states in his conclusion (multiple authors, but they're his words, I'm sure)

Tide gauge data have been cited in support of an on-going rise of mean sea level (Singh et al., 2001). Tide gauge records, however, do not provide simple and straight-forward measures of regional eustatic sea level. They are often (not to say usually) dominated by the effects of local compaction and local loading subsidence. With this perspective, our multiple mor-phological and sedimentological records appear more reliable and conclusive. Besides, satellite altimetry does not record any significant rise in global sea level in the last decades (Mörner, 2003a, Fig. 2). In order fully to investigate the situation, however, available tide gauge records, now extending from 1990 to 2002, were re-examined. This reveals a total absence of any rising secular trend (Mörner, 2003b).

He's saying that tide gauge records don't provide reliable indications of sea level trends, but uses them to prove a point he's making, because he claims that what they show conveniently proves that point? There's a word (several words in fact) for that kind of thing. It's also difficult to understand how 13 to 14-year-span gauge records could possibly reveal anything about the presence or absence of a "secular trend", as in this context "secular" means long-term. He also says "extending from 1990" - the Gan record starts in 1988. Incidentally, the reference "Mörner, 2003b" was never published. There's a whole story here but it's for a future post.

I take issue with his statement that "Tide gauge records, however, do not provide simple and straight-forward measures of regional eustatic sea level. They are often (not to say usually) dominated by the effects of local compaction and local loading subsidence". They are not so "dominated", but many are affected. Many others are affected by just the opposite, where the land is rising, and sea-level rise is understated, something he fails to mention here. His term "regional eustatic sea level" is confusing and ambiguous  It's a term he's proud to have invented - eustatic means worldwide or global so he's effectively using the term "regional global sea level". It's worse than confusing and ambiguous. It's meaningless, a contradiction in terms.

In my opinion, and in the opinion of others too, the worldwide tide-gauge network provides a more reliable source of environmental data than do thermometers in weather stations. Tide gauges have inherent sources of error, they're subject to vertical movement caused by subsidence or rebound of their mountings (piers, jetties, etc) and the land or seabed beneath, but those errors can be and are identified, quantified, and allowed for by adjustment of the data. What's more. such adjustments are usually very small and full details are published on the 'net. Many gauge stations have multiple gauges as at Malé, and an increasing number use CGPS to monitor stability and record the slightest movement. Most transmit their data at regular intervals by telephone line or satellite link, so any malfunction is soon spotted. All have a linear and uniform response, unlike temperature sensors whose non-linear response has to be allowed for by the recording circuitry, and they're not subject to sensor drift which may occur for some time before it's identified. There's no equivalent of UHI either, and so far no organisation has been able to get their grubby hands on the data and homogenise it or apply suspicious "adjustments".

Australia's CSIRO - manufacturing sea level hockeysticks?

The CSIRO's "State of the Climate" report was published today (14th March 2010). Missing a link to the report in the newspaper article summarising the dire consequences of climate change in store for Diggers Down Under, I did a Google News search, and came up with a rather interesting ABC page on the release of the report, titled "Sea Levels are Rising", which is not exactly news, but never mind that. Below the obligatory (and of course irrelevant) statement

It notes that the long-term warming trend has not changed, with each decade having been warmer than the previous decade since the 1950s.

... is an "interactive data visualisation" which "explores sea level rise patterns across Australia using data sets provided to the ABC by CSIRO". What struck me immediately was the uptick in sea level at three of the 62 tide gauge stations on the superimposed charts below the interactive map. I've selected Wyndham (WA) on the map - which shows the largest 2010 uptick. I've almost completed an interactive map for Australia myself (similar to the "South Pacific Seal Level 2011" in the reference pages, top-right sidebar), so I'm familiar with the chart for Wyndham. That uptick is definitely not kosher, not kosher at all. here it is:

Source: ABC

... and the RHS embiggened:

Source: ABC

By reading the data points from the ABC chart, I've established that all the preceding annual data points are correct, but 2010 stands out like a sore thumb Here's the corresponding chart for Wyndham, over the same time period, 1993-2010, with the last two points from the ABC chart added in red:

Data Source: NTC

The lower pair of points are Darwin (346) and Carnarvon (322). If the offsets from the real data for all three stations are added, they total 526, which averaged over all 62 stations comes to 526/62 = 8 mm rounded down. If that's deducted from the all-station average for 2010 (dark green line), it brings it down from 120 mm to 112, which means that the three outliers have increased that 112 figure by almost 11%.

