Origin of methane in high-arsenic groundwater of Taiwan – Evidence
from stable isotope analyses and radiocarbon dating
Tsung-Kwei Liu
a,*, Kuan-Yu Chen
a,b, Tsanyao Frank Yang
a, Yue-Gau Chen
a, Wen-Fu Chen
c, Su-Chen Kang
a,
Chih-Ping Lee
aa
Department of Geosciences, National Taiwan University, Taipei, Taiwan, ROC
bEnergy and Environment Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan, ROC c
Institute of Hot Spring Industry, Chia Nan University of Pharmacy and Science, Tainan County, Taiwan, ROC
a r t i c l e
i n f o
Article history:
Received 29 August 2008
Received in revised form 12 June 2009 Accepted 25 June 2009
Keywords: Methane (CH4)
Dissolved inorganic carbon (DIC) High-arsenic groundwater Taiwan
Stable isotopes Radiocarbon dating
a b s t r a c t
Groundwaters in the confined aquifers of the Chianan and Ilan coastal plains of Taiwan are rich in dis-solved methane (CH4). Serious endemic ‘‘blackfoot disease”, which occurred in the Chianan plain,
espe-cially during AD1950-1970, has been demonstrated to have arisen from drinking highly reducing groundwater with abnormal arsenic and humic substance levels. In order to explore the origin of CH4
and its hydrological implications, stable carbon isotope ratios (d13C) and radiocarbon (14C) ages of
exsolved CH4, dissolved inorganic carbon (DIC), and sedimentary biogenic sediments from a total of 34
newly completed water wells at 16 sites were determined. The main results obtained are as follows: (1) The d13C
CH4 (65‰ to 75‰) values indicate that, except for one thermogenic sample
ðd13C
CH4¼ 38:2‰Þ from the Ilan plain, all CH4samples analyzed were produced via microbially mediated
CO2reduction. Many d13CDICvalues are considerably greater than 10‰ and even up to 10‰ due to
Rayleigh enrichment during CO2reduction. (2) Almost all the14C ages of CH4samples from the shallow
aquifer (I) (<60 m depth) are greater than the14C ages of coexisting DIC and sediments, suggesting the
presence of CH4from underlying aquifers. (3) The14C ages of coexisting CH4, DIC and sediments from
aquifer (II) of the Chianan plain are essentially equal, reflecting in-situ generation of CH4and DIC from
decomposition of sedimentary organic matter and sluggishness of the groundwater flow. On the other hand, both CH4and DIC from each individual well of the relatively deep aquifers (III) and (IV) in the
Chi-anan plain are remarkably younger than the deposition of their coexisting sediments, indicating that cur-rent groundwaters entered these two aquifers much later than the deposition of aquifer sediments. (4) Each CH4sample collected from the Ilan plain is older than coexisting DIC, which in turn is distinctly
older than the deposition of respective aquifer sediments, demonstrating the presence of much older CO2and CH4from underlying strata.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Considerable amounts of dissolved methane (CH4) are often
present in groundwaters pumped from wells tapping confined aquifers in the Chianan and Ilan alluvial coastal plains of SW and NE Taiwan, respectively (Fig. 1). These groundwaters are also char-acterized by high concentrations of dissolved humic substances and arsenic (Chen and Liu, 2007), both of which were considered to be responsible for endemic ‘‘blackfoot disease” through drinking of groundwater (Tseng et al., 1961; Lu, 1990). It has long been rec-ognized that occurrence of CH4 generally implies a considerably
reducing state in the geochemical environment (Stumm and
Mor-gan, 1996). In addition, strongly reducing waters tend to be con-taminated by heavy metals and natural discharges are likely to be minimal (Smedley and Kinniburgh, 2002; Gooddy and Darling, 2005). Previous studies of dissolved gas from the six water wells drilled by the Chinese Petroleum Corporation in the Chianan plain focused mainly on the potential for profitable production (Hsu, 1984), and the composition of the gas was shown to contain more than 90% CH4, up to 6% CO2, and a few percent N2. However, the
origins and source of CH4 in these aquifers have not been well
studied.
