Skip to content

Sea Level Rise–A Look At Church & White 2009

May 18, 2012

By Paul Homewood




Further to my reconstruction of Bruce Douglas’s assessment of sea level changes during the 20thC, I have also now had a chance to analyse the study by John Church and Neil White “Sea Level Rise from the Late 19th to the Early 21st Century”.

As with the Douglas study, this work is also based on tidal gauges, but is based on more sites (about 200) and is taken up to 2009. Their results are shown on the above graph and conclude that between 1993 and 2009 sea level rose by 2.8mm/year (fairly much in line with satellite trends).

They do point out though that

However, the reconstruction indicates there was little net change in sea level from 1990 to 1993, most likely as a result of the volcanic eruption of Mount Pinatubo in 1991.

This can be clearly seen in Figure 1. Clearly a large volcanic eruption such as this can not be regarded as a normal event and this has certainly acted to skew the trend upwards, but by how much?

From 1992 to 1993 sea level fell by 4.5mm, so if the trend was taken from 1992, the trend would be about 2.4mm not 2.8mm, quite a sizeable difference.  (I should point out the authors used 1993 as the starting point as they were comparing their numbers with results from satellites which began that year ).

It is also true to say that the trend would be less if 2010 or 2011 had been used as the end point, as sea levels began to drop in those years, probably close to 2.1mm.


Church   White 2009_htm_2477bf9b

Figure 1


But, more importantly, what do the longer trends tell us? Figure 2 shows a steady rise since 1880. There have been times, such as the 1920’s and 1930’s, when things have slowed down and it would appear we have crept above trend in the last decade.


Church   White 2009_htm_m2073a1b6

Figure 2


However sea level rise was relatively low in the first couple of decades of the record, so let’s take a look at how the graph looks from 1920 onwards, as in Figure 3.


Church   White 2009_htm_mbc01fc8

Figure 3

In fact, there has been very little movement around the trend line, except in the last 3 years. As climatologists never cease to tell us, we should look at trends over 30 years or more, which the next two graphs show with 30 year running averages.


Church   White 2009_htm_77778943

Figure 4

Church   White 2009_htm_7f8ac800

Figure 5

Figure 4 indeed shows a long term increasing trend from about 1mm/year at the start to 2mm/year. However it is clear there is a step change around 1950 (in other words, as this is a 30 year running average, sea levels began rising much faster after 1920 ). (The green line is the mean). The effect of this step change can be graphically seen in Figure 5. Since 1950, the 30 year average has meandered around the trend line, which is essentially flat. It currently stands at 1.93mm/year, slightly above trend, but at a similar level to many other years in the past. In 1957, for instance, it stood at 2.02mm.

While the Church & White studies suggest a slightly faster rate of rise in the last decade, it is too soon to assume this rate will continue. As Bruce Douglas pointed out

It is well established that sea level trends obtained from tide gauge records shorter than about 50-60 years are corrupted by interdecadal sea level variation

  1. Brian H permalink
    May 19, 2012 2:34 am

    All within the broad bounds of historical natural variation. No floods in prospect.

  2. Hugh J. permalink
    July 24, 2012 6:52 pm

    “As climatologists never cease to tell us, we should look at trends over 30 years or more, which the next two graphs show with 30 year running averages.”

    I think you misunderstand this point. How many years of data you need to look at depends specifically on the data set you are using. It just so happens that for Global Mean Surface Temperature 30 years of data centered on the date you wish to establish a trend for is about right. This is not true for all GMST datasets, and it is especially not true for all climate related data. For some it may be longer, others shorter, but that quote is a gross oversimplification and either a misunderstanding of how to study trends or a misuse of this idea which has crept into public consciousness.

    For a better description that I have the skills to provide, see this:

    Needless to say, applying the “30 years” requirement to every dataset related to climate completely misunderstands the mathematics.

Comments are closed.

%d bloggers like this: