Geochemical Journal, Vol. 48, pp. 9 to 18, 2014
doi:10.2343/geochemj.2.0279
Characteristics of hydrogen and oxygen stable isotope ratios in precipitation
collected in a snowfall region, Aomori Prefecture, Japan
HIDENAO HASEGAWA,1* NAOFUMI A KATA,1† H ITOSHI KAWABATA,1 TADAHIRO SATO,2
Y UKI CHIKUCHI1‡ and SHUN’ICHI HISAMATSU1
1
Institute for Environmental Sciences, Obuchi, Rokkasho, Aomori 039-3212, Japan
Tohoku Nuclear Co. Ltd., Higashi-Okamisawa, Misawa, Aomori 033-0024, Japan
2
(Received January 25, 2013; Accepted August 17, 2013)
Characteristics of δ2H and δ18O in precipitation samples collected at 16- to 17-d intervals at three stations at the
northern end of Honshu Island, Japan, were investigated. Rokkasho and Ajigasawa stations are approximately 100 km
apart, on the Pacific Ocean coast and the Sea of Japan coast, respectively, of Aomori Prefecture. Hakkoda station is
between Rokkasho and Ajigasawa stations at 1334 m ASL. Precipitation samples were continuously collected during
2000–2011 at Rokkasho, and during 2003–2006 at Ajigasawa and Hakkoda. At all stations, δ2H and δ18O of the precipitation samples showed weak seasonality, with higher values in summer and lower values in winter. In the summer (June,
July, and August), the intercept of the regression line between δ2H and δ18O, that is, the local meteoric water line (LMWL),
was lower than that of the global meteoric water line (GMWL) with similar slope, but in winter (December, January, and
February), the LMWL intercept was higher than the GMWL intercept with similar slope. Temporal variations of δ18O at
the two coastal stations were similar, although the timing of precipitation events was different. At all stations, d-excess
values showed clear seasonality, with high values in winter and low values in summer, indicating a seasonal change in the
source of the water vapor in air masses arriving over northeast Japan. In winter, d-excess values were higher at Hakkoda
than at the other stations, suggesting a precipitation at Hakkoda was derived from water vapor from a remote area of the
Sea of Japan. The long-term precipitation data set for δ2H and δ 18O obtained in this study will be useful for investigations
of regional hydrology and validation of numerical models.
Keywords: hydrogen isotope ratio, oxygen isotope ratio, d-excess, seasonal variation, Aomori Prefecture
Rozanski et al. (1993) determined the global relationship between 2H/1H and 18O/ 16O ratios in precipitation
by using data from 379 stations belonging to the Global
Network of Isotopes in Precipitation, established by the
International Atomic Energy Agency (IAEA) and World
Meteorological Organization. The relationship between
2
H/1H and 18O/16O ratios in precipitation in a particular
region, however, is affected by local geographical and
meteorological parameters such as latitude, altitude, distance from a coast, atmospheric temperature, and precipitation intensity (Dansgaard, 1964). For example, large
isotopic variations have been observed between coastal
and inland areas of North America separated by major
mountain ranges (Taylor, 1974). Because the Japanese
islands are situated off the eastern margin of the Asian
continent and extend more than 3000 km from north to
southwest, large variations in the 2H/ 1H and 18O/16O ratios in precipitation in Japan can be expected. In addition, seasonal changes in air mass sources are expected
to enhance the variation; air masses from the Pacific
Ocean cover Japan in summer, whereas in winter air
masses from the eastern Asian continent extend across
the Sea of Japan.
INTRODUCTION
Stable isotope ratios of hydrogen ( 2H/ 1H) and oxygen
( O/ 16O) in atmospheric water vapor and precipitation
vary spatially and temporally because of equilibrium and
kinetic isotopic fractionations associated with condensation and evaporation. Investigations of isotopic ratios of
water have significantly contributed to our understanding of atmospheric circulation (e.g., Epstein and Mayeda,
1953; Fritz et al., 1987; Hoffman et al., 2000). Moreover, many hydrological studies have investigated 2H/1H
and 18O/16O ratios in precipitation in relation to surface
water (Gonfiantini, 1985; Telmer and Veizer, 2000;
Gibson and Edwards, 2002), groundwater (Gat, 1981;
Yonge et al., 1989), and their interaction.
18
*Corresponding author (e-mail: [email protected])
†
Present address: National Institute for Fusion Science, Oroshi-cho,
Toki, Gifu 509-5292, Japan.
‡
Present address: JGC PLANTECH AOMORI Co. Ltd., Obuchi,
Rokkasho, Aomori 039-3212, Japan.