Dissolved methane in groundwater can be formed via bacterial reduction, or from thermogenic decomposition of organic matter at relatively high temperatures. Furthermore, the former type can proceed through either reduction of dissolved forms of CO2 (i.e.,
CO2+ 4H2?CH4+ 2H2O) or acetate fermentation (i.e., CH3COO+
H2O ? CH4+ HCO3) (Schoell, 1980, 1988; Oremland et al., 1982).
1367-9120/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jseaes.2009.06.009
*Corresponding author. Tel./fax: +886 2 2365 7380. E-mail address:liutk@ntu.edu.tw(T.-K. Liu).
Contents lists available atScienceDirect
Journal of Asian Earth Sciences
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j s e a e sThe best diagnostic tool for identifying the origin of CH4in the
groundwater environment appears to be the13C and2H
fraction-ation between coexisting CH4and CO2(Oremland, 1988; Coleman
et al., 1988; Whiticar et al., 1986; Whiticar, 1999; Lansdown et al., 1992; Aravena et al., 1995, 2003). It has long been recognized that the stable carbon isotope ratios (d13C) of thermogenic methane range from 30‰ to 50‰, and are mostly greater than 45‰ (Barker and Fritz, 1981; Tyler, 1991). By contrast, d13C values for
microbial CH4 are less than 50‰ and mostly fall in the range
50‰ to 65‰ for acetate fermentation pathway and 65‰ to 90‰ for CO2reduction pathway. In addition to the problem of
the origin of CH4, it is also interesting to know whether CH4was
formed in-situ. The purpose of this study is to verify the origin and source of CH4 associated with groundwater in the Chianan
and Ilan plains by using stable carbon isotopes and radiocarbon dating of CH4, DIC and shell (or plant) sediments collected from
34 recently drilled groundwater-monitoring wells. A d2H value
was measured for a CH4sample to further support its origin
iden-tified by d13C criterion.
2. Hydrogeological setting 2.1. Chianan plain
The Chianan plain is a part of the coastal plain of southwestern Taiwan, covering about 1100 km2with a N–S length of
approxi-mately 40 km and an average E–W width of 30 km. The ground surface slopes seaward with a very low gradient. It is bounded on the east by the western edge of the Western Foothills, which is mostly composed of siltstone and fine-grained sandstone of the late Tertiary epoch and provides the source rocks for the Chia-nan plain (Ho, 1975). Accordingly, the upper-most 300 m of which is mostly silt and clay, intercalated with thin lenticular fine-sand layers (Fig. 2a; Chen and Liu, 2007); these sediments
were deposited in mixed sedimentary environments, including la-goon, estuarine, shallow marine, and fluvial plains during late Pleistocene and Holocene. The average rate of sediment accumula-tion during Holocene reached a high value of 1 cm/yr in the coastal areas as determined by radiocarbon dating (Liu et al., 1997).
Based on grain-size of sediments, the upper 300-m thick stra-ta at each well locality are generally divided into four aquifers: I, II, III and IV from top to bottom. Unconfined aquifers are limited to the upper few meters of the plain, even at the eastern parts adjoin-ing the hills. Hydrogeological characteristics are consistent with the very low permeability shown in pumping tests and the occur-rence of saline water in the aquifers deposited in a marine environ-ment.Chen and Liu (2007)interpreted the high chlorinity for some wells in aquifer (I) as being due to strong evaporation in the coastal zone.
2.2. Ilan plain
The Ilan plain is located in the western tip of the rifting back-arc basin of the Okinawa Trough. This plain is triangular in shape and bounded by low-grade metamorphic rocks of the Hsueshan Range and medium-high grade metamorphic rocks of the Central Range to the northwest and the south, respectively. The upper 120-m thick layer was deposited during Holocene and its average sedi-ment accumulation rate is similar to the Chianan plain, also reach-ing a high of 1 cm/yr. Grain-sizes of sediments in most parts of the plain decrease distinctly from gravel in the proximal part to silt/clay in the coastal area. A relatively shallow (<20 m depth) and thick (10 m) clayey layer acts as the upper-most aquitard overlying the confined aquifers in the middle and distal parts. The CH4samples analyzed were taken from the southern margin
of the plain, where the whole pile of strata drilled through for this study is basically composed of dark-gray clay intercalated with thin layers of silt and fine-grained sand (Fig. 2b).
