1
Long term trend of chemical oxygen demand in Saroma-ko Lagoon, Japan; possible effects of
climatic warming
Akihiro Shiomoto
Tokyo University of Agriculture, 196 Yasaka, Abashiri 099-2493, Japan
E-mail: a3shiomo@bioindustry.nodai.ac.jp
A trend of shortening duration of ice coverage has been reported in many rivers, ponds
and lakes around the world due to climatic warming
1-4
. The shortening will have much
influence on organic production due to the greater supply of light intensity and hence on
water quality. Chemical Oxygen Demand (COD), a useful measure of water quality,
has been monitored over the past thirty years in Saroma-ko Lagoon, a boreal lake in
Japan characterized by wintertime ice coverage
5
. Here the data set of water quality in
the lagoon is analyzed to reveal long-term trends in COD and the influence of climatic
warming on such trends. There has been a significant increasing trend, caused
primarily by an abrupt increase in the early 1990s. However, the increase could not be
explained by the increased load of organic matter on the lagoon from the river basin and
organic production in the lagoon. On the other hand, periods of freezing of the entire
surface have tended to be shorter with inevitable thinning of ice over the past forty
years
6
, probably due to climatic warming. Ice and planktonic algae are exposed to the
low light intensity
in situ, which is not optimum for the algal production of organic
matter
7-9
. The shortening and thinning should thus bring an improvement in ambient
light conditions for algae, resulting in an increase of organic matter in winter. Most of
the organic matter possibly remains even after spring because of the low water
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
2
temperature. It is highly possible that climatic warming promotes deterioration in
water quality in ice-covered lakes.
Saroma-ko Lagoon, which is150 km
2
in area and 19.5 m in maximum depth, is
located on the northeast coast of Hokkaido (ca. 44°N, 144°E) and is connected to the
Sea of Okhotsk by two narrow channels. The lagoon is characterized by wintertime
ice coverage. Long-term monitoring of water quality has been carried out by the
Hokkaido Government since the early 1970s
5
. A trend of increasing COD has been
noted during the past ten years
10
, arousing fear of worsening water quality. It has been
suggested that the increase is due to an increase in the load of organic matter from the
river basin
10
; however, a detailed study has not yet been conducted. Hence, the data
set of COD during 1975-2006 was analyzed to reveal trends over the past thirty years
and factors causing such trends. As a result, there has been a significant increase in
COD over the past thirty years, though temporal and spatial variations were observed in
each year (P<0.01; Fig. 1a). For detailed analyses, the mean value in each year was
investigated. The mean is affected by extremely high and low values but a significant
linear relation was found between the mean and the median by using the data in this
study from 1975 to 2006 (r = 0.948, n=32, p<0.001), meaning that the mean values
represent the COD in each year.
Relatively high mean values of COD were observed in 1975 and 1976 and an abrupt
decrease occurred in 1977 (Fig. 1b). The second channel was opened in 1979 and the
interaction of water between the ocean and the lagoon was therefore not as active in the
two years. The lowest salinity and relatively low salinity were observed in 1975 and
1976, respectively (Fig. 1c), meaning a large inflow of river water. Remarkably high
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
3
SS in 1975 supports the large inflow. The high COD can be due to a large load of
organic matter into the lagoon due to the large inflow of river water.
The mean COD values shifted from 1977 to 1993 within a range of relatively low
values, and neither a significantly increasing nor decreasing trend was found in the
values in spite of year-to-year variation (P>0.05; Fig. 1b). An abrupt increase in mean
COD was observed in 1994 and the mean COD values varied from 1994 to 2006 in a
range of high values. Neither an increasing nor decreasing trend was significant
during the twelve years (P>0.05), though an increasing trend was observed in the
running mean (Fig. 1b). Furthermore, there was a significant difference between the
COD values in 1977-1993 and in 1994-2006 (U test, P<0.01) with a mean value in
1994-2006 (2.5 mg/L) that was 1.6 times higher than that in 1977-1993 (1.6 mg/L).
The trend of increasing COD over the past thirty years can therefore be due to the high
COD after 1994.
