“Perhaps you could point us to some citation that shows acceleration of rise in the 20th century” – with pleasure:

Church and White, 2006. A 20th century acceleration in global sea-level rise, GRL, Vol 33, L01602, doi:10.1029/2005GL024826.

(but I still think that a pre-20th century rise of 0.25 mm/year, a 20th century (averaged) rise of 1-2 mm/year, and a late-20th century rise of over 3 mm/year is a pretty good indication of an acceleration – it fits the above paper pretty well also.

Thanks for the Church and White reference, Jim, but of course I had it already. The problem is, I don’t take much stock in it.

Why not? Because the errors are greatly underestimated, there has been inadequate adjustment for autocorrelation, and the analysis results disagree with other datasets.

DISAGREEMENT WITH OTHER DATASETS

Let’s look first at the problems with estimating sea level from coastal gauges. Here’s what the TOPEX satellite radar says the change over the last ten years has been like …

Now you tell me, Jim … how accurately could you predict this rise from 10 sea level gauges on the coast in 1870, mostly in Europe and the US? Or even from the 170 current gauges? Most of the coast world-wide hasn’t seen much rise at all in ten years, and sea level’s dropping around San Francisco … but it’s rising where there are no gauges.

Church and White claim that from 10 tide gauges in 1870, we can tell the relative global sea level within ±44 mm (95% confidence) … that’s an inch and three quarters accuracy for the sea level a century an a quarter ago, from 10 tide gauges … hmmm. Given the map above, do you think that’s possible?

ERROR UNDERESTIMATION

To see if their errors were reasonable, I decided to compare their error estimates with the error estimates from another dataset for which we have 95% confidence intervals, the HadCRUT3 surface air temperature dataset. Here’s the comparison …

Now, air temperature is easy to measure. We have now, and have had historically, many more air temperature measuring sites than tidal stations. In addition, the sea level must be measured for at least 50 years before we can have any confidence in the results, because that is the length of the tidal cycle. Air temperature, on the other hand, has no such restriction. Sea levels must be measured at high tide and low tide to be of any use, and these times change constantly. Early records are often quite inaccurate, because the levels were not measured at the highest or lowest times. Again, air temperatures do not have this restriction. Automated air temperature measurements of daily highs and lows, using mercury thermometers that measure the extremes, have been around for half a century. Automated tide gauges are a recent development. Sea level measurements are affected by the local rise or fall of the ground level, where air temperatures are absolute. Finally, air temperature can be measured anywhere on the globe, while sea levels can only be measured on the coast.

Because of all of these factors, it is totally unreasonable to say that we can measure historical sea levels with greater accuracy than historical air temperature levels, and yet that is what they are saying.

AUTOCORRELATION

Here’s the real problem in all of this. C&W say that they have adjusting for autocorrelation, viz:

For the 20th century, the rise is about 160 mm and the linear least-squares trend is 1.7 ± 0.3 mm yr-1 (95% confidence limits). This error includes allowance for the serial correlation of the time series, (four years of data per degree of freedom), uncertainties in GIA corrections (0.09 mm yr 1 from the rms difference between GMSL trends calculated using three different GIA models [Church et al., 2004]) and uncertainties in the EOFs (0.1 mm yr 1, see below).

However, the problem runs much deeper than that. The lag(1) autocorrelation of their yearly data is so extreme (0.90) that:

1) We cannot say that there is a trend in the data (p>0.05, so we cannot reject the null hypothesis).

2) We cannot say whether the quadratic they propose as a fit to the data is any better than a straight line. Thus, we cannot claim any acceleration.

For all of these reasons, as I said above, I don’t put much stock in this study …

w.

Mr and Mrs welirocks (68-71): all you are doing by concentrating on San Francisco is illustrating the common strategy of the contrarians of cherrypicking – just go and look at the literature on GLOBAL sea-level rise. You’ll also find in that literature information on how sea-level rise varies AROUND THE GLOBE, why it varies and how results from one station are NOT measures of global sea-level.

