Sunday, May 27, 2012

Pumping sea-level up

Global sea-level has risen at a pace of about 1.8 mm/year in the period 1960 to 2003. Several factors are contributing to this rise: ocean thermal expansion, glacier melting, and ground water depletion. Which factor has been the strongest contributor ?

According to a recent paper, the strongest contribution stems from pumping out ground water for irrigation, which subsequently either evaporates and precipitates over the ocean or each channelled to the oceans by rivers. The other side of the coin is construction of artificial reservoirs, which capture water that otherwise would have ended up in the oceans as precipitation or river flow. Thus groundwater depletion causes sea-level rise, whereas water storage in reservoirs depletes sea-level. These two anthropogenic contributions, in particular ground water depletion (GWD), are found to be one of the largest factors. GWD has attained a value of 1.05 mm/year, compared to about 0.5 mm/year caused by glacier and ice-cap melting in this period.

Estimates of various contributions to the observed SLC. The estimates of the IPCC AR4 and that of this study are for the period 1961–2003, whereas the estimates of the other studies are for the later half of the twentieth century, except for Wada et al.19 (1960–2000). Error bar indicates the uncertainty range in the residual of the sea-level budget estimated by the AR4. From Pokhrel et al. Nature Geosciences 

This most recent study provides estimations of GWD that are larger than previous studies so far. But it has to be borne in mind that these estimations are the results of model simulations and not of direct observations. Previous studies were also based in some way or another based on model estimations, but Pokhrel et al claim that their integrated model is superior to previous estimations. As usual, I would interpret that these estimations may be better than previous ones, but I would put more confidence on these figures once they have been confirmed by other groups in the next years. This paper will be likely discussed in the next IPCC report, but any follow-up studies, confirming or rebutting these numbers, will not, as the dead line to submit articles to be considered by the IPCC is due in 2 months.

The high contribution to sea-level rise by GWD indicated in this study doe not imply that the contribution of ocean expansion or glacier melting have to be readjusted. This is because the 'sea-level budget' is still not really closed for the period up to year 2000, and there is still a margin of uncertainty that is not accounted for by any of the known potential contributors. This is not the case for the last decade, for which the nearly global and more accurate measurements from satellite altimetry, satellite gravimetry and the Argo monitoring system, do allow to disentangle the individual contributions to the observed sea-level rise.

The new estimations of GWD may have, however, consequences for the so called 'semi-empirical' estimations of future sea-level rise. These methods are based on statistical relationships between global mean temperature and the rate of sea-level rise. If a larger part than previously thought of the observed sea-level rise is not climate-related, these statistical estimation may be then biased high.


intrepid_wanders said...

While I can not fathom how this research can even begin (in the concept of adjustments for iso-static rebound but not considering the the fissure genesis with both subduction and induction zones and many other plate tectonic movements).

Anyhow, I very much enjoy papers that are published on the Argo system by Josh Willis.

Here are some conversations with Pielke Sr.:

Even though he has been tacked on with J. Hansen papers of the past, he is very balanced and I concider him an honest broker.

Here is a PNAS from Josh on GWD:

eduardo said...


The esimation of Gorund Water Depletation are based on a model of water requirements:


Unsustainable groundwater use is estimated based on the total water demand (domestic, industrial and agricultural) and the availability of water from near-surface sources. This approach proposed by previous studies29,30 estimates
the unsustainable groundwater use implicitly, as the model does not explicitly account for groundwater dynamics. Water is withdrawn unlimitedly when needed because there are no global data sets on the availability of groundwater
sources. Nevertheless, it has been shown that the model-estimated global and country-scale groundwater depletion is within the range of the reported statistics for circa 2000 (ref. 17).


It is thus not based on observations of relative sea-level change i.e. with respect to the a reference anchored in earth crust, like a tide gauge. I cannot tell if this model is realistic, as I have no expertise in this field

Alex Harvey said...

Dear Eduardo,

I note you have written,

The high contribution to sea-level rise by GWD indicated in this study does not imply that the contribution of ocean expansion or glacier melting have to be readjusted.

The paper is behind a pay-wall and I have some difficulty interpreting the figure. It shows in black the 'Net SLC owing to TWS change'. I assume 'TWS' is total water storage. There is no 'Net SLC owing to TWS change' given for either the AR4 or 'other studies', however, leading in my mind to an apples-with-oranges comparison. Is this detail buried in the paper somewhere? Exactly how much of a departure from previous studies is this new one?

Paul S said...


TWC stands for 'Terrestrial Water Storage'.

AR4 simply repeats the TAR finding for this value, in terms of bottom-up estimates: 'which estimated the rather wide ranges of –1.1 to +0.4 mm yr–1 for 1910 to 1990 and –1.9 to +1.0 mm yr–1 for 1990'

Pay attention to the colours. The other studies show individual estimates for either groundwater depletion or reservoir storage. These can be compared with Pokhrel et al.'s estimates for the respective individual contributions. The values from AR4 are included presumably to compare the residual (i.e. unexplained) sea level rise with their own Net SLC from TWS change.

ghost said...

Stefan Rahmstorf hat auch einen Artikel darüber:

Sein Fazit:
Insgesamt würden die neuen Zahlen also die prozessbasierten Projektionen erhöhen und die semi-empirischen absenken, was die beiden Methoden näher zusammen brächte und insofern aus meiner Sicht erfreulich wäre. Allein mir fehlt der Glaube.

Eigentlich hat Prof. Rahmstorf schon immer geschrieben, dass prozess- und statistikorientierte Methoden sich angleichen werden.

eduardo said...

man sollte nicht unbedingt alle Kommentare, die in blogs geschrieben werden, kritiklos akzeptieren, auch die von der Klimazwiebel nicht.

Wenn es sich in den nächsten Drafts des IPCCs Berichts nicht viel ändert, werden die IPCC-Schätzungen zu globalem Meeresspiegel weit weg von denen der semi-empirischen Methoden liegen.

Alex Harvey said...

Paul S., thanks very much; that's very helpful. Well, I guess the issue's important given the tone of Prof. Rahmstorf's RealClimate post.

ghost said...

Es ist eben Prof. Rahmstorfs Meinung und er verteidigt seine Arbeit. Hm, im Moment sieht es nicht so aus, als ob die statistischen (was bedeutet semi-empirisch eigentlich?) Modelle richtig sind. Zum Glück. Ich finde prozessorientierte Modelle + Beobachtungen eigentlich sehr überzeugend.

Aber doch ist es eben eine Meinung, keine Beleidigung, keine Unterstellung, keine Lüge, kein Vergleich von Grünen mit Nazis oder so. Und deswegen kann man es Beitrag einfach zitieren, ohne eigene Meinung. Ich hoffe, sie merken den Unterschied.

North Carolina macht es aber einfacher für die wissenschaftliche Diskussion:

Da wird einfach per Gesetz bestimmt, was man annehmen soll:

These rates shall only be determined using historical data, and these data shall be limited to the time period following the year 1900. Rates of sea-level rise may be extrapolated linearly to estimate future rates of rise but shall not include scenarios of accelerated rates of sea-level rise.


eduardo said...


natürlich darf Rahmstorf seine Meinung darüber schreiben und auch seine Methoden verteidigen, gar keine Frage. Ich wollte aber zum Ausdruck bringen, dass diese Methoden nicht den 'Konsensus' darstellen.

In der Vergangenheit wurde eben der wissenschaftlichen Konsensus dafür missbraucht, die wissenschatliche Diskussion abzuwürgen.

Es ist nun interessant zu sehen, wie plötzlich der Konsensus seine Autorität verliert- wenn es nicht passt