2004-12-13

Sea level rise

A science post again. It was about time.

The standard septic lines-of-retreat are something like this:


  1. Global warming isn't happening
  2. OK, so its happening, but its not our fault
  3. OK, so its happening and its our fault but it won't continue
  4. OK, so its happening and its our fault and it will continue but its not a problem


Now whether it will be a problem is (to my mind at least) quite hard to say: certainly a harder problem than the first three points. My best guess is that rapid warming would be a problem because it will change things that we have got used to. Thats in terms of temperature, precipitation, crops. But one area where its pretty hard to say "its not a problem" is sea level rise, which is one reason people are interested in it.

If you're interested in future sea level changes, then a good place to start is... the future section of IPCC chapter on sea level and the table here. Note that (though numbers vary from different models) the biggest contributor is thermal expansion of the oceans as they warm. Next biggest is glaciers and small ice caps. Antarctica tends to contribute a reduction. And this is all on the assumption of no surprises.

However, I didn't want to write about the future but the present. Because the sea level rise over the past century is a slight puzzle - "The Enigma", according to Munk - because different estimates produce different results. Much of what I'm writing here is based on a recent Reviews of Geophysics paper by A Cazenave and R S Nerem VOL. 42, RG3001, doi:10.1029/2003RG000139, 2004 Present-day sea level change: Observations and causes (argh, I included a DOI: ah well) and the IPCC tar ch 12, section 11.4: Can 20th Century Sea Level Changes be Explained?.

So: there are essentially two ways to estimate sea level changes over the last century: firstly, you can total up (over the last century) thermal expansion, glacier melt, ice sheets, terrestrial storage and so on, and you get a value of 0.7 +/- 1.5 mm/y. Or, you can take the observations of sea level from tide gauges (having done your best to remove or avoid effects from the land movements) and you get 1.8 mm/y (+/- something like 0.5). And since the satellite era in 1992, you can do it from satellite altimetry (if you're clever enough) and you get 2.8 +/- 0.4 mm/y (plus maybe a bit from changes in basin shape).

Broadly speaking, to get the central values (0.7 and 1.8) to fit, you have to either push the tide gauge value down, or the individual-contributions value up. Pushing the tide gauge value down is unattractive though, because it then implies a huge acceleration over recent decades to reach the (probably fairly well determined value) from the satellites. Most promising possibilities are an underestimation of the thermal expansion component (which would possibly have consequences for the future projections, i.e. make them larger, but would make the ocean GCM people unhappy); or an underestimate of the extra water from ice melt and terrestrial storage changes.

See-also: SLR on wikipedia.

No comments: