Acceleration Of Global Sea Level In The th Century

It was mentioned in Chapter 2 that sea level began to rise at the modern rate approximately at the middle of the 19th century. In addition to the geological, geomorphic, and archeological evidence cited, there are supporting data from a few very long water level records. Woodworth (1999) has reconstructed some records reaching back to the early 18th century, which show a distinct increase in the rate of sea level rise in the 19th century. Figure 3.15 presents his data for Amsterdam sea levels. Similar results are seen at Brest, Liverpool, Sheerness, and Stockholm. Data taken prior to the late 19th century are from tide staff or other visual readings and have a significantly larger scatter of values than later data taken with a recording tide gauge. But the old data are adequate to show that there was a change in the rate of sea level rise near the middle of the 19th century. In Fig. 3.15 the data set was divided into two groups arbitrarily at 1850, but an alteration of a decade one way or the other would not change the results very much. The rate of relative sea level rise from 1850 onward is about 1.5 mm/ yr, about 5 times greater than prior to that time.

Has there been any acceleration of global sea level in the 20th century? Fortunately, global acceleration is much easier to address than global trend. This is so because GIA, and vertical crustal movements arising from plate tectonics at locations with long earthquake recurrence times, are both temporally linear over the tide gauge era. A polynomial regression of the 2nd degree in time can absorb the total linear component of the signal into the linear term whatever its source, and separately accommodate an acceleration in the quadratic term, if the record is long enough. Woodworth (1990), Gornitz and Solow (1991), and Douglas (1992) exploited this fact to examine the issue of an acceleration of sea level during the last 100-150 years. No author found conclusive evidence of a

Figure 3.15 Amsterdam mean annual relative sea levels.

global acceleration of sea level, especially compared to what is predicted to accompany future global warming. Douglas (1992) used the largest number of tide gauge records for his estimate of global sea level acceleration and found for the period of the latter half of the 19th century to 1980 the global acceleration value -0.01 ± 0.01 mm/yr2. He also concluded that 50 or more years would be needed to detect an increase in the rate of sea level rise (i.e., an acceleration) from ordinary tide gauge records. Figure 3.16 (after Douglas, 1992) illustrates the problem. It presents the coefficients in millimeters per year-squared of the acceleration terms derived by linear regression for PSMSL RLR sites with records >10 years, from 1880 to 1990. This figure is analogous to Fig. 3.11, which showed trends of sea level rise. However, in contrast to the trend case, it shows only a very small scatter of estimates for records longer than about 50-60 years. This demonstrates that vertical crustal movement rates are nearly constant at most sites; they are absorbed into the linear trends. The scatter of accelerations in Fig. 3.16 for the shorter records exists because interdecadal and longer variations of sea level, prevalent in sea level records, can be absorbed into the acceleration term for the shorter records. No reason is known for the precipitious drops in magnitude of the quadratic term at about 30 years and 60 years. In any case, Fig. 3.16 further underscores previous comments about the impact of interan-nual and longer variations of sea level, and the absolute necessity of using long records.

Concerning the issue of global warming and a possible increase in the rate of 21st-century sea level rise, Douglas (1992) has pointed out that predictions

■* i ■


»" V '

ii'S'' •••• ••

• •

Record Length, Years

Figure 3.16 Acceleration component of relative sea levels.

0 20 40 60 80 100 120

Record Length, Years

Figure 3.16 Acceleration component of relative sea levels.

(Warrick et al., 1996) imply an acceleration of sea level corresponding to a coefficient of (time)2 of order 0.1-0.2 mm/yr2. This is small compared to the scatter of accelerations shown in Fig. 3.16 for sea level records less than about 60 years inlength. However, Woodworth (1990) demonstrated by a numerical simulation that applying a meteorological correction for the long-period sea level variations at some sites may enable detection of a significant change in the rate of sea level rise in a few decades. Sturges and Hong in Chapter 7 of this book show explicitly how key U.S. east coast sea level records can be corrected to a very high degree for interdecadal variations of sea level. This gives some confidence that properly corrected tide gauge records do have the potential to detect a change of the underlying trend of regional sea level rise in a matter of decades.

0 0

Post a comment