The COD test is to measure the amount of organic compounds in the water. There
are thus three ways to account for the high COD after 1994. The first is an increase
in the discharge of river water with organic compounds. Saroma-ko Lagoon has
many rivers and thus variation in the inflow of river water appears as variation in
salinity in the lagoon. Saroma-betsu River accounts for about 60% of the area of the
river basin and about 80% of the inflow of all rivers
11
. The year-to-year trend in
salinity from 1975 to 2006 was examined using the median surface salinity between
May and November except October in each year reported at the station located the
mouth of Saroma-betsu River (Fig. 1c). As a result, neither an increasing nor a
decreasing trend was found during the thirty years (P>0.05) and there was furthermore
no significant difference in salinity between 1977-1993 and 1994- 2006 (U test,
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
4
P>0.05). In addition, Biological Oxygen Demand (BOD), an indicator of water
pollutants, has been measured downstream from the river
5
. A significantly
decreasing trend was found in mean BOD from 1975 to 2006 (P<0.05; Fig. 1d). It
can thus be said that the load of organic matter from the river basin has not increased
over the past thirty years.
The second is an increase in organic production by phytoplankton (primary
production) from spring to autumn in the lagoon. An increase in primary production
should appear as an increase in chlorophyll
a, an index of phytoplankton standing stock.
Unfortunately, the data set provided by the Hokkaido Government does not have
sufficient chlorophyll
a data to analyze long term trends. On the other hand, a
sufficient data set has been prepared for transparency, which can be empirically related
to variability in chlorophyll
a
12
. A significantly negative relationship was observed
between transparency and chlorophyll
a reported in the lagoon using data previously
reported (r
s
= -0.295, n = 518, p<0.01), indicating that transparency also becomes an
index of chlorophyll
a in the lagoon and hence primary production. Neither an
increasing nor a decreasing trend was found in mean transparency (P>0.05; Fig. 1e) and
there was furthermore no significant difference in the values between in 1977-1993 and
in 1994-2006 (U test, P>0.05). It can thus be said that organic production from spring
to autumn has not increased during the past thirty years.
Saroma-ko Lagoon has been the site of a very successful scallop culture operation.
Scallops excrete organic compounds. Thus, the third is an increase in the quantity of
scallops being raised. Although the scale of the aquaculture operation varied
year-to-year, a significant increasing trend was not found in the scale from 1975 to 2005
(r
s
=0.011, n=31, P>0.05) and there was no significant difference in quantities between
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
5
1975-1993 and in 1994-2005 (U test, P>0.05)
6
.
The above indicates that high COD after 1994 was not caused by an increase in
organic compounds from spring to autumn, necessarily suggesting that high COD after
1994 was due to events during the winter. The freezing of the entire surface is a
unique wintertime event at Saroma-ko Lagoon. A significantly deceasing trend was
found in the period of the freezing of the entire surface from 1964 to 2007 (Fig. 1f;
P<0.01). In particular, a remarkable trend of reduction was found from the late 1980s
to the early 1990s. The monthly mean air temperature, monthly mean daily maximum
temperature and monthly mean daily minimum temperature in Tokoro Town near
Saroma-ko Lagoon tended to increase during the past thirty years (Fig. 1g; P<0.05).
Relatively high values were observed in the three temperature measures especially
around 1990 (Fig. 1g). The warming of the air temperature has been indicated
throughout Eurasia in some years since 1988 due to the climate regime shift
13
. The
long term decreasing trend and remarkable reduction from the late 1980s to the early
1990s are considered to be due to climatic warming and the climate regime shift,
respectively.
The decrease in the periods of the freezing of the entire surface causes the thinning of
ice while also reducing to area and duration of the ice coverage
9
, resulting in higher
light intensity into the lower part of the ice and the seawater under the ice. In winter,
ice algae in the lower part of the ice and phytoplankton in the seawater under the ice
produce organic matter in the lagoon through photosynthesis
14, 15
. Photosynthetic rates
of both algae do not saturate at the
in situ light intensity and maximum photosynthetic
rates are several times higher than the rates at the
in situ light intensity
9, 16, 17
. Organic
production in Saroma-ko Lagoon in winter is thus easily expected to increase due to the
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
6
shortening of the periods of the freezing of the entire surface. Furthermore, most of
the organic matter produced possibly still remains with the coming of spring because of
low temperatures during the winter, resulting in an increase in COD from spring to
autumn. It would be no exaggeration to say that climate warming has contributed
much to the high COD since 1994.