Copyright © 2014 by The Geochemical Society of Japan.
9
Fig. 1. Locations of the three stations at which precipitation samples were collected (open diamonds), in Aomori Prefecture,
Japan. The open circle indicates the place of Hachinohe where a typical climate of Pacific coast of Japan is observed in winter.
Mizota and Kusakabe (1994) mapped hydrogen and
oxygen isotopic data for surface and shallow groundwater
in Japan and the surrounding area using their own measurement data and data available in the literature, and others have examined the hydrogen and oxygen isotopic data
of local meteoric waters in relation to the regional hydrological system of central Japan (e.g., Kusakabe et al.,
1970). Data from northern Japan are relatively scarce,
however; only a few data covering relatively short time
periods have been reported from Aomori Prefecture,
northern Honshu Island (Aoki, 1992; Toyama et al., 2010,
2011). No long-term data set is available for analysis of
atmospheric circulation and the hydrological system in
this region.
Therefore, the objectives of this study were (1) to
document the characteristics of hydrogen and oxygen isotopic ratios in precipitation in Aomori Prefecture; (2) to
estimate the important factors controlling water isotopes;
(3) and thus to establish for this region a long-term data
set of isotopic ratios suitable for hydrologic studies.
MATERIALS AND METHOD
Study sites
Situated at the northern end of Honshu Island, Aomori
Prefecture is surrounded by ocean on three sides. The Ou
Mountains, which are located in the center of the prefecture, divide it into two parts, creating a complicated topography of inland and coastal features and, consequently,
highly varying climatic conditions in different areas of
the prefecture (Fig. 1).
Under typical winter conditions, a cold, wet air mass
from over the Sea of Japan collides with the Ou Moun10
H. Hasegawa et al.
tains, bringing heavy snowfall to the Sea of Japan side of
the prefecture. However, the mountains shield the southeastern part of the prefecture on the Pacific Ocean side,
with the result that many winter days in an area within
several tens of kilometers of Hachinohe city are sunny
and dry (Nibe, 1989). In summer, cool, humid easterly
winds (known as Yamase) from the Okhotsk high cause
many days on the Pacific Ocean side of the prefecture to
be characterized by low temperatures and high humidity.
Precipitation samples were collected at three stations
in Aomori Prefecture (Fig. 1). Rokkasho station is on the
rooftop of the Institute for Environmental Sciences building (40°57′56″ N, 141°21′38″ E, 30 m ASL) in Rokkasho
Village, approximately 2 km from the Pacific Ocean on
the east and 12 km from Mutsu Bay on the west. Rokkasho
Village experiences rather low temperatures in summer
(Pacific Ocean side type summer) and heavy snowfall in
winter (Sea of Japan side type winter). Ajigasawa station
is approximately 0.3 km from the Sea of Japan, on the
rooftop of the Saikai Elementary School building
(40°46′49″ N, 140°12′27″ E, 40 m ASL) in Ajigasawa
Town. Hakkoda station is at the summit of Mt.
Tamoyachidake (40°40′33″ N, 140°51′33″ E, 1334 m
ASL) in the Hakkoda Mountains, the northward extension of the Ou Mountains.
Sample collection and analytical methods
At Rokkasho and Ajigasawa stations, precipitation
samples were collected with a wet and dry deposition
sampler (US-750, Ogasawara Keiki Seisakusho Co. Ltd.,
Tokyo, Japan), which has an opening of 0.0314 m2 and
collects wet and dry deposition separately by means of
an automatically moving lid. In winter, the collection area
Fig. 2. Time series of Rokkasho station measurement results for October 2000 to December 2011: (a) δ2H and δ18O; (b) monthly
precipitation amount (bars) and mean temperature (line).
of the sampler is kept at 5–10°C by a circulating antifreeze solution that also melts any deposited snow. At
Hakkoda station, precipitation samples were collected in
summer by a total deposition sampler consisting of a
polyethylene tank connected to a polyethylene funnel with
an opening of 0.0798 m2. In winter, snow samples were
collected in an open polyethylene bag set in a vinyl chloride pipe with an opening of 0.0798 m 2.
At the three stations, samples were collected at a mean
interval of 16–17 d (range, 9–28 d). Samples were collected around 09:00 local time (LT) on each collection
day at both Rokkasho and Hakkoda stations, and at around
13:00 LT at Ajigasawa station. At Rokkasho station, samples were collected continuously from October 2000 to
December 2011. At Ajigasawa and Hakkoda stations, continuous sampling was carried out from November 2003
to March 2006. Each collected sample was weighed and
then passed through a filter with a pore size of 0.45 µm.
