Palaeo-hydrogeological control on groundwater As levels in Red River
delta, Vietnam
Søren Jessen a,*, Flemming Larsen b, Dieke Postma b, Pham Hung Viet c, Nguyen Thi Ha d, Pham Quy Nhan e, Dang Duc Nhan f, Mai Thanh Duc c, Nguyen Thi Minh Hue c, Trieu Duc Huy d, Tran Thi Luu c, Dang Hoang Ha e, Rasmus Jakobsen a
a Depart ment of Envi ron men tal Engi neer ing, Tech ni cal Uni ver sity of Den mark (TUD), 2800 Kgs. Lyn gby, Den markb National Geo log i cal Sur vey of Den mark and Green land (GEUS), Den markc Research Cen tre for Envi ron men tal Tech nol ogy and Sus tain able Devel op ment (CET ASD), Ha noi Uni ver sity of Sci ence, Viet Namd Viet nam North ern Hydro geo log i cal and Engi neer ing Geo log i cal Divi sion (NHEGD), Viet Name Ha noi Uni ver sity of Min ing and Geol ogy (HUMG), Viet Namf Insti tute for Nuclear Sci ence and Tech nol ogy, Viet Nam
a r t i c l e i n f o a b s t r a c t
Article histry:
Available online 4 July 2008
To study the geo log i cal con trol on ground wa ter As con cen tra tions in Red River delta,
depth-spe cific ground wa ter sam pling and geo phys i cal log ging in 11 mon i tor ing wells
was con ducted along a 45 km tran sect across the south ern and cen tral part of the delta,
and the lit er a ture on the Red River delta’s Qua ter nary geo log i cal devel op ment was
reviewed. The water sam ples (n = 30) were ana lyzed for As, major ions, Fe2+, H2S, NH4,
CH4, d18O and dD, and the geo phys i cal log suite included nat u ral gamma-ray, for ma-
tion and fluid elec tri cal con duc tiv ity. The SW part of the tran sect inter sects depos its of
grey estu a rine clays and del taic sands in a 15–20 km wide and 50–60 m deep Holo cene
incised val ley. The NE part of the tran sect con sists of 60–120 m of Pleis to cene yel low ish
allu vial depos its under neath 10–30 m of estu a rine clay over lain by a 10–20 m veneer
of Holo cene sed i ments. The dis tri bu tion of d18O-val ues (range ¡12.2‰ to ¡6.3‰) and
hydrau lic head in the sam ple wells indi cate that the estu a rine clay units divide the flow
sys tem into an upper Holo cene aqui fer and a lower Pleis to cene aqui fer. The ground wa-
ter sam ples were all anoxic, and con tained Fe2+ (0.03–2.0 mM), Mn (0.7–320 lM), SO4
S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126 3117
1. Intro duc tion
Ele vated con cen tra tions of ge o gen ic As in ground wa ter
poses a threat to the health of tens of mil lions of people
living in the large delta areas of South east Asia. In the Red
River delta, Viet nam, an esti mated 11 mil lion people are
at risk (Berg et al., 2001). Although sev eral pro cesses lead-
ing to the release of As have been pro posed, the reduc tion
of As-con tain ing Fe-oxides with nat u ral organic mat ter
is gen er ally con sid ered the most impor tant mobi li za tion
mech a nism (McAr thur et al., 2001; Akai et al., 2004;
Islam et al., 2004; Post ma et al., 2007). On a local scale,
the ground wa ter As con cen tra tion often shows a patchy
dis tri bu tion (Har vey et al., 2005; Char let and Polya, 2006)
prob a bly deter mined by the local hydro ge ol ogy and/or
vari a tions in abstrac tion depth (Smed ley and Kinni burgh,
2002; Har vey et al., 2005). How ever, on a larger scale,
regional sur veys in the Ben gal and Me kong del tas indi-
cate the exis tence of areas in which new wells are likely
to pro duce low-As water (McAr thur et al., 2001; Berg et
al., 2007). Regional sur veys are costly and time con sum-
ing and a pre dic tion of the ground wa ter As con tent based
on exist ing data would there fore be pref er a ble. Recently
maps that pre dict the As con cen tra tion in the ground wa ter
based on geo log i cal and sur face-soil param e ters have been
val i dated with rea son able suc cess against sur vey data sets
from the Ben gal, Me kong, Red River, Myan mar and Suma-
tra delta areas (Polya et al., 2005; Hoss ain et al., 2007;
Berg et al., 2007; Win kel et al., 2008; Rodri guez-Lado et
al., 2008) and also global ground wa ter As pre dic tion mod-
el ling has been con ducted (Amini et al., 2008). For the
Red River delta, a val i da tion and refine ment of pre dic tion
maps should be pos si ble, because thousands of ground-
wa ter sam ples have been ana lyzed in two recent regional
sur veys by UNICEF (Bad loe et al., 2004) and the Swiss Fed-
eral Insti tute of Aquatic Sci ence and Tech nol ogy, EA WAG,
in coop er a tion with CET ASD (Unpub lished data, Michael
Berg, pers. comm.). The EA WAG/CET ASD sur vey shows a
high ground wa ter As con cen tra tion in a 20 km wide band
along the NW–SE bound ary of the delta plain, par al lel to
the position of the pal ae o-Red River main chan nel (Fig.