I considered whether this might not be deliberate fiddling (generous and understanding to a fault!), but possibly a result of using incomplete data for 2010, for example January-June, but that doesn't explain it - any average with January as a start point is much lower than the ABC/CSIRO figures. Not only that, but 2009/10 were El Niño years, and there's usually a drop in sea level during one of these events; it shows up (though much less pronounced than 1997/8) at most stations except some in the far south. One such rogue plot might be a genuine error, but three? Here are my plots for Darwin and Carnarvon:

Data Source: NTC

Data Source: NTC

I've checked a random sample of the other stations on that rather clever (I'm green with envy - don't take that as weakness) interactive page, and can detect no further excursions from NTC data whatsoever, even for Hobart, home of the CSIRO. Currently, this is a mystery within an enigma, to paraphrase Churchill's description of Russia at the beginning of WW2.

UPDATE 15th March 2012

I've tracked down the source of two of the "adjustments". The CSIRO (funded by the Australian taxpayer) doesn't obtain its sea level data from the National Tidal Centre (funded by the Australian taxpayer, and where you'd expect them to get it as I do), but from an independent archiver, the Permanent Service for Mean Sea Level, based in Liverpool, UK. It's one which I use regularly for data unobtainable elsewhere, but I'm surprised that the CSIRO uses it. The PSMSL relies on data submission from national or other bodies which collect it, and it's rarely up-to-date, often months or even years of the latest data are missing. This explains why CSIRO web pages and publications quote sea level statistics up to Dec. 2010 for the NTC SEAFRAME stations, even though the NTC publish the data monthly.

Anyway, PSMSL has, apparently arbitrarily, adjusted 2010 monthly data for Carnarvon and Wyndham upwards by 140 mm and 260 mm respectively, with no acknowledgement (as is customary) on the station pages. here's a chart for Wyndham for 2009-10 illustrating what's been done:

Data Sources: PSMSL & NTC

That 260 mm is larger by 63 mm than the change in sea level at Wyndham since the gauge was installed in 1966, and larger than estimates of global sea level rise since 1900, so it's no small matter. What are they playing at? The Darwin data (to 2010) at PSMSL appears to be identical to the NTC data (except measured from a different baseline, which is irrelevant), so where did that 128 mm adjustment come from? Sherlock is on the case, with his (t)rusty magnifying glass at the ready. No violin though. On second thoughts, perhaps I'll just ask them. Watch this space for exciting developments.

Australia's CSIRO - The Truth, (but not) The Whole Truth, and nothing but Cherry Picking

Bruce Mapstone, chief, CSIRO Marine and Atmospheric Research, Hobart, Tasmania (note Hobart, there's more) wrote to The Australian "Sea-level observations". February 9th, 2012

BOB Carter's references to sea level ("Scientific research drowning in a sea of alarmism", 7/2) require clarification.

Sea level observations come from measurement by satellites and, as Carter notes, a network of tide gauges around the Australian coastline operated by the Bureau of Meteorology and similar gauges internationally.

These observations show that rates of sea-level rise around Australia range from 2.6mm a year to 9.5mm a year since the early 1990s, compared with the global average rate over a similar period of about 3.2mm a year.

Variation in sea levels from region to region is to be expected, as Carter also notes, but continuing analysis indicates the dominant trend around Australia for the past 20 years is upward and above the global average.