The periods of the freezing of the entire surface has decreased gradually since 1964
(Fig. 1f), implying a gradual increase in COD. It is a fact, however, that an abrupt
increase in COD was found in 1994 as well as high COD since 1994 (Fig. 1b). The
periods of the freezing of the entire surface have been less than 80 days since 1994 with
the exception of 2001-2003, whereas a duration of less than 80 days was observed only
in 1974 (0 days) and in 1982 (79 days) before 1993 (Fig. 1f). The shortening of the
periods of the freezing of the entire surface improves light conditions for ice algae and
phytoplankton and thus a duration of less than 80 days is considered to be a necessary
condition for the occurrence of high COD in the lagoon. Durations during the period
of the regime shift (1988-1993) were less than 80 days as well. However, air
temperatures were high during the regime shift, implying high water temperature and
hence acceleration in the consumption of organic matter. Thus, high COD possibly did
not occur.
COD is composed of particulate and dissolved organic matter. There has been no
increasing or decreasing trend in transparency, which is influenced by particulate matter
(Fig. 1e), and the same result was found for suspended solids (SS) (P>0.05; Fig. 1h).
There was no significant difference in SS between 1977-1993 and 1994-2006 (U-test,
P>0.05). Consequently, most of the high COD after 1994 is possibly composed of
dissolved organic matter (DOM). An increase of DOM promotes microbial loop
18, 19
,
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
7
which then diminishes fish production
20
. It is possible that climatic warming promotes
not only a deterioration in water quality in boreal lakes characterized by ice coverage
but also the development of microbial loop and diminished fish production. In
addition, global warming possibly deteriorates water quality through increase of organic
production in winter. If global warming goes on, measures as usual, e.g. restriction of
load of organic compounds, may not stop the progress in the deterioration in water
quality.
METHODS SUMMARY
The items analyzed in this study were chemical oxygen demand (COD), salinity,
transparency, chlorophyll
a and suspended solids (SS) in Saroma-ko Lagoon, biological
oxygen demand (BOD) in Saroma-betsu River, periods of the freezing of the entire
surface of Saroma-ko Lagoon and air temperature at Tokoro Town near Saroma-ko
Lagoon. The COD, salinity, chlorophyll
a, transparency, SS and BOD were cited from
the annual reports of the Results of Water Quality Survey of Public Water Areas from
FY 1975 to FY2006
5
. Periods of the freezing of the entire surface were cited from the
website of the Saroma-ko Aquaculture Union
6
. Air temperature was cited from the
Climatic Statistics on the website of the Meteorological Agency of Japan
21
. The COD,
salinity, transparency and BOD were obtained from 1975 to 2006 (32 years), the
freezing periods from 1964 to 2007 (44 years) and air temperatures from 1977 to 2006
(30 years).
The surface data (0.5 m in depth) at seven stations (Stations 1-7)
5
located throughout
the lagoon from May to November except October were used for analyzing COD,
salinity, chlorophyll
a, SS and BOD in this study. The methods for water sampling and
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
8
analysis are found elsewhere
5
. COD was measured by the permanganate reflux
method. This method is inferior in oxidizing ability to the dichromate reflux method;
however, it poses no problems in elucidating trends over the past forty years. The
duration of complete ice coverage was monitored visually.
References
1. Schindler, D.W., Beaty, K.G., Fee, E.J., Cruikshank, D.R., DeBruyn, E.R., Findlay, D.L., Linsey, G.A.,
Shearer, J.A., Stainton, M.P. & Turner, M.A. Effects of climatic warming on lakes of the central boreal
forest. Science 250, 967-970 (1990).
2. Schindler, D.W., Bayley, S.E. & Parker, B.R. The effects of climatic warming on the properties of
boreal lakes and streams at the Experimental Lakes Area, northwestern Ontario. Limnol. Oceanogr.,
41, 1004-1017 (1996).