3. Samples and analytical methods
During the establishment of the groundwater-monitoring net-work of Taiwan, continuous sediment cores of up to 250 m or more in length were taken and up to four wells tapping individual aqui-fers were drilled at each monitoring site. Most of the water sam-ples were collected from the same wells as those used byChen and Liu (2007)for arsenic study. The water samples were ensured to be fresh and representative by pumping each well long enough to replace the free-standing water in the casing before taking sam-ples. About 50 l of each water sample was collected from these wells for precipitation of dissolved inorganic carbon (DIC) for14C
dating using the procedures described previously by Hackley et al. (1992)andLiu (1995a). Collection of exsolved gases for gas isotopic analyses was carried out using inverted glass bottles sub-merged in sample water at the discharge outlet of the flow cham-ber. The depositional ages of aquifer sediments corresponding to the depth of each well screen were either determined by14C dating
of biogenic (i.e., plant or shell) fragments from the sediment cores corresponding to well screens, or estimated by correlation of neighboring strata with known 14C ages (Chen and Liu, 2007).
The original CO2,if any, in the gas samples was eliminated by
bub-bling the gas through 0.5 M BaCl2solution under pH 13 to facilitate
carbonate (BaCO3) precipitation, and further passing through an
Fig. 2. Hydrogeological profiles of (a) the Chianan plain and (b) the Ilan plain. The strata within the drilling range are divided into four aquifers: (I), (II), (III), and (IV) from top to bottom. Location of profile lines is shown inFig. 1.
ascarite column before any hydrocarbon gas was combusted and converted to CO2for14C dating and d13C determinations. The14C
activity was measured by the ultra-low level liquid scintillation counter, Quantulus 1220TM(Gupta and Polach, 1985). The maxi-mum determinable radiocarbon age, defined by the two above-background conventions described byStuiver and Polach (1977), is 50 Ka for our laboratory. The zinc technique (Coleman et al., 1988) was used to convert H2O resulting from combustion of a
CH4sample to H2 for d2H analyses. Stable isotope analyses were
performed on a triple-collecting mass spectrometer and are ex-pressed in the usual delta notation relative to the PDB (d13C) and
SMOW (d2H) standards. Analytical reproducibility was better than
0.1‰ for d13C and 3‰ for d2H. The14C results of DIC and CH4were
reported as conventional14C ages (Stuiver and Polach, 1977) for
comparison with depositional ages of aquifer sediments. 4. Results and discussion
The analytical results of d13C and14C ages for methane and
dis-solved inorganic carbon are listed inTable 1. In addition, strati-graphic 14C ages for each screen level are estimated by lateral
correlation and average depositional rates based on the age data ofChen and Liu (2007). It is evident that except for sample IL-01(IV) from the Ilan plain all other d13C values for CH
4samples fall
in the range 67‰ to 78‰, demonstrating that CH4was formed
via bacteria mediated CO2reduction (Fig. 3). This is further
sup-ported by the d2H (211.8‰) and d13C (65.31‰) values for
meth-ane from well CN-03(III) on the classification scheme ofColeman et al. (1995)and the generally positive correlation between the
14C
DICand14CCH4ages (Fig. 4). A general increasing trend of
14C DIC
Table 1
14C ages and d13C values for dissolved methane (CH
4) and inorganic carbon (DIC), and stratigraphic age of each well in the Chianan and Ilan plains.