1A), while ground wa ters in the cen tral and north ern delta
plain gen er ally have low lev els of As. These results are
con sis tent with the results of the UNICEF sur vey, though
the lat ter are reported per prov ince. In the Ben gal delta, a
high ground wa ter As con cen tra tion is found in Holo cene
aqui fers, while Pleis to cene aqui fers have a low-As level
(Rav ens croft et al., 2001, 2005). The South east Asian del tas
all derive their sed i ments ulti mately from the Hima la yas
(Stan ger, 2005; Char let and Polya, 2006; Guil lot and Char-
let, 2007), and depo si tion takes place in Ceno zoic sub si-
dence basins under the influ ence of Qua ter nary eu stat ic
sea level changes (Tan a be et al., 2006). The sub si dence
rate in the Ben gal delta (Good bred and Ku ehl, 2000) has
been much higher than that in the Red River delta, and in
the onshore Red River delta the Qua ter nary sequence is
only up to 200 m in total (Ma thers and Zal asiewicz, 1999).
Dur ing the mid-Holo cene trans gres sion the sea cov ered
a large part of the Red River delta (Tan a be et al., 2006).
This trans gres sion, com bined with the mod est total thick-
ness of the Qua ter nary aqui fers, influ ences the pres ent day
ground wa ter chem is try in the Red River delta.
In this study ground wa ter sam ples have been col-
lected from mon i tor ing wells along a 45 km tran sect run-
ning per pen dic u lar to the Red River, from the south ern
delta bound ary to the cen tral part of the delta (Fig. 1A
and B). The aims of the pres ent study are (i) to pro vide
a first assess ment of the over all dis tri bu tion of Pleis to-
cene and Holo cene sed i ments in the Red River delta, (ii)
to inves ti gate if the ground wa ter As lev els in the Red
River delta are related to geo log i cal age of the aqui fer
sed i ments, and (iii) to inves ti gate the con trols on the
ground wa ter As con cen tra tion in the brack ish-marine
sed i ments of the Red River delta.
Fig. 1. (A) The Red River delta, pres ent day sit u a tion and pal ae og e og ra phy. A–A9 indi cates the position of the stud ied tran sect. (B) The position of bore holes
for water sam pling (Q82–Q131) and sed i ment descrip tions along the tran sect.
3118 S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126
2. Meth ods
2.1. Field cam paign
The field work was car ried out from 27 May to 5 June
2006 and sam pling sites included 11 of the Viet nam ese
National Mon i tor ing Net work wells (Fig. 1B). The wells
are nests with up to three sep a rate holes. Typ i cally, the
bore holes are equipped with OD 120 mm PVC cas ings and
screens; the upper one or two screens are 6 m long while
the deep est screen is 8–10 m long.
2.2. Bore hole log ging and water table mea sure ments
Geo phys i cal log ging was car ried out in the deep est
bore hole at each mon i tor ing nest. The log suite included
nat u ral gamma-ray, for ma tion elec tri cal con duc tiv ity
(focused induc tion), and fluid tem per a ture and con duc tiv-
ity (Rob ert son Geo log ging). The gamma-ray log is a proxy
for sed i men tary clay con tent, while the for ma tion con-
duc tiv ity log is a proxy for the salin ity of the for ma tion
pore water. To trans late for ma tion con duc tiv ity log-val ues
into esti mates of pore water salin ity, the for ma tion fac tor
Ff = rw/rf was used, where rw and rf are the con duc tiv i ties
of the water sam ple (mea sured by sam pling) and of the
for ma tion (mea sured by log ging), respec tively. Thus, after
esti mat ing the for ma tion fac tor for a given sed i men tary
unit in the tran sect, the pore water salin ity expressed as
rw can be cal cu lated from the for ma tion con duc tiv ity log
response.