3. Magnuson, J.J., Robertson, D.M., Benson, B.J., Wynne, R.H., Livingstone, D.M., Arai, T., Assel, R.A.,
Barry, R.G., Card, V., Kuusisto, E., Granin, N.G., Prowse, T.D., Stewart, K.M. & Vuglinski, V.S.
Historical trends in lake and river ice cover in the northern hemisphere. Science, 289, 1743-1746
(2000).
4. Futter, M.N. Patterns and trends in southern Ontario lake ice phenology. Environ. Monit. Assess. 88,
431-444 (2003).
5. Hokkaido Government. Results of the Water Quality Survey of Public Water Areas during FY1975 to
FY2006 (1976-2006). The results of the Water Quality Survey are opened to the public in the net:
http: //envgis.hokkaido-ies.go.jp/ index.asp?category = mizu; http://www.pref.hokkaido.lg.jp/
ks/khz/contents/mizukankyo/down/ down.htm
6. Saroma-ko Aquaculture Union: http://homepage3.nifty.com/ saromako/
7. Satoh, H., Yamaguchi, Y., Watanabe, K. & Aruga, Y. Light conditions and photosynthetic productivity
of ice algal assemblages in Lake Saroma, Hokkaido. Jpn. J. Phycol. 37, 274-278 (1989a).
8. Satoh, H., Yamaguchi, Y., Watanabe, K., Tanimura, A., Fukuchi, M. & Aruga, Y. Photosynthetic
nature of ice algae and their contribution to the primary production in lagoon Saroma Ko, Hokkaido,
Japan. Proc. NIPR Symp. Polar Biol. 2, 1-8 (1989b).
9. Kudoh, S., Robineau, B., Suzuki, Y., Fujiyoshi, Y. & Takahashi, > Photosynthetic acclimation and the
estimation of temperate ice algal primary production in Saroma-ko Lagoon, Japan. J. Mar. Sys. 11,
93-109 (1997).
10. Anonymous, Lakes and Marshes in Hokkaido, revised edition (Hokkaido Institute of Environmental
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
9
Sciences, Sapporo, 2005).
11. Takeuchi, T., Sakata, T. & Hayase, Y. A study on environmental conservation in lagoon on cold
regions. Rept. Civil Eng. Res. Inst. 92, 1-103 (1990).
12. Falkowski, P.G. & Wilson, C. Phytoplankton productivity in the North Pacific ocean since 1900 and
implications for absorption of anthropogenic CO
2
. Nature 358, 741-743 (1992).
13. Watanabe, M. & Nitta, T. Decadal changes in the atmospheric circulation and associated surface
climate variations in the northern hemisphere winter. J. Climate 12, 494-510 (1999).
14. Kudoh, S. Responses of microalgal primary productivity to the reduction of sea ice coverage in Lake
Saroma during winter. Bull. Plankton Soc. Jpan. 39, 155-156 (1993).
15. Watanabe, K., Satoh, H. & Yamaguchi, Y. Phytoplankton and ice algae in Lake Saroma. Bull.
Plankton Soc. Jpn. 39, 165-166 (1993).
16. Priscu, C., Lizotte, M.P., Cota, G., Palmisano, A.C. & Sullivan, C.W. Comparison of the irradiance
response of photosynthesis and nitrogen uptake by sea ice microalgae. Mar. Ecol. Prog. Ser. 70,
201-210 (1991).
17. Lizotte, M.P., Sharp, T.R. & Priscu, J.C. Phytoplankton dynamics in the stratified water column of
Lake Bonney, Antarctica. Polar Biol. 16, 155-162 (1996).
18. Azam, F., Fenchel, T., Field, J.G., Gray, J.S., Meyer-Reil, L.A. & Thingstad, F. The ecological role of
water-column microbes in the sea. Mar. Ecol. Prog. Ser. 10, 257-263 (1983).
19. Azam,F. Microbial control of oceanic carbon flux: the plot thickens. Science 280, 694-696 (1998).
20. Williams, N. The Mediterranean beckons to Europe's oceanographer. Science 279, 483-484 (1998).
21. Meteorological Agency of Japan. Climatic Statistics (http:// www.jma.go.jp/jma/index.htm).
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
10
Figure legend
Figure 1. Long-term changes in Saroma-ko Lagoon. (a) COD. Linear relationship shown by solid
line (r = 0.297, n = 1260, P<0.01) and Spearman rank correlation (r
s
= 0.423, P<0.01) are significant.