The global average of observations is near the high end of the Inter-governmental Panel on Climate Change model projections over the same period, indicating that the range of model projections encompass changes that have occurred, and indeed that the models may be underestimating observed changes.

Lets' jump right in and have a look at the Hobart tide gauge data - right on Bruce's doorstep at CSIRO Mansions on Castray Esplanade, overlooking the harbour, and probably within sight of the gauge -


View Larger Map

(click charts to enlarge)

Hobart, Tasmania - Sea Level 1987-2010

Oh dear, things are not looking too good for Bruce's "clarification", the rate from 1987 is -0.438 mm/year (0.67 from 1962). The odd 1987 start date is due to the fact that there are large gaps and anomalous readings prior to then. The gauge was displaced several times by ship's masters who couldn't park neatly, and had to be replaced (the gauge, not the masters, but who knows?).

He was referring to an opinion piece by Bob Carter two days previously, in which he attacked a "a science-fiction prediction of a 91cm rise in sea level in the district by 2100". Carter ridiculed Lake Macquarie City Council's new coastal planning regulations, which control new developments around the lake, mandating higher floor levels to "account for sea level rise and wave action". Yet the NSW Department of Planning has said there's "uncertainty surrounding the manner in which sea level will rise", which means whatever floor levels they enforce are likely to be either inadequate or excessive. The "91" figure (notice it's not a round 90 or "almost a metre") implies an estimate accuracy of +/- 1cm. The range is given as LOW: 18 cm. MEDIUM: 55 cm. HIGH: 91 cm. So the overall spread is 1:5, yet the figures are quoted to a single cm accuracy? Over a period of 90 years? That really is "science fiction". I could stick my neck out and say 50 +/-50 centimetres, which should cover most eventualities. Of course, I could be wrong. In fact I know I'll be wrong, which is why my neck will stay in.

The regulations are based on that 91cm maximum predicted rise by 2100, and Carter says

This prediction comes from the NSW government, which in turn sourced it from a UN political body, the Intergovernmental Panel on Climate Change. As with its unnecessarily alarmist projections of global warming, the IPCC's estimate of future sea-level rise derives from speculative computer models.

He also lists what he terms "four facts"

First, that rates of sea-level change vary around the Australian coast. This means any new coastal planning regulations (if and where they are needed) should be based on the appropriate local sea-level measurements rather than a hypothetical global average.

Second, the longest east coast tide-gauge record, from Fort Denison (Sydney), records an average rate of rise over the past 100 years of about 1mm a year (10cm a century).

Third, that other tide gauges, scattered around Australia as part of the national tidal network, mostly record rates of long-term rise between about 0.5mm and 2.5mm a year with no change in behaviour in the late 20th century that might reflect a human (global warming) influence.

And, fourth, that the Sydney tide gauge, as well as other long tidal records from nearby (Fremantle, Auckland) and overseas, exhibits a slowing rate of sea-level rise over the past 40 years.

How do Carter's "facts" stack up? His first is common sense. A health service bases its policies on local , not world, nor even regional demographic statistics and forecasts. It's logical to expect local and state authorities to do the equivalent as regards local planning.

His second is indisputable (except by Bruce Mapstone and others, who just ignore it. "Post-normal" science doesn't need facts).

His third is correct if you accept his qualification of mostly, and I accept it, having analysed data from most of the 95 National Tidal Centre tide gauges (plus two privately owned) around Australia. How does it square with what Bruce Mapstone says? First note that Brucie (I feel I know him well already) doesn't actually refute what Carter wrote. In short, he can't - it's fact. Note also what Mapstone actually said - "from 2.6mm a year to 9.5mm a year since the early 1990s". "Early 1990s" is a bit vague, isn't it? Well no, it isn't, because the CSIRO have chosen to use a subset of 14 of those 97 gauges, installed and monitored by the ABSLMP project at the National Tidal Centre of the Bureau of Meteorology. In other words, they've chosen to use just one seventh of the data available to them. Cherry anyone?