Sample sitea
Aquifer Well screen depth (m) CH4 DIC Estimated Stratigraphic14C age (yBP)
14 C age (yBP) d13C (‰ PDB) 14 C age (yBP) d13C (‰ PDB) CN-01 (24)b (II) 62–86 10010 ± 60 74.3 10050 ± 70 18.2 10000 CN-02 (27) (I) 32–50 6210 ± 60 76.6 5070 ± 50 4.5 7000–9000 CN-03 (22) (I) 32–50 14400 ± 80 74.3 9400 ± 50 43.5 7000–9000 (II) 104–119 17280 ± 100 75.8 17590 ± 140 1.4 17000 (III) 150–168 25730 ± 230 82.2 27800 ± 200 15.7 S35000 CN-04 (23) (II) 94–114 13500 ± 75 74.2 13200 ± 70 7.3 12000 (III) 170–182 13360 ± 80 74.1 13840 ± 90 0.2 S40000 (IV) 216–228 13520 ± 80 73.8 17300 ± 100 6.2 >40000 CN-05 (14) (I) 18–30 10650 ± 110 82.2 5390 ± 50 2.5 5100 (II) 114–126 14340 ± 100 68.1 14580 ± 200 8.0 12000 (III) 172–190 15820 ± 90 68.3 15070 ± 220 3.3 S40000 (IV) 241–250 12800 ± 60 69.5 18600 ± 250 9.8 >40000 CN-06 (17) (I) 26–44 10370 ± 60 71.2 1410 ± 40 8.8 7000–9000 (II) 98–110 11310 ± 70 70.9 10300 ± 150 4.8 12000 CN-07 (18) (II) 101–113 11180 ± 60 76.3 8270 ± 70 7.0 12000 CN-08 (15) (III) 230–239 30070 ± 280 78.2 22850 ± 200 21.4 >40000 (IV) 265–283 17490 ± 100 71.2 n.d.c 6.3 S40000 CN-09 (13) (II) 94–106 11820 ± 70 77.1 9600 ± 70 0.2 12000 (III) 121–133 12620 ± 70 69.3 13500 ± 140 9.9 13000 CN-10 (10) (III) 191–203 42250 ± 1100 80.8 37900 ± 700 10.3 >40000 CN-11 (7) (I) 18–33 9530 ± 70 76.9 5030 ± 80 9.9 5100 CN-12 (4) (IV) 215–227 33930 ± 450 84.2 31000 ± 450 13.0 >40000 CN-13 (32) (III) 176–188 19150 ± 100 73.8 18670 ± 100 0.8 S40000 (IV) 230–248 17720 ± 120 69.4 20150 ± 150 6.9 >40000 CN-14 (1) (III) 180–192 16680 ± 95 72.9 18300 ± 120 3.8 >40000 (IV) 233–251 23070 ± 150 68.1 31500 ± 700 5.2 >40000 IL-01 (I) 13–37 11770 ± 120 75.6 5120 ± 90 11.6 1000–4000 (II) 60–78 n.d. n.d. 11700 ± 70 8.0 6000 (III) 100–118 22000 ± 850 71.5 15350 ± 110 10.1 8500 (IV) 136–151 35700 ± 200 38.2 21700 ± 100 12.1 S10000 IL-02 (I) 3–15 n.d. n.d. 3510 ± 40 13.0 n.d. (II) 29–41 n.d. n.d. 14900 ± 100 9.0 n.d. (III) 64–82 21300 ± 1050 75.4 16900 ± 80 17.6 6000–8000 (IV) 152–176 23050 ± 820 78.4 19100 ± 140 12.1 15000–17000 a
CN, Chianan plain; IL, Ilan plain.
b
Number in the bracket denotes the well number used byChen and Liu (2007)for the identical well.
c
n.d., not determined.
Fig. 3. Plot of d13
CDICvs. d13CCH4for groundwater from the Chianan and Ilan plains. The general ranges of d13
C values for methane generated by different pathways according to the classification scheme ofBarker and Fritz (1981)are also shown.
and14C
CH4ages with well depths can also be found (Fig. 5). In
addi-tion to the characteristics menaddi-tioned above, we make the following relevant comments.
4.1. Chianan plain
It is worth pointing out that d13C
DICvalues for about 18 of the 25
water samples from the Chianan plain are considerably greater than 0‰ (up to 9.9‰), suggesting the occurrence of Rayleigh enrichment during CO2reduction as was invoked byNissenbaum
et al. (1972)to explain unusually heavy d13C values of interstitial
water CO2. According to the Rayleigh model, successive fractions
of CO2are reduced to produce12C-enriched CH4, leaving residual
CO213C-enriched. Similar enrichment in d13C (as high as +35‰)
was reported for DIC in groundwater with dissolved methane in the United States (Scott et al., 1994; Aravena et al., 2003) and Aus-tralia (Smith and Pallasser, 1996). In contrast, the very negative value for DIC (43.5‰) in CN-03 (I) suggests that methane oxida-tion occurred.
