Download Sea Level Variation in the Eastern Asia

Journal of Oceanography
Vol. 50, pp. 643 to 651. 1994
Sea Level Variation in the Eastern Asia
TETSUO YANAGI and TATSUYA AKAKI
Department of Civil and Ocean Engineering, Ehime University, Matsuyama 790, Japan
(Received 25 November 1993; in revised form 7 April 1994; accepted 22 June 1994)
Mean sea level variations in the eastern Asia region during 1950 to 1991 are
investigated with the use of observed sea level data at 16 stations. It is suggested
from the data analysis, that the main cause of long-term sea level variation in this
region may be the plate tectonic processes. The mean sea levels along the eastern
coasts of Japan and the Philippines, and that along the southern coast of Indonesia
have risen due to the subsidence of Pacific, Philippine and Australian plates under
the Eurasian plate, respectively. On the other hand, the mean sea levels along the
western coasts of Japan and the Philippines, and that along the northern coast of
Indonesia have fallen. The distribution map of mean sea level rise at the year 2030
from 1985 in this region is presented on the basis of the results of this work and
IPCC (1990).
1. Introduction
The global mean sea level has been rising in these 100 years and such mean sea level rise
is largely due to the thermal expansion of oceans, and the melting of mountain glaciers and the
Greenland ice sheet (Gornitz et al., 1982; Barnett, 1984). IPCC (Intergovernmental Panel of
Climate Change, 1990) summarized the global mean sea level rise rates in these 100 years and
obtained the best estimate value of 1.05 mm/year. Moreover, IPCC (1990) reported that the most
suitable predicted mean sea level rise will be 18 cm (8–29 cm) at the year 2030 and 44 cm (21–
71 cm) at the year 2070 from 1985 and they will be mainly due to the thermal expansion of oceans,
and the increased melting of mountain glaciers and small ice caps. If we would have some
countermeasures to the emissions of the major greenhouse gases, the mean sea level rise would
be one-third lower than those without any countermeasures (IPCC, 1990).
Because the sea level is observed in relation to the fixed bench mark on land, “the relative
sea level” may be changed not only due to the change of ocean level but also due to the vertical
land movements. Then, the mean sea levels have not changed in the same manner in the world,
e.g. the mean sea levels have fallen in parts of Scandinavia by as much as 10 cm/year due to
isostatic rebound of land following the last major glaciation.
Therefore, it is expected that the future mean sea level rises will not appear in the same
manner in the world. We reveal in this paper the past main reason of mean sea level change in
the eastern Asia at first and present the map of future mean sea level rise distribution there.
2. Data Analysis
The sea level data at 16 stations in the eastern Asia (shown by full circles in Fig. 1) from 1951
to 1991 are obtained from the Permanent Service for Mean Sea Level (PSMSL) at Proudman
Oceanographic Laboratory, U.K. Such observation stations are selected because the yearly mean
sea level data during these 41 years are obtained there. The partial lack of data are linearly
interpolated with use of data before and after the lacking. There are many other observation
644
T. Yanagi and T. Akaki
Fig. 1. Observation stations of sea level (closed circle) and air pressure (open circle) variations.
stations around Japan, which satisfy such condition, but we only select representative four
stations around Japan because the detailed characteristics of sea level variations there has been
already investigated by many authors (e.g. Yanagi and Akaki, 1993). There is no sea level data
around Indonesia in PSMSL. The yearly mean sea level data and least-square regression linear
trend are shown in Fig. 2. The mean sea levels have risen at some stations and they have fallen
at some stations.
The rapid increase of mean sea levels at Manila and at Fort Phrachula Chomklao (Bangkok)
from 1965 to 1982 were caused by the large withdrawal of groundwater (Emery and Aubrey,
1991), and the rapid sea level rise has stopped after 1982 at both stations due to some regulation
for the withdrawal of groundwater. The data at these two stations are out of consideration from
the following analysis because we are concerned to the mean sea level variation in rather wide
area due to the natural processes without the anthropogenic sea level variation at local station in
this paper.