Hydrogen isotope ratios were measured using H 2 gas
generated by reduction of the sample water with zinc
metal at 490°C in a sealed quartz tube (Coleman et al.,
1982). For determination of the oxygen isotope ratio, the
water sample was equilibrated with CO2 at 25°C in a plastic syringe (Yoshida and Mizutani, 1986), and then the
isotope ratio of the CO 2 was measured. A dual-inlet stable isotope mass spectrometer (Optima, VG Isotech, UK,
or MAT252, Thermo Quest, USA) was used to measure
both ratios.
The conventional delta notation is used to express the
stable isotope ratios of hydrogen (δ2H) and oxygen (δ18O)
in the water samples; δ = (Rsample/Rstandard – 1), where
Rsample and R standard denote the 2H/1H or 18O/16O ratios in
the sample and the standard material, respectively. Vienna Standard Mean Ocean Water (VSMOW) distributed
by the IAEA (Vienna, Austria) was used as the standard
for both ratios. The measurement system was calibrated
by using IAEA reference materials, VSMOW, Greenland
Ice Sheet Precipitation, and Standard Light Antarctic Precipitation (SLAP). Values of –428‰ and –55.5‰ were
used for δ 2H and δ18O of SLAP, respectively, following
the guidelines of Coplen (1996). The estimated
reproducibilities (1σ), which were evaluated from routine measurements of the laboratory standard water, were
within 1.0‰ for δ 2H and within 0.1‰ for δ18O. Reported
monthly averaged δ2H and δ18O in this paper are means
of the measured isotope ratios weighted by the proportion of the total monthly precipitation amount.
RESULTS AND DISCUSSION
Seasonal variation of δ2H and δ18O in precipitation
The observed δ2H and δ18O in precipitation at the three
stations are shown in Figs. 2–4. At Rokkasho station during 2000–2011, δ2H of all precipitation samples ranged
Hydrogen and oxygen isotope ratios in precipitation, Aomori
11
Fig. 3. Time series of Ajigasawa station measurement results
for November 2003 to March 2006: (a) δ 2H and δ18 O; (b)
monthly precipitation amount (bars) and mean temperature
(line). Mean temperature data were collected by the Ajigasawa
Automated Meteorological Data Acquisition System (AMeDAS)
station operated by the Japan Meteorological Agency.
Fig. 4. Time series of Hakkoda station measurement results
for November 2003 to March 2006: (a) δ2 H and δ 18O; (b)
monthly precipitation amount (bars) and mean temperature
(line). Mean temperature data were collected at the Sukayu
AMeDAS station (890 m ASL; 444 m lower than Hakkoda station).
from –91.1‰ to –15.4‰, and δ 18O values varied from
–13.7‰ to –4.0‰ (Fig. 2). Mean δ2H and δ18O, weighted
by precipitation, were –56.2‰ and –8.9‰, respectively
(n = 132). During 2004–2005 (n = 24), when sampling
was also being conducted at Ajigasawa and Hakkoda stations, the weighted means of δ 2H and δ18O were –52.5‰
and –8.6‰, respectively.
At Ajigasawa station during 2003–2006, δ2H and δ18O
in the precipitation samples ranged from –70.2‰ to
–25.0‰ and from –10.9‰ to –4.5‰, respectively (Fig.
3). The weighted means during 2004–2005 were –52.3‰
for δ2H and –8.6‰ for δ18O (n = 24). Thus, they were
almost equal to the means at Rokkasho station during
2004–2005.
At Hakkoda station, which is at the highest elevation
(1334 m ASL; corresponding to the lower part of the free
atmosphere) among the three stations, δ 2H and δ18O in
the precipitation samples varied from –90‰ to –35.2‰
and from –14.0‰ to –6.9‰, respectively, during 2004–
2005 (Fig. 4). In this study, the minimum values of both
δ 2H and δ 18O were observed at this station. Weighted
means of all data at Hakkoda station were –64.3‰ and
–10.3‰ for δ2H and δ 18O (n = 24), respectively. These
means are clearly lower than those at Rokkasho and
Ajigasawa stations, but they are consistent with the δ2H
and δ18O values of shallow groundwater in this region of
Japan (Mizota and Kusakabe, 1994).
In general, the δ2H and δ18O values in precipitation at
the three stations do not show clear seasonal changes. The
δ2H and δ18O values (Figs. 2–4) show only weak seasonal variation, being generally higher in summer and
lower in winter. Araguas-Araguas et al. (1998) reported
similar weak seasonality in δ2H and δ18O of precipitation
at Tokyo and Ryori, Japan, and at Pohang, South Korea.
At Rokkasho station, however, 11-year monthly averaged δ18O values show clear seasonality, with higher values in summer and lower values in winter (Fig. 5).
12
H. Hasegawa et al.
Relation between δ 2H and δ18O, and deuterium excess in
precipitation
Craig (1961) reported that the stable isotope ratios of
hydrogen and oxygen in global precipitation were related
as follows: δ2H = 8 × δ 18O + 10. This relationship was
later defined as the Global Meteoric Water Line (GMWL).
Fig. 5. Variations of the 11-year averaged monthly δ18O value
(upper panel) and monthly averaged air temperature and
monthly precipitation (lower panel) at Rokkasho station. The
error bar means standard deviation.
The relationship between δ 2H and δ 18O at each station was compared with the GMWL (Fig. 6). The slope
of the regression line for summer (June, July, and August; δ 2H = 7.5 × δ 18O + 4.2, r = 0.98) at Rokkasho station was similar to the GMWL slope, but its intercept was
lower than the GMWL intercept. The slope of the regression line for winter (December, January, and February;
δ2H = 7.9 × δ18O + 22.6, r = 0.96) was also similar to the
GMWL slope, but its intercept was higher. The δ2H and
δ 18 O values in precipitation collected during spring
(March, April, and May) and autumn (September, October, and November) were distributed between the summer and winter regression lines. The results obtained for
Ajigasawa and Hakkoda stations are essentially in agreement with the Rokkasho station results.
The different intercepts can be explained by reference
to the deuterium excess (d-excess) concept, where dexcess = δ 2 H – 8 × δ 18O (Dansgaard, 1964). The dexcess value has been used as a tracer of conditions affecting evaporation in oceanic moisture source regions,
such as sea surface temperature and the relative humidity
of the overlying air mass (Merlivat and Jouzel, 1979).
Several studies have reported that d-excess values in precipitation in Japan vary seasonally (e.g., Waseda and
Nakai, 1983; Satake et al., 1984; Uemura et al., 2012),
and Uemura et al. (2008) have shown that d-excess val-
Fig. 6. Relationships between δ2H and δ18O in precipitation
samples collected at (a) Rokkasho, (b) Ajigasawa, and (c)
Hakkoda stations. The solid line in each panel is the global
meteoric water line (GMWL; δ2H = 8δ18O + 10), and dotted
lines in each panel are the local meteoric water lines (LMWL)
for summer (δ2H = 7.5 × δ18O + 4.2, r = 0.98) and winter (δ 2H
= 7.9 × δ18O + 22.6, r = 0.96) calculated from observed isotope data at Rokkasho station.
ues of water vapor above the ocean surface correlate negatively with the relative humidity above the ocean and positively with sea surface temperature. In general, higher dexcess values indicate lower relative humidity in the oceanic moisture source region. In this study, the d-excess
values of precipitation show clear seasonal variation at
all three stations, with lower values in summer and higher
values in winter (Fig. 7).
Hydrogen and oxygen isotope ratios in precipitation, Aomori
13
monthly precipitation amount, and the logarithm of
monthly precipitation (Log P) at each station are shown
in Table 1. δ 18O and surface air temperature showed a
weak positive correlation at Rokkasho and Ajigasawa stations, but at Hakkoda station, the correlation was not significant (p > 0.05). Monthly precipitation amounts did
not show a clear correlation with δ18O at any of the three
stations. When we compared the slopes of the regression
lines between δ 18 O and surface air temperature, the
monthly precipitation amount and the logarithm of
monthly precipitation (Log P) among seasons at Rokkasho
station (Table 2), however, we found a positive correlation of δ18O with air temperature in spring, and a negative correlation of δ18O with the two precipitation parameters in winter.
Johnson and Ingram (2004) applied a multiple regression model to the analysis of spatial and temporal variations of stable isotopes in modern precipitation in China,
as follows:
δ18O = β0 + βtT + βlogP logP
Fig. 7. Temporal variation of monthly averaged (a) d-excess,
(b) δ18O in precipitation, and (c) monthly precipitation amount
at Rokkasho (solid line), Ajigasawa (dashed line), and Hakkoda
(chain line) during November 2003 to March 2006.
Relation to local meteorological parameters
At mid-latitudes, both temperature and the rainfall
amount are important factors controlling the final isotopic
compositions of precipitation (Dansgaard, 1964; Lawrence and White, 1991). The slopes of the linear regression lines between δ18O and surface air temperature, the
14
H. Hasegawa et al.
where β0 is the y-intercept, and βt (‰/°C) and βlogP (‰/
log mm) are the partial regression coefficients for temperature and the logarithm of precipitation, respectively.
We applied this multiple regression model to the data at
the three stations (Table 3). As a result, at Rokkasho station, we obtained a regression equation (R2 = 0.37) with
βt = 0.14 and βlogP = –2.64. These partial regression coefficients are similar to those at Yantai, on the coast of northeastern China (βt = 0.11, βlogP = –3.26), where the δ 18O
time series is controlled by both air temperature and precipitation amount (Johnson and Ingram, 2004). The δ 18O
data at the other two stations, however, were not significantly correlated with air temperature or monthly precipitation amount. The higher coefficient of determination
(R2 = 0.37) of the multiple regression analysis compared
with the linear regression analysis (R2 = 0.03–0.21) suggests that δ18O in precipitation at Rokkasho station may
be controlled by both local meteorological factors (temperature and precipitation amount). To assess the
seasonality of both βt and βlogP, we applied the multiple
regression model to the δ 18 O data of each season at
Rokkasho station (Table 4). We found significant correlations in three seasons. In spring and autumn, δ18O was
positively correlated with air temperature and negatively
correlated with logP. These correlations are consistent
with well-known effects (temperature effect, amount effect; e.g., Dansgaard, 1964). In summer, δ18O did not
show any clear relation with air temperature or precipitation amount, but in winter, the results of the model analysis show that δ18O was negatively correlated with both T
and logP. The liquid-vapor equilibrium isotope
fractionation factor increases with decreasing tempera-
Table 1. Slope of regression lines between δ18O and air temperature (T), precipitation
amount (P), and logarithm of the precipitation amount (logP) during the study period
Station
Rokkasho
Ajigasawa
Hakkoda
n
T
R2
p
P
R2
p
logP
R2
p
132
29
29
0.10
0.07
0.06
0.21
0.15
0.10
0.00
0.04
0.09
−0.005
−0.006
0.0006
0.03
0.04
0.00
0.03
0.29
0.78
−1.40
−0.81
0.77
0.05
0.01
0.01
0.01
0.54
0.54
Table 2. Slopes of regression lines between δ18O and air temperature (T), precipitation
amount (P), and logarithm of precipitation amount (logP) during each season at Rokkasho
station
Season
n
T
R2
p
P
R2
p
logP
R2
p
Spring
Summer
Autumn
Winter
33
33
34
32
0.22
0.08
0.12
−0.28
0.31
0.02
0.10
0.13
0.00
0.41
0.07
0.04
−0.01
−0.006
−0.009
−0.02
0.09
0.08
0.15
0.35
0.09
0.10
0.03
0.00
−2.13
−1.96
−2.45
−2.89
0.11
0.09
0.14
0.35
0.06
0.09
0.03
0.00
Table 3. Summary of the multiple regression
analysis results for the three stations
Table 4. Summary of the multiple regression
analysis results for each season at Rokkasho
station
n
βT
βP
R2
p
Rokkasho
132
0.14
−2.64
0.37
0.00
Season
Ajigasawa
Hakkoda
29
29
0.08
0.06
−0.94
−0.31
0.17
0.10
0.09
0.24
Spring
Summer
Autumn
Winter
Station
ture (e.g., Horita and Wesolowski, 1994). If the condensation temperature is assumed to be correlated with the
surface air temperature, condensed precipitation should
be isotopically lighter under the lower air temperature
condition (Dansgaard, 1964). The observed negative correlation between air temperature and δ18O is contrary to
this expectation. According to our observations in this
study, monthly precipitation and the monthly averaged
air temperature were positively correlated in winter. There
is a possibility that precipitation amount is more effective factor controlling δ18 O of winter precipitation at
Rokkasho station. However, the relationships between
δ18O and local air temperature or amount of precipitation
are not sufficient to explain variation of δ18O in precipitation at all three stations. Because the variation of isotope ratios of precipitation at three stations are controlled by various meteorological parameters. The isotope
system in precipitation at all three stations thus needs to
be analyzed in relation to large-scale water circulation.
Relation to air mass pathways
Figure 7 shows time series of monthly averaged dexcess, δ18O in precipitation, and monthly precipitation
amounts at the three stations from November 2004 to
βT
βP
R2
p
33
0.25
−2.86
0.50
0.00
33
34
32
0.08
0.15
−0.11
−1.94
−2.79
−2.61
0.11
0.28
0.37
0.17
0.01
0.00
n
March 2006. At Hakkoda station, d-excess values tended
to be higher and δ18O in precipitation tended to be lower
than at the other two stations.