The fourth is indeed fact. Data from most of the long-term records show a slowing rather than an acceleration, and most sites show a flat or declining trend since 1999. I've posted on the CSIRO's data selection bias and more recent, lower rates here.

Back to Bruce's last paragraph

The global average of observations is near the high end of the Inter-governmental Panel on Climate Change model projections over the same period, indicating that the range of model projections encompass changes that have occurred, and indeed that the models may be underestimating observed changes.

Beat that for gobbledygook - if the model "projections" didn't "encompass changes that have occurred" then they can't be much use. If the models may be underestimating "observed changes" it doesn't automatically follow that they are also underestimating future "projections". I don't know, and he doesn't know what assumptions are made in those models.

Then there's a (partial but flawed) rebuttal of Mapstone's "non-rebuttal" the following day. "Levels all at sea" from one William Kininmonth:

THE clarification on Australian sea-level observations by Bruce Mapstone (Letters, 9/2) fails to reveal all of the uncertainties relating to the data.

Australian tide gauges accurately measure the relative movements of sea level with respect to the gauges. However, the vertical movement of individual tide gauges is uncertain.

The vertical movement isn't uncertain, at least not for some of the gauges. A global network of Continuous Global Positioning System (CGPS) stations is already established, designed to monitor land movement and/or gauge movement. There are 12 around Australia, near the ABSLMP stations. Here's the current plot for Sydney - appropriately the station's in Macquarie Park!

Latest ULR GPS position time series for Sydney

A note for those who've heard of the CGPS stations used by the NTC - the stations are remote from the gauges (up to several km inland), they monitor land movement, and do not "automatically adjust" the vertical position of the gauge. That's both impractical (the gauges are of necessity firmly fixed to their supports), and undesirable. It's not only the sea that rises and falls with the pull of the sun and moon, but the Earth's crust too (crustal tide), up to about 10cm range, depending on latitude, greatest near the point nearest to the sun and/or moon. If the gauge position was kept constant relative to the GPS satellites, the gauge would be dropping relative to the land as the tide and land rose, and would register a higher reading than it should, and a lower reading as tide and land fell.

And for the last word, from a body which even Bruce might take note of, The Tasmanian Planning Commission, in its sober but a trifle "consensus science" "State of the Environment (SoE) Tasmania 2009" report is precise and factual (my bold):

Sea-level rise for Australia is only slightly less than the global average (NTC 2007; Church and White 2006) and coastal observations confirm that sea-levels have been rising since at least 1920 by approximately 1.2 mm/yr (Church et al. 2006; IPCC 2001; see also Pittock and Wratt 2001; Lambeck 2002). A common feature in many model projections is a higher than the global average sea-level rise off the southeast coast of Australia and in a band stretching across the Indian and southern Pacific oceans at about 30°–45°S (CSIRO 2009). Sea-level rise is greatest (about 3 mm/yr) in the eastern equatorial Pacific and western equatorial Indian Ocean. Observed rates of rise are smallest (about 1 mm/yr) in the western equatorial Pacific and eastern Indian Ocean, particularly the northwest coast of Australia. Two of the longest continuous Australian tide gauge records are from Fremantle in Western Australia (92 years) and Fort Denison in New South Wales (83 years) indicate that the observed rate of sea-level rise relative to the land has been 1.38 mm/yr and 0.86 mm/yr respectively (DPIWE 2004). Regional variations in the rate of sea-level rise are weaker for much of the rest of the global oceans (AGO 2003). 

They haven't got the last word, though, I have. If the CSIRO chooses to use the state-of-the-art NTC SEAFRAME gauge data for (some of) Australia, why didn't they use data from similar NTC gauges across the Pacific islands for their "Climate Change in the Pacific" reports instead of spatially coarse satellite data? Why did they choose to use (coarse) gridded HADSST2 sea surface temperature data rather than localised in-situ measurements recorded by those same gauges? You answer the question, I already know the answer.