Observed sea level variations obtained here are not adjusted by the air pressure variations
though the air pressure variation of +1 hPa induces the sea level variation of –1 cm. The yearly
mean air pressure variations at three representative stations in this region, Fukuoka, Manila and
Songkura (shown by open circles in Fig. 1) are shown in Fig. 3. The analysis is limited from 1960
to 1990 because the air pressure data at Manila and Songkura from 1951 till 1959 are not obtained
Sea Level Variation in the Eastern Asia
645
Fig. 2. Variations of yearly mean sea level and least-squared regression line. Numbers mean the mean sea
level rise rate and its standard deviation in mm/year.
646
T. Yanagi and T. Akaki
Fig. 3. Variations of yearly mean air pressure and least-squared regression line. Numbers mean the mean
air pressure rise rate and its standard deviation in hPa/year.
in Japan Meteorological Agency. The least-square regression linear trend are +0.00073 hPa/year
at Fukuoka, –0.032 hPa/year at Manila, and –0.014 hPa/year at Songkura. Such air pressure trend
may induce the mean sea level change at –0.073 mm/year to 0.32 mm/year and such mean sea
level change rates are much smaller than the past global mean sea level change rate of 1.05
mm/year. Therefore, we may ignore the effect of air pressure variation to the mean sea level
variation in this paper.
The estimated rates of mean sea level variation in these 41 years obtained by the least-square
regression method are summarized in Fig. 4. The mean sea levels have risen along the eastern
coasts of Japan and the Philippines but they have fallen along the western coasts of Japan and the
Philippines, the eastern coast of China and the southern coast of Indochina Peninsula. The rates
of mean sea level rise along the eastern coasts of Japan and the Philippines of 3.8–5.9 mm/year
are much larger than that of global mean sea level rise of 1.05 mm/year.
Along the coasts of Indonesia, we cannot have sea level data now. Proudman (1940)
analyzed the mean sea level variation there from 1925 to 1931. His results are shown in Fig. 5
though the reliability is low due to analytical duration which is too short. The mean sea level rose
along the southern coast of Indonesia as much as 11.9–13.8 mm/year but it fell along the northern
coast as much as 3.9–6.9 mm/year.
Sea Level Variation in the Eastern Asia
647
Fig. 4. Obtained mean sea level rise (+) or fall (–) rates in mm/year from 1951 to 1991 in the eastern Asia.
Fig. 5. Land subsidence rate in mm/year from 1925 to 1931 obtained by the available tide-gage data along
the coasts of Indonesia as reported by Proudman (1940) (From Emery and Aubrey, 1991).
648
T. Yanagi and T. Akaki
3. Discussion
Gornitz et al. (1982) and Barnett (1984) showed that the mean sea level variation in the
eastern Asia is not obvious in these 50 or 100 years. However, they obtained their results from
an averaged sea level data at some stations in this region. When we average sea level variations
in this region, it means that the rising sea level trend along the eastern coasts of Japan and the
Philippines, and the southern coast of Indonesia are canceled by the falling one along the western
coasts of Japan and the Philippines, and the northern coast of Indonesia. Therefore, we must not
average the sea level data in this region in order to reveal the horizontal distribution of mean sea
level variation there.
The main causes of mean sea level change are thought to be 1) the change in volume of ocean
water, 2) the change of the land level and 3) the change of wind and current, except of the local
sea level change due to anthropogenic effects such as large withdrawal of groundwater. The
change of mean sea level due to 3) is considered to have short-term periods of several or decadal
years and we can neglect here because we estimated the 41 years linear trend here. The long-term
mean sea level trend shown in Figs. 4 and 5 may be due to 1) or 2). The cause 1) consists of the
Fig. 6. Activities of plates in the eastern Asia (from Gross, 1987).
