Monday, 14 November 2011

South Pacific Sea Level to September 2011

Note: A permanent and (to be) regularly updated page has been created (see top of sidebar) located here. I've updated this page with data to December 2011.

The island sea level charts are drawn from monthly data from the South Pacific Sea Level and Climate Monitoring Project. I've included a chart for a second Fiji station, and also one at the end for a New Zealand station, Jackson Bay (South Island). Neither is part of the project, but data is provided on the website.

The high resolution SEAFRAME (Sea Level Fine Resolution Acoustic Measuring Equipment) monitoring stations comprise modern integrated housings which measure and record sea level, barometric pressure, water temperature and air temperature. Most of the stations were installed in 1992 and 1993, though a few were later. It's important to note that the effect of local land movement is eliminated from sea level data:
The Continuous Global Positioning System (CGPS) network monitors vertical movement in the earth's crust, such as subsidence or tectonic shifts, at the SEAFRAME tide gauges and adjacent land. Sea level data can then be adjusted to compensate for the earth's movement to within a millimeter, enabling the absolute sea level to be determined. 
I have voiced criticism of Australia's Bureau of Meteorology on various topics in the past, but as far as the presentation of sea level data is concerned, I rate their National Tidal Centre as the best. The page I linked to above has a table of the 12 stations in the project (and the two others I mentioned) which links to PDF plots and data tables for sea level, barometric pressure, water temperature and air temperature. The data tables in turn link to online graphical plots and text files for easy import to spreadsheets. A map links directly to the data tables. I liked that Idea so much I've pinched it for use here.

What should be evident from the plots is that any generalisation of the situation over this wide area is invalid. Apart from an almost universal downward spike during the 1997/98 ENSO event, the history and trends differ widely. It should also be clear that claims of "25 mm/year" or "no rise" since the early 1990s are also invalid.

Most of the charts are dominated by a downward "spike" in 1997/98. The level drop was due to unusually high barometric pressure during the 1997/98 ENSO ("El Niño/La Niña-Southern Oscillation") event. The correlation is well illustrated for the Marshall Islands (Majuro Atoll), so I've placed this first. Kiribati and Tuvalu have been given much attention in the news media and on the 'net recently, so they're listed next.

Readers are welcome to reproduce any of the plots - all I ask is that attribution be given, preferably with a link to this post. I haven't used thumbnails - right-click on the image and select "save image as" or whatever your browser prompts.

Level data has been converted from metres to millimetres to overcome loss of precision in Excel's trend data. Gaps in the plots indicate gaps in the original data. Note that the trend slope is monthly - multiply by 12 to get the annual value (e.g. y = 0.2269x gives 2.7228 mm/year).

Select a location from the map to view a graph of the monthly sea level statistics for that location. Click on the bottom of the blue area for the NZ station, Click on your browser's back button to return to the map.

Lombrum, Manus Island, PNG 02° 02' S 147° 22' E Honiara, Solomon Islands 09° 26' S 159° 57' E Port Vila, Vanuatu 17° 45' S 168° 17' E Lautoka, Fiji 17° 36' S 177° 26' E Nuku'alofa, Tonga 21° 08' S 175° 10' W Rarotonga, Cook Islands 21° 12' S 159° 46' W Apia, Samoa 13° 49' S 171° 45' W Funafuti, Tuvalu 08° 23' S 179° 13' E Tarawa, Kiribati 01° 22' N 172° 56' E Nauru, Nauru 00° 32' S 166° 54' E Majuro, Marshall Islands 07° 06' N 171° 22' E Pohnpei, FSM 06° 59' N 158° 14' E Jackson Bay, New Zealand 43° 58' S 168° 37' E
Click on a red dot to jump to the relevant sea level graph
Source: Bureau of Meteorology     

Marshall Islands

Island: Majura   Location: Uliga

Note the 1997/98 ENSO "spike" and the correspondence with the abnormally high atmospheric pressure from late 1997 to late 1998 shown in the barometric pressure plot below:


Island: Tarawa   Location: Betio

The trend line is pulled down by the ENSO dip from the end of 1997 to end 1998. To give an better view of the trend from 1992 to present, I replotted the chart with that data excluded:
The resulting trend is effectively zero.
A downward trend is evident from end 2001 to present:


Atoll:Funafuti   Island: Fongafale

As with Kiribati, the deep ENSO dip in 1997/98 pulls the trend line down on the left; it's below the 1994/1997 average. A zero trend is evident from 1999 to present:

Papua New Guinea

Island: Manus   Location: Lombrum

Here's a clearer view of the 1999-2011 trend of 3.4 mm/year, slightly above the global average:

Solomon Islands

Island: Guadalcanal   Location: Honiara

Again the trend is pulled down on the left; the average level 1999 to present is around 750 mm, and the trend 2.1 mm/year.


Island: Efate   Location: Port Vila

The big 1997/98 downward spike evident in most of the other plots is absent; instead there's a relatively steady upward trend.


Island: Viti Levu   Location: Lautoka

Island: Viti Levu   Location: Suva


Island: Tongatapu   Location: Nuku'alofa

Although there's a steep trend from 1993, current levels are not dissimilar to those between 2000 and 2003.

Cook Islands

Island: Rarotonga   Location: Avatiu

Levels seem to have stabilised from 2006.


Island: Upolu   Location: Apia

The familiar ENSO dip pulls the trend down at the LHS, but there is a sharp upward "spike" from 2010.


Island: Nauru   Location: Aiwo

Overall trend is effectively flat - levels at present similar to those in 1993. However, the trend from 2002 is downward.

Federated States of Micronesia

Island: Pohnpei   Location: Dekehtik

An overall upward trend, though little change from 2007 to present.

New Zealand

Island: South Island   Location: Jackson Bay

There seems to be little overall change from mid-1998 (after the ENSO dip) to present. A plot from 1999 confirms that:


  1. I have always wondered why BoM only used mid pacific locations when sites like Norfolk Island, Brisbane, Townsville and Thursday Island would have substantially expanded the sample area. Has this curiosity ever been explained?

    Ian Mott

  2. They have, except it's a separate project - The Australian Baseline Sea Level Monitoring Project, which covers 15 stations around Oz. See my post

    for details & links. I'm planning a similar post to this one, both of which will become permanent pages and updated regularly.