Comparisons of14C ages between CH
4and DIC, and between DIC
and sediment deposition for each well are shown inTable 2. In view of the probability of water mixing during pumping, wide span of well screen, and the uncertainty of depositional ages,14C ages of
CH4, DIC, and strata are arbitrarily defined as significantly different
if their difference is greater than 10 times that of the larger
Fig. 4. Plot of14C ages for dissolved methane (CH
4) and inorganic carbon (DIC) from
the Chinana (CN) and Ilan (IL) plains. (I), (II), (III), (IV): number of aquifers. A positive correlation between the ages of the two components can be found.
0 50 100 150 200 250 300 0 5 10 15 20 25 30 35 40 14
CDIC age (kaBP)
well depth (m) 0 50 100 150 200 250 300 0 5 10 15 20 25 30 35 40 45 14 CCH4age (kaBP) well depth (m) CN-10(II) CN-07(II) CN-12(IV) CN-14(III) CN-14(IV) CN-02(I) CN-03(I) CN-05(II) CN-06(I) CN-05(I) CN-05(IV) CN-06(II) CN-08(III) CN-08(IV) CN-09(II) CN-09(III) CN-11(I) CN-13(III) CN-13(IV) IL-01(I) IL-01(III) IL-01(IV) IL-02(III) IL-02(IV) CN-05(III) IL-01(II) IL-02(II)
(a)
(b)
CN-04(II) CN-01(II) CN-03(III) CN-04(III) CN-04(IV) CN-03(II) CN-04(IV) CN-11(I) IL-01(III) CN-09(II) CN-09(III) CN-02(I) CN-03(I) CN-05(I) CN-06(I) IL-02(I) IL-01(I) IL-02(III) CN-03(II) CN-04(II) CN-05(II) CN-06(II) CN-07(II) CN-13(III) CN-03(III) CN-04(III) CN-05(III) CN-05(IV) CN-08(III) CN-10(II) CN-12(IV) CN-13(IV) CN-14(III) CN-14(IV) IL-01(IV) IL-02(IV) CN-01(I)Fig. 5. Plot of well depth vs.14
C age of (a) CH4and (b) DIC from the Chianan (CN)
and Ilan (IL) plains. A generally positive correlation can be found.
Table 2 Comparison of14
C ages between coexisting dissolved methane (CH4) and inorganic
carbon (DIC), and between DIC and strata from each individual well.
Aquifer Sample sitea 14
C ageb CH4 DIC Strata (I) CN-02 > = CN-03 > = CN-05 > = CN-06 > = CN-11 > = (II) CN-01 = = CN-03 = = CN-04 = = CN-05 = = CN-06 = = CN-07 > < CN-09 > < (III) CN-03 = < CN-04 = < CN-05 = < CN-08 > < CN-09 = = CN-10 = < CN-13 = < CN-14 = < (IV) CN-04 < < CN-05 < < CN-12 = < CN-13 < < CN-14 < < (I) IL-01 > > (II) IL-01 (n.d.)c > (III) IL-01 > > IL-02 > > (IV) IL-01 > > IL-02 > >
aCN, Chianan plain; IL, Ilan plain. b
Two ages are defined as equal if their difference is less than 10 times of the larger standard error (r) of the two ages. Otherwise, the greater (>) or less (<) symbol is used.
c
standard deviation of two ages. Otherwise, two ages are considered to be essentially equal. Accordingly, it is evident that the14C age of
DIC is essentially the same as that of aquifer sediments from each individual shallow well of aquifer (I) (<60 m depth) in the Chianan plain. By contrast, the14C ages of CH
4from all aquifer (I) wells are
significantly greater than those of coexisting DIC, which in turn are essentially coeval with coexisting sediments except CN-02(I) and CN-06(I), indicating the presence of CH4from the underlying
aqui-fers. It is interesting to note that seeps of older methane from underlying sediments are commonly found in some other coastal waters and shallow aquifers (e.g.Laier et al., 1996; Laier, 2003). The relatively shallow and up-gradient CN-02(I) and CN-06(I) have remarkably younger 14CDIC ages than their respective sediment
depositional ages, very probably due to mixing of younger CO2
from overlying layers.