Even though Rokkasho and Ajigasawa stations are 100
km apart and precipitation events were not synchronous
between them, temporal changes in δ 18O in precipitation
and the d-excess value were remarkably similar at these
two stations. As shown in Figs. 2–4, the relationships
between δ18O in precipitation and local air temperature
or monthly precipitation amount were not very clear. In
addition, the expected isotope variation of approximately
2‰ (based on the observed summer-winter air temperature difference of 22°C at Rokkasho station), calculated
from the temperature dependency of equilibrium
fractionation (0.1‰/°C; e.g., Majoube, 1971), was less
than the observed summer-winter variation of δ18O (10‰)
at Rokkasho station. These facts suggest that the important factor controlling monthly averaged δ 18O in precipitation was synoptic-scale vapor circulation rather than
local meteorological parameters.
Backward air mass trajectory analysis is a useful tool
for estimating air mass origins, and many researchers have
Hydrogen and oxygen isotope ratios in precipitation, Aomori
15
Fig. 8. Results of the backward air mass trajectory analysis for summer (left panels) and winter (right panels) at the three
stations: (a) Rokkasho, (b) Ajigasawa, and (c) Hakkoda.
demonstrated a relationship between air mass origins and
isotopic signals (e.g., He et al., 2006; Pfahl and Wernli,
2008). Therefore, we calculated 72-h backward air mass
trajectories starting at the altitude of 500 m (1500 m at
16
H. Hasegawa et al.
Hakkoda) using the NOAA HYSPLIT model (Draxler and
Rolph, 2013; Rolph, 2013) for the three stations in summer (August 2005) and winter (December 2004) (Fig. 8).
In winter, most trajectories passed over the Sea of Japan,
whereas in summer, they did not pass primarily over any
particular area. In winter, the pathways by which the air
masses arrived at the two coastal stations did not differ
very much, and they transited east coast of Russia. The
pathway of the air masses arrived at Hakkoda station
transited through the northeast of Korean Peninsula. In
summer, however, though the pathways of air masses arriving at Rokkasho and Ajigasawa were almost the same,
the pathways of those arriving at Hakkoda frequently
crossed over the Asian continent.
Yoshimura and Ichiyanagi (2009) used an atmospheric
general circulation model that incorporates isotope data
(IsoGSM; Yoshimura et al., 2008) to analyze the mechanism of the remarkable seasonal change in d-excess values in precipitation observed in East Asia. In winter, the
East Asian seas and western Pacific Ocean are much
warmer than the overlying air, which causes the effective
relative humidity (h*, vapor pressure normalized by the
saturated vapor pressure at sea surface temperature) to
be low and the water vapor to have high d-excess values.
The model results showed a higher d-excess value in the
western Sea of Japan (east coast of Korean Peninsula)
and the Pacific Ocean adjacent to south coast of Japan.
In contrast, the vapor distributed over the central Sea of
Japan and the Pacific Ocean adjacent to northeast coast
of Japan showed lower d-excess values.
The results of backward air mass trajectory analysis
suggest that the vapor arrived at Hakkoda station have
different moisture source. The higher d-excess values at
Hakkoda station than at the other stations reflect the water vapor arrived at the station derived from relatively
remote ocean areas. Thus, the precipitation that falls at
Hakkoda station in winter is mainly derived from water
vapor from the western Sea of Japan areas (east coast of
Korean Peninsula). In contrast, the relatively lower dexcess values observed at the two coastal stations indicate that the winter precipitation was derived from water
vapor that evaporated from the central or eastern parts of
the Sea of Japan.
CONCLUSIONS
We reported the measurement results of δ 2H and δ18O
values in precipitation samples collected at Rokkasho,
Ajigasawa, and Hakkoda stations. Values were generally
lower at Hakkoda station than at the two coastal stations,
and they showed weak seasonal variation at all three stations. In particular, the 11-year monthly averaged δ18O
in precipitation at Rokkasho station showed clear
seasonality, with higher values in summer and lower values in winter. However, δ18O was not clearly correlated
with local meteorological parameters. Temporal variations
of δ18O at the two coastal stations were similar, although
they face different oceans, and the timing of precipita-
tion events differed between them. Thus, conditions in
the source area of the water vapor are a more important
control on δ18O values at the three stations than local climatic conditions.
The d-excess values at all stations showed clear
seasonality, being high in winter and low in summer, reflecting seasonal changes in the water vapor sources of
the air masses arriving over northeast Japan. Higher dexcess values were observed at Hakkoda station in winter than at the other two stations, however, suggesting
that precipitation at Hakkoda was derived from water
vapor from a remote area of Sea of Japan.