The number of data in each year is 21-42. (b) Mean COD. Bars indicate standard errors. Linear
relationship shown by broken line (r = 0.585, n = 32, P<0.01) and Spearman rank correlation (r
s
= 0.573,
P<0.01) are significant from 1975 to 2006. Linear relationship (r = 0.089, n = 17, P>0.05) and
Spearman rank correlation (r
s
= -0.003, P>0.05) are not significant from 1977 to 1993. Linear
relationship (r = 0.077, n = 13, P>0.05) and Spearman rank correlation are not significant (r
s
= -0.033,
P>0.05) from 1994 to 2005. Gray line shows 3-years running mean. (c) Median salinity. Bars
indicate standard errors. Linear relationship by broken line (r = 0.198, n=30, P>0.05) and Spearman
rank correlation (r
s
= 0.114, P>0.05) are not significant. The number of data in each year is 20-41.
Gray line shows 3-years running mean. (d) Mean BOD. Bars indicate standard errors. Linear
relationship shown by broken line (r = -0.365, n=32, P<0.05) and Spearman rank correlation (r
s
= -0.447,
P<0.05) are significant. The number of data in each year is 3-6. Gray line shows 3-years running
mean. (e) Mean transparency. Bars indicate standard errors. Linear relationship shown by broken
line (r = -0.285, n=32, P>0.05) and Spearman rank correlation (r
s
= -0.274, P>0.05) are not significant.
Gray line shows 3-years running mean. (f) Periods of the freezing of the entire surface. Linear
relationship shown by broken line (r = -0.576, n=44, P<0.01) and Spearman rank correlation (r
s
= -0.747,
P<0.01) are significant. Gray line shows 4-years running mean. (g) Monthly mean air temperature
(solid diamond), monthly mean daily maximum temperature (solid square) and monthly mean daily
minimum temperature (solid triangle) in Tokoro Town near Saroma-ko Lagoon. Linear relationships
shown by broken lines (r = 0.371-0.494, n = 29, P<0.05) and Spearman rank correlation (r
s
=
0.405-0.496, P<0.05) are significant. Gray line shows 3-years running mean. (h) Mean suspended
solid (SS). Bars indicate standard errors. Linear relationship shown by broken line (r = -0.251, n=31,
P>0.05) and Spearman rank correlation (r
s
= -0.309, P>0.05) are not significant. The number of data in
each year is 21-42. Gray line shows 3-years running mean.
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008
Figure 1
0
2
4
6
8
10
12
14
1975
1980
1985
1990
1995
2000
2005
CO
D (
m
g
/
L
)
Year
a
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1975
1980
1985
1990
1995
2000
2005
CO
D (m
g
/
L
)
Year
b
29.5
30.0
30.5
31.0
31.5
32.0
32.5
33.0
33.5
1975
1980
1985
1990
1995
2000
2005
Sa
l
i
n
i
t
y
Year
c
0
1
2
3
4
5
6
1975
1980
1985
1990
1995
2000
2005
BO
D
(
m
g
/
L)
Year
d
0
1
2
3
4
5
6
7
8
9
1975
1980
1985
1990
1995
2000
2005
Tr
a
n
s
p
a
r
e
n
c
y
(
m
)
Year
e
0
2
4
6
8
10
12
14
16
18
20
1975
1980
1985
1990
1995
2000
2005
SS
(
m
g
/
L
)
Year
h
-2
0
2
4
6
8
10
12
14
1975
1980
1985
1990
1995
2000
2005
Te
m
p
e
r
a
t
u
r
e
(
)
Year
g
0
20
40
60
80
100
120
140
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
P
e
r
i
od
s
of
t
h
e
f
r
e
e
z
i
n
g of
t
h
e
e
n
t
i
r
e
s
u
rf
a
c
e (
d
a
y
s
)
Year
Nature Precedings : doi:10.1038/npre.2008.1857.1 : Posted 5 May 2008