Although CH4 ages for samples CN-07(II) and CN-09(II) are
slightly older than their respective DIC ages, all the other14C ages
of CH4, DIC and aquifer sediments for individual wells in aquifer (II)
are essentially equal. This demonstrates that dissolved methane
and DIC of aquifer (II) were formed in-situ via bacterial degrada-tion of sedimentary organic matter within the aquifer and implies that the groundwater flow rate is very slow.
Like aquifer (II), all the14C ages of CH
4from aquifer (III) (except
sample CN-08) are also essentially the same as those of their coex-isting DIC. In contrast to aquifer (II), however, both CH4and DIC
from each individual well of aquifer (III) (except CN-09) and aqui-fer (IV) are remarkably younger than their respective sediment depositional ages. Obviously, current groundwaters in aquifer (III) and (IV) were recharged much later than the deposition of their sediments. Note that the sediments of aquifers (III) and (IV) were deposited during the interval between 20 and 90 KaBP (Liu et al., 1997; Chen, 2008) when the eustatic sea-level was much lower than present and hydraulic gradients were much larger caus-ing faster discharge of groundwater. The CH4from CN-08(III) is
sig-nificantly older than its coexisting DIC and the CH4 from the
underlying CN-08 (IV).Hackley et al (1992)also noted that CH4
in their study is generally older than DIC. They interpreted this as resulting from the fact that CO2from degradation of
sedimen-tary organic matter can react to form CH4before equilibration with
DIC. We believe the same effect might have occurred in CN-08(III). Except for sample CN-12(IV), the14C ages for DIC samples from
CN-04, -05, -13, and -14 wells tapping the deepest aquifer (IV) are significantly older than those of their respective CH4(Fig. 4),
prob-ably due to dissolution of carbonate sediments. Moreover, CH4
samples from CN-05(IV) and 08(IV) are much younger than those from their respective overlying wells in aquifer (III), suggesting that some portions of aquifer (IV) have faster groundwater dis-charge than their overlying aquifer (III).
4.2. Ilan plain
The CH4of IL-01(IV) has a much older14C age than its coexisting
DIC. In addition, it is of thermogenic origin as is shown by the d13C
CH4of 38.2‰, very probably reflecting the thermocatalytic
ef-fect of shallow and young intrusive igneous rocks under the plain (Liu, 1995b; Yang et al., 2005). This interpretation is further sup-ported by the much higher CO2content of 16 vol%, as compared
with <1 vol% in other wells, although no higher hydrocarbon (e.g. ethane) was found. It is worth pointing out that the well screen of IL-01(IV) is close to the underlying metamorphic basement. Be-sides, weak-alkaline carbonic acid springs, often with CO2gas
bub-bles, are common in and around the near-by metamorphic mountain range.
Except for IL-01(IV), all CH4samples from the Ilan plain have
d13C values in the range of 71‰ to 78‰, suggesting that they originated from CO2reduction via microbial decomposition. These
microbial CH4 gases also have much older 14C ages than their
accompanying DIC, indicating a mixture of older CH4 from the
underlying strata. Moreover, both CH4and DIC of each well are
dis-tinctly older than the depositional age of sediments at depths cor-responding to each individual well screen. The most likely cause for the occurrence of much older CH4 and DIC is that the CH4
and DIC in these aquifers is mixing by CO2and CH4from the
under-lying strata. 5. Conclusions
The stable carbon isotope ratios (d13C) and 14C ages for
dis-solved methane and inorganic carbon from a total of 34 groundwa-ter samples, in addition to depositional ages of sediments, at 16 sites in the Chianan and Ilan plain of Taiwan were measured. The results show that, except for one thermogenic CH4 sample from
the Ilan plain, all other CH4samples were produced via microbially
mediated CO2 reduction. Most d13CDIC values are considerably
greater than 10‰ and even up to 10‰ due to Rayleigh enrich-ment during CO2reduction. Almost all the14C ages of CH4samples