To clarify in more detail the relationship between the
water isotopic system and climatic parameters, eventbased observations and numerical model-based analyses
are required.
Acknowledgments—The authors are grateful to Mr. Koichi
Nishimura and Mr. Kaoru Kogawa for their assistance in collecting the precipitation samples. We also thank all of the staff
members at Hakkoda Ropeway Co., Ltd. and the teachers at
Saikai Elementary School for supporting the precipitation sample collection. The manuscript benefitted from comprehensive
review by three anonymous reviewers.
This work was performed under a contract with the government of Aomori Prefecture, Japan.
REFERENCES
Aoki, M. (1992) Magmatic fluid discharged to the surface from
the Osorezan geothermal system, northern Honshu, Japan.
Rep. Geol. Surv. Jpn. 279, 16–21.
Araguas-Araguas, L., Froehilich, K. and Rozanski, K. (1998)
Stable isotope composition of precipitation over southeast
Asia. J. Geophys. Res. 103(D22), 28721–28742.
Coleman, M. L., Shepherd, T. J., Durham, J. J., Rouse, J. E.
and Moore, G. R. (1982) Reduction of water with zinc for
hydrogen isotope analysis. Anal. Chem. 54, 993–995.
Coplen, T. B. (1996) New guidelines for reporting stable hydrogen, carbon, and oxygen isotope-ratio data. Geochim.
Cosmochim. Acta 60, 3359–3360.
Craig, H. (1961) Isotope variations in meteoric waters. Science
133, 1702–1703.
Dansgaard, W. (1964) Stable isotopes in precipitation. Tellus
16, 436–468.
Draxler, R. R. and Rolph, G. D. (2013) HYSPLIT (HYbrid
Single-Particle Lagrangian Integrated Trajectory) Model access via NOAA ARL READY Website (http://
www.arl.noaa.gov/HYSPLIT.php). NOAA Air Resources
Laboratory, College Park, MD.
Epstein, S. and Mayeda, T. (1953) Variation of 18O content of
waters from natural sources. Geochim. Cosmochim. Acta 4,
213–224.
Fritz, P., Drimmie, R. J., Frape, R. K. and O’Shea, K. (1987)
The isotopic composition of precipitation and groundwater
in Canada. Isotope Techniques in Water Resource Development, 539–550, International Atomic Energy Agency, Vienna.
Hydrogen and oxygen isotope ratios in precipitation, Aomori
17
Gat, J. R. (1981) Paleoclimate conditions in the Levant as revealed by the isotopic composition of paleowaters. Israel
Meteorological Research Paper 3, 13–28.
Gibson, J. J. and Edwards, T. W. D. (2002) Regional surface
water balance and evapotranspiration partitioning from a
stable isotope survey of lakes in northern Canada. Global
Biogeochem. Cycles 16, doi:10.1029/2001GB001839.
Gonfiantini, R. (1985) On the isotopic composition of precipitation tropical stations. Acta Amazonica 15, 121–139.
He, Y. Q., Pang, H. X., Theakstone, W. H., Zhang, Z. L., Lu, A.
G. and Gu, J. (2006) Isotopic variations in precipitation at
Bangkok and their climatological significance. Hydrol.
Process. 20, 2873–2884.
Hoffman, G., Jouzel, J. and Masson, V. (2000) Stable isotopes
in atmospheric general circulation models. Hydrolog. Process. 14, 1385–1406.
Horita, J. and Wesolowski, D. J. (1994) Liquid-Vapor
fractionation of oxygen and hydrogen isotopes of water from
the freezing to the critical temperature. Geochim.
Cosmochim. Acta 58(16), 3425–3437.
Johnson, K. R. and Ingram, B. L. (2004) Spatial and temporal
variability in the stable isotope systematics of modern precipitation in China: implications for paleoclimate reconstructions. Earth Planet. Sci. Lett. 220, 365–377.
Kusakabe, M., Wada, H., Matsuo, S. and Horibe, J. (1970)
Oxygen and hydrogen isotope ratios of monthly collected
waters from Nasudake volcanic area, Japan. J. Geophys. Res.
75, 5941–5951.
Lawrence, J. R. and White, W. C. (1991) The elusive climate
signal in the isotopic composition of precipitation. In Stable Isotope Geochemistry: A tribute to Samuel Epstein. The
Geochemical Society Special Publication 3 (Taylor, H. P.,
O’Neil, J. R. and Kaplan, I. R., eds.), 169–185, Lancaster
Press.