Jim, we all understand that. The issue is the uncertainty surrounding the trends. Examples have been given of widely divergent trends from sites that are only miles apart. Taking a large number of sites from across the globe and averaging them doesn’t solve that problem.

Added to the measurement error is the problem of subsidence or otherwise as mentioned by the Wellirocks, which can apparently be important on quite small scales.

I think anyone that pretends that sea-levels in the 19th century can be reliably estimated or reconstructed is out of their mind, and in the 20th century, prior to satellite observations, I don’t think the situation is much better.

It was Steve Bloom who chose to focus on San Francisco, not welikerocks, and not I. If you want to blame someone for cherry-picking, blame him.

If not cherry-picking, what would you call selecting out responses for criticism, rather than the initiating post?

And yet your compere in arms, Steve B., in #56 employed two sites from an unpublished unrefereed internal California government report to suggest that sea levels have been rising all along the west coast of north america.

Is that not egregious cherry-picking, too? And why haven’t you protested it?

“A Coastal Hazards Data Base for the U.S. West Coast”
link is to one page (home and back buttons provided bottom of page)

http://tinyurl.com/qcd5n

It’s from the late 90′s, but go for it. Now it’s your turn. Show us the Hockey Stick. It should still be there.

Also we know perfectly well how sea level varies around the globe. Re-read.

you say:

“You are going to claim that the present global sea-level rise is simply a continuation of the recovery from the last glaciation, then it might be an idea to take a look at the literature on sea-level reconstructions over the last glacial cycle (and especially during the Holocene). And make sure you come up with some actual numbers.”

No we don’t. We don’t need actual numbers to comment on a well known geological conditions for San Fransisco and San Diego Bay areas and the Holocene. We cited sources. Look it up yourself in the library. And for that matter Gov.Arnold should too.

Do you understand the difference between tidal gauge measurements and sea level rise?
Do you understand how local geological condition plays a major role in this discussion?
Look up what subsidence means.
Go over Willis’ and Frank’s points again.
You are wasting everyone’s bandwith and time.

Also, if you are going to claim that the present global sea-level rise is simply a continuation of the recovery from the last glaciation, then it might be an idea to take a look at the literature on sea-level reconstructions over the last glacial cycle (and especially during the Holocene). And make sure you come up with some actual numbers and not just qualitative generalisations!

Just up the coast a short distance the sea level is actually dropping because of uplift. Remember Steve Bloom, sea level is a relative term that depends on where you are like Pat Frank is showing as well. That paper you cited is complete nonsense and not based on any real geology from what I have seen in these references, and without reading the thing yet.

Read:
Califonia’s changing landscapes: A Guide to the Geology of the State. By Gordon B Oakeshott. 1978 (This is the bible for California Geology)

Or try:
California Geology By Deborhah Harden, 1995

Or:

Geology of California By Robert M. Norris and Robert W. Webb 2nd addition 1990

Therein you will find a real understanding of San Fransisco Bay. All three will tell you the sea is not rising there, the bay is sinking.

The tidal gauge data needs to be correct for uplift and subsidence and the data has not been corrected. This data is worthless unless this is taken into account. Tidal gauges measure tide not techtonics hence the difference between the San Fransisco Tidal Gauge and the Crescent City Tidal Gauge.
BTW San Diego Bay is also under going tidal subsidence hence the apparent sea level rise. It’s all relative.

On a personal note 30yrs of surfing the local breaks in So Cal show no evidence of sea level rise.

In the last 100 yrs my understanding from my education is that sea level has actually only come up a few cm in California. All natural. From the beginning of the Holocene, sea level in California has risen several hundred feet. All natural.

The other colored chart on the page is sea level in mm/yr

Man, I hate to fix it when it bumps up on the side bar que over more inportant pin-point discussions. Sorry!