from the shallow aquifer (I) (<60 m depth) of the Chianan plain are greater than the14C ages of coexisting DIC and sediment
deposi-tion, suggesting the presence of CH4that has migrated from the
underlying aquifers. In contrast, 14C ages of coexisting CH 4, DIC,
and deposition of sediments for aquifer (II) in the Chianan plain are essentially equal, reflecting in-situ generation of CH4and DIC
and sluggishness of the groundwater flow. Moreover, both CH4
and DIC from each individual well of aquifers (III) and (IV) in the Chianan plain are remarkably younger than the respective deposi-tional ages of these aquifers. Obviously, these groundwaters re-charged much later than the deposition of aquifer sediments. However, the age of methane depends on the age of the organic matter source as well as groundwater flow. Finally, the CH4of each
well in the Ilan plain is distinctly older than coexisting DIC, which in turn is much older than the depositional age of aquifer sedi-ments, demonstrating dissolution of much older CO2and CH4from
underlying strata. Acknowledgements
The authors thank Drs. B.M. Jahn, T. Laier and an anonymous re-viewer for valuable comments and suggestions. We are grateful to Dr. Chao-Li Liu for technical instructions and measuring stable hydrogen isotopes. Thanks are also due to the Central Geological Survey, Ministry of Economic Affairs (ROC) for providing sediment core samples and related geological information, and the Ground-water Exploitation and Conservation Center of Taiwan Sugar Com-pany for helping in water sampling. We also wish to thank Su-Chin Kang and Chun-Yen Chou for their laboratory assistance. This study is supported financially by National Science Council, Republic of China (Grant: NSC95-2116-M-002-017-MY3).
References
Aravena, R., Wassenaar, L.I., Barker, J.F., 1995. Distributuion and isotopic characterization of methane in a confined aquifer in southern Ontario, Canada. Journal of Hydrology 173, 51–70.
Aravena, R., Harrison, S.M., Barker, J.F., Abercrombie, H., Rudolph, D., 2003. Origin of methane in the Elk Valley coalfield, southeastern British Columbia, Canada. Chemical Geology 195, 219–227.
Barker, J.F., Fritz, P., 1981. The occurrence and origin of methane in some groundwater flow systems. Canadian Journal of Earth Science 18, 1802–1816. Chen, K.Y., Liu, T.K., 2007. Major factors controlling arsenic occurrence in the
groundwater and sediments of the Chianan coastal plain, SW Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 18, 975–994.
Chen, Y.W., 2008. Luminescence dating on Quaternary sediments: cases of active structures in western Taiwan. Ph.D. Thesis, Institute of Geosciences, National Taiwan University, 115p.
Coleman, D.D., Liu, C.L., Riley, K.M., 1988. Microbial methane in the shallow Paleozoic sediments and glacial deposits of Illinois. Chemical Geology 71, 23–40. Coleman, D.D., Liu, C.L., Hackley, K.C., Pelphrey, S.D., 1995. Isotopic identification of
landfill methane. Environmental Geosciences 2, 95–103.
Gooddy, D.C., Darling, W.G., 2005. The potential for methane emissions from groundwaters of the UK. Science of the Total Environment 339, 117–126. Gupta, S.K., Polach, H.A., 1985. Radiocarbon Dating Practices at ANU. Radiocarbon
Laboratory, Research School of Pacific Studies, Australian National University, Canberra, 176p.
Hackley, K.C., Liu, C.L., Coleman, D.D., 1992.14
C dating of groundwater containing microbial CH4. In: Long, A., Kra, R.S. (Eds.), Proceedings of 14th International14C
Conference, Radiocarbon, vol. 34, pp. 686–695.
Ho, C.S., 1975. An Introduction to the Geology of Taiwan: Explanatory Text of the Geologic Map of Taiwan. Ministry of Economic Affairs, Taipei, Taiwan. 153p. Hsu, L.M., 1984. Pleistocene formation with dissolved-in-water type gas in the
Chianan plain, Taiwan. Petroleum Geology of Taiwan 20, 199–213.
Lansdown, J.M., Quay, P.D., King, S.L., 1992. CH4production via CO2reduction in a
temperate bog: a source of13
C-depleted CH4. Geochimica et Cosmochimica Acta
56, 3493–3503.