Majoube, M. (1971) Oxygen-18 and deuterium fractionation
between water and steam. J. Chim. Phys., 68, 1423–1436
(in French).
Merlivat, J. and Jouzel, J. (1979) Global climatic interpretation of the deuterium-oxygen 18 relationship for precipitation, J. Geophys. Res. 84, 5029–5033.
Mizota, C. and Kusakabe, M. (1994) Spatial distribution of δD–
δ18O values of surface and shallow groundwaters from Japan, South Korea and east China. Geochem. J. 28, 387–
410.
Nibe, H. (1989) Aomori no Tenki. Hoppo Shinsha, 282 pp. (in
Japanese).
Pfahl, S. and Wernli, H. (2008) Air parcel trajectory analysis of
stable isotopes in water vapor in the eastern Mediterranean.
J. Geophys. Res. 113, doi:10.1029/2008JD009839.
Rolph, G. D. (2013) Real-time Environmental Applications and
Display
sYstem
(READY) Website
(http://
www.ready.noaa.gov). NOAA Air Resources Laboratory,
College Park, MD.
Rozanski, K., Araguas-Araguas, L. and Gonfiantini, R. (1993)
Isotopic pattern in modern global precipitation. Climatic
Change in Continental Isotopic Records, Geophys. Monogr.,
18
H. Hasegawa et al.
78 (Swart, P. K., Lowhan, K. L., Mckenzie, J. and Savin,
S., eds.), 1–37, American Geophysical Union, Washington,
D.C.
Satake, H., Mukai, T. and Mizutani, Y. (1984) Environmental
isotope hydrology of precipitations and river waters in the
Hokuriku district, Japan. Res. Rpt. Tritium Res. Cent.,
Toyama Univ. 3, 45–56 (in Japanese with English abstract).
Taylor, H. P., Jr. (1974) The application of oxygen and hydrogen isotope studies to problems of hydrothermal alteration
and ore deposition. Econ. Geol. 69, 843–883.
Telmer, K. and Veizer, J. (2000) Isotopic constraints on the transpiration, evaporation, energy and GPP budgets of a large
boreal watershed: Ottawa River basin, Canada. Global
Biogeochem. Cycles 14, 149–166.
Toyama, S., Ii, H., Uehara, T., Koizumi, Y. and Wamuro, H.
(2010) Stable isotopic composition and origin of hot spring
water in Shimokita Peninsula, Aomori Prefecture. J. Hot
Spring Sci. 59, 261–272 (in Japanese with English abstract).
Toyama, S., Suzuki, Y., Nakashita, R. and Aikawa, Y. (2011)
Correlation of isotope ratios of light elements between ambient water and polished rice, Tsugaru Plain, Aomori Prefecture. BUNSEKI KAGAKU 60, 563–570 (in Japanese with
English abstract).
Uemura, R., Matsui, Y., Yoshimura, K., Motoyama, H. and
Yoshida, N. (2008) Evidence of deuterium excess in water
vapor as an indicator of ocean surface conditions. J.
Geophys. Res. 113(D19114), doi:10.1029/2008JD010209.
Uemura, R., Yonazawa, N., Yoshimura, K., Asami, R., Kadena,
H., Yamada, K. and Yoshida, N. (2012) Factors controlling
isotopic composition of precipitation on Okinawa Island,
Japan: Implications for reconstruction in the East Asian
Monsoon region. J. Hydrol. 475, 314–332.
Waseda, A. and Nakai, N. (1983) Isotopic compositions of meteoric and surface waters in Central and Northeast Japan.
Chikyukagaku, 17, 83–91 (in Japanese with English abstract).
Yonge, C. J., Goldenberg, L. and Krouse, H. R. (1989) An isotope study of water bodies along a traverse of southwestern
Canada. J. Hydrol. 106, 245–255.
Yoshida, N. and Mizutani, Y. (1986) Preparation of carbon dioxide for oxygen-18 determination of water by use of a plastic syringe. Anal. Chem. 58, 1273–1275.
Yoshimura, K. and Ichiyanagi, K. (2009) A reconsideration of
seasonal variation in precipitation deuterium excess over
east Asia. J. Japan Soc. Hydrol. and Water Resour. 22(4),
262–276 (in Japanese with English abstract).
Yoshimura, K., Kanamitsu, M., Noone, D. and Oki, T. (2008)
Historical isotope simulation using Reanalysis atmospheric
data. J. Geophys. Res. 113, doi:10.1029/2008JD010074.
S UPPLEMENTARY MATERIALS
URL (http://www.terrapub.co.jp/journals/GJ/archives/
data/48/MS279.pdf)
Table S1