Sea Level Variation in the Eastern Asia
649
water expansion due to water temperature rise and the melting of land ice. The mean sea level
rise due to 1) has occurred in the same manner in the world and the past global mean sea level
rise rate of 1.05 mm/year is thought to be due to 1) (IPCC, 1990). On the other hand, the mean
sea level rise due to 2) has regional characteristics, that is, the mean sea level rises in the region
where the plate subducts under the other plate, and it falls in the region where the isostatic land
rebound occurs. The activity of plates is very high in the eastern Asia as shown in Fig. 6 (Gross,
1987). The Pacific, Philippine and Australian plates subduct under the Eurasian plate and the line
of plate subducting well coincides with the boundaries of mean sea level rising and falling shown
in Figs. 4 and 5. Such facts suggest that the main cause of mean sea level variation in the eastern
Asia is 1) and 2) and the trend of mean sea level variation due to 2) may be obtained by subtracting
the global mean sea level trend of 1.05 mm/year in these 100 years from the observed mean sea
level trend shown in Figs. 4 and 5. Therefore, the contour lines of 2 mm/year and 0 mm/year in
Fig. 4 mean that the land level has fallen at 0.95 mm/year and has risen at 1.05 mm/year in these
41 years, respectively.
For comparison, the mean sea level rise rates at eight stations on the Pacific and Australian
plates are shown in Fig. A-1. They are between –0.1 and +1.6 mm/year and close to +1.05
mm/year, which is the global mean sea level rise rate. Such fact also suggests that the mean sea
level rise rate on the plate are mainly regulated only by 1) and that along the rim of plates by 1)
and 2). Yanagi and Akaki (1993) obtained similar result, that is, the mean sea level variation
Fig. 7. Predicted sea level rise at 2030 from the level at 1985 in cm in the eastern Asia.
650
T. Yanagi and T. Akaki
around Japan is mainly caused by 1) and 2) with use of many mean sea level data as much as 30
stations.
The horizontal distribution of future mean sea level rise at the year 2030 from 1985 in this
region may be obtained by the summation of the future global sea level rise of 18 cm at 2030
(IPCC, 1990) and the sea level rise due to 2) which is estimated from the rates of land level rise
or fall in these 41 years which are obtained by the procedures mentioned above. The result is
shown in Fig. 7. The mean sea level will rise more 22 cm at the year 2030 from 1985 along the
eastern coasts of Japan and the Philippines but the mean sea level rise will be below 13 cm along
the coasts of the Yellow/East China Seas, South China Sea, and the Gulf of Thailand.
Such figure will be useful when we plan to take some countermeasures for the mean sea level
rise such as increasing the height of dike, protecting the beach and so on.
Acknowledgements
The authors express their sincere thanks to Dr. M. Odamaki who guided us to the data base
of PSMSL. This study is partly defrayed from the research fund by the Ministry of Education,
Science and Culture, Japan.
Appendix
The mean sea level variations at eight stations on the Pacific and Australian plates shown
in Fig. A-1 are also investigated with the use of the yearly mean sea level data during 1948 to
Fig. A-1. Observation stations of mean sea level variation during 1948 to 1980 and mean sea level rise
rate in mm/year.
Sea Level Variation in the Eastern Asia
651
Fig. A-2. Mean sea level variations during 1948 to 1980 at eight stations in the western Pacific. Numbers
mean the mean sea level rise rate and its standard deviation in mm/year.
1980, which are obtained from the PSMSL. The analysis duration is a little different from that
in the eastern Asia because a good quality long-term data is only obtained there during this period.
The yearly mean sea level data, the least-square regression lines, estimated sea level variation
rates and its standard deviations are shown in Fig. A-2. The mean sea level rise rates are
summarized in Fig. A-1.
References
Barnett, T. P. (1984): The estimation of global sea level change: a problem of uniqueness. J. Geophys. Res., 89, C5,
7980–7988.
Emery, K. O. and D. G. Aubrey (1991): Sea Levels, Land Levels, and Tide Gauges. Springer-Verlag, New York, 237
pp.
Gornitzs, V., L. Lebedeff and J. Hansen (1982): Global sea level trend in the past century. Science, 215, 1611–1614.
Gross, M. G. (1987): Oceanography. Fourth Edition. Prentice-Hall, London, 406 pp.
IPCC (1990): Climate Change. Cambridge University Press, Cambridge, 365 pp.
Proudman, J. (1940): Monthly and annual mean heights of sea level, up to and including the year 1936. Association
d’Ocianographie Physique, Union Geodesique et Geophisique Internationale, Vol. 5, 255 pp.
Yanagi, T. and T. Akaki (1993): Variation of the mean sea level around Japan. Umi no Kenkyu, 2, 423–430 (in Japanese).