Laier, T., 2003. Migration patter of methane related to glacio-tectonic deformation of marine deposits in the Kattegat-Skagerrak area. In: Proceedings of the 7th International Conference on Gas Geochemistry, pp. 44–46.
Laier, T., Kuijpers, A., Denneggard, B., Heier-Nielsen, S., 1996. Origin of shallow gas in Skagerrak and Kattegat – evidence from stable isotopic analyses and radiocarbon dating. Bulletin of Geology Survey of Norway 430, 129–136. Liu, T.K., 1995a. Estimating flow and recharge rates of groundwater in western
Taiwan using radiocarbon and tritium. Radiocarbon 37, 531–542.
Liu, C.C., 1995b. The Ilan plain and the southwestward extending Okinawa Trough. Journal of Geological Society of China 38, 229–242.
Liu, T.K., Sung, Q.C., Chen, K.Y., Pi, Z.L., Yang, C.H., Cheng, P.H., 1997. Tectonic subsidence and uplift in the Zaikang-Hopi area of south-western Taiwan since the Late Pleistocene. Journal of Geological Society of China 40, 155–165. Lu, F.J., 1990. Fluorescent humic substances and blackfoot disease in Taiwan.
Applied Organometallic Chemistry 4, 191–195.
Nissenbaum, A., Presley, B.J., Kaplan, I.R., 1972. Early diagenesis in a reducing fjord, Saanich Inlet, British Columbia, I. Chemical and isotopic changes in major components of interstitial water. Geochimica et Cosmochimica Acta 36, 1007– 1027.
Oremland, R.S., Marsh, L.M., Polcin, S., 1982. Mathane production and simultaneous sulfate reduction in anoxic salt marsh sediment. Nature 296, 143–145. Oremland, R.S., 1988. Biogeochemistry of methanogenic bacteria. In: Zehnder, A.J.B.
(Ed.), Biology of Anaerobic Microorganism. Wiley Interscience, pp. 641–706. Schoell, M., 1980. The hydrogen and carbon isotopic compositions of methane from
natural gases of various origins. Geochimica et Cosmochimica Acta 44, 649– 661.
Schoell, M., 1988. Multiple origins of methane in the earth. Chemica Geology 71, 1–10.
Scott, A.R., Kaiser, W.R., Ayers Jr., W.B., 1994. Thermogenic and secondary biogenic gases, San Juan basin, Colorado and New Mexico – implications for coalbed gas producibility. American Association of Petroleum Geologists Bulletin 78, 1186– 1209.
Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behavior and distribution of arsenic in natural waters. Applied Geochemistry 17, 517–568. Smith, J.W., Pallasser, R.J., 1996. Microbial origin of Australian coalbed methane.
American Association of Petroleum Geologists Bulletin 80, 891–897. Stuiver, M., Polach, H.A., 1977. Discussion: Reporting of14
C data. Radiocarbon 19, 355–363.
Stumm, W., Morgan, J.J., 1996. Aquatic Chemistry, third ed. Wiley-Interscience, 1022p.
Tseng, W.P., Chen, W.Y., Sung, J.L., Chen, J.S., 1961. A clinical study of Blackfoot disease in Taiwan, an endemic peripheral vascular disease. Memoir of College of Medicine, National Taiwan University 7, 1–18.
Tyler, S.C., 1991. The global methane budget. In: Rodgers, J.E., Whitman, W.B. (Eds.), Microbial Production and Consumption of Greenhouse Gases: Methane, Nitrogen oxides, and Halomethanes. American Society of Microbiology, Washington DC, pp. 7–38.
Whiticar, M.J., Faber, E., Schoell, M., 1986. Biogenic methane formation in marine and freshwater environments: CO2reduction vs. acetate fermentation – isotope
evidence. Geochimica et Cosmochimica Acta 50, 693–709.
Whiticar, M.J., 1999. Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chemical Geology 161, 291–314. Yang, T.Y., Lan, T.F., Lee, H.F., Fu, C.C., Chuang, P.C., Lo, C.H., Chen, C.H., Chen, C.T.,
Lee, C.S., 2005. Gas compositions and helium isotopic ratios of fluid samples around Kueishantao, NE offshore Taiwan and its tectonic implications. Geochemical Journal 39, 469–480.