To assess the aqui fer hydro dy nam ics the rel a tive
static water lev els in the bore holes at each loca tion were
recorded and cor rected for den sity vari a tions using fluid
con duc tiv ity mea sure ments to obtain com pa ra ble heads.
2.3. Water sam pling and field anal y sis
To ensure that the water sam ples orig i nated from the
screened inter vals of the bore holes the sam ples were
col lected by a low-flow-rate Whale-pump (0.5–2 L/min)
posi tioned at the top of the screen, and below a high flow-
rate Grund fos MP1-pump (typ i cally, 10–20 L/min). In two
of the sam pled screens, two sam ples (as opposed to one
sam ple) were col lected from sep a rate in-flow zones iden-
ti fied by the geo phys i cal log ging. Before sam pling, 3–12
bore hole vol umes were flushed. Dis solved O2, pH, tem per-
a ture and elec tri cal con duc tiv ity was mea sured by WTW
elec trodes in a flow cell and mon i tored through the flush-
ing to ensure sta ble val ues before sam pling. Fer rous iron
(Fe2+), PO4 and H2S were mea sured on a HACH DR/2010
spec tro pho tom e ter in the field, using, respec tively, the
Fer ro zine (Stoo key, 1970), molyb date blue and meth y lene
blue meth ods (Cline, 1967). The Fe2+ con cen tra tions mea-
sured in the field closely matched total Fe mea sured in the
lab o ra tory by flame atomic absorp tion spec tro pho tom e try
(slope 1.00; r2 = 0.99). Alka lin ity was mea sured in the field
by Gran-titra tion (Stumm and Mor gan, 1981). Sam ples
were col lected by poly eth yl ene tub ing and syrin ges, and
fil tered through 0.20 lm cel lu lose ace tate syrin ges-fil ters
(Sar to rius Min is art) to poly eth yl ene vials. Prior to sam-
pling, syrin ges and fil ters were pre-flushed with N2-gas
to pre vent As seques tra tion by Fe oxide pre cip i ta tion. Sam-
ples for CH4 were injected into pre-weighed evac u ated
glass vials equipped with a pierce able sep tum (ex e tain ers;
Lab co, ord.co. 819W), through a syringe nee dle mounted
on the sam pling tube. The ex e tain ers were stored upside-
down to trap the CH4 in the head space vol ume of 2–3 mL.
Sam ples for anions, NH4 and CH4 were put on ice in the
field and fro zen later on the day of sam pling. Sam ples for
all other param e ters were acid i fied by add ing 2 vol% of a
7 M HNO3 solu tion, then put on ice and stored refrig er ated.
Detec tion lim its for Fe2+, H2S and PO4 were ca. 0.1 lM.
2.4. Lab o ra tory anal y sis
Cat ions were ana lyzed by flame atomic absorp tion
spec tro pho tom e try on a Shi ma dzu AAS 6800 instru ment.
Aque ous As was deter mined on the same instru ment using
a HVG hydride gen er a tor and a graph ite fur nace. Anions
were ana lyzed by ion chro ma tog ra phy using a Shi ma dzu
LC20AD/HIC-20ASu per. Ammo nium was deter mined by
spec tro pho tom e try using nitro prus side. Meth ane head
space con cen tra tions were deter mined by gas chro ma tog-
ra phy using a Shi ma dzu GC-14A with a 1 m packed col umn
(3% SP1500, Car bo pack B) and a FID detec tor. The aque ous
CH4 con cen tra tion was cal cu lated using Henry’s law. Detec-
tion lim its were: As 0.013 lM; Mn 0.91 lM; Ca 0.50 lM;
NH4 5.6 lM; NO3 3.2 lM; SO4 2.1 lM and CH4 0.01 mM.
The sta ble iso tope ratio of O (18O/16O) and H (2H/1H)
of the water rel a tive to the VSMOW stan dard was ana-
lyzed using a Mi cro Mass spec trom e ter (Iso Prime, GV
Instru ments, UK) equipped with an Eu ro vec tor ele men tal
ana lyzer (Eu roEA 3000, Italy) and the Mass lynx Pro gram
(GV Instru ments, UK) for data pro cess ing. The results are
expressed in ‰ units using the d-nota tion with a stan dard
devi a tion not larger than ±0.20‰ (for d18O), as cal cu lated
from a min i mum of five rep li cate injec tions into the ele-
men tal ana lyzer.
2.5. Spe ci a tion cal cu la tions
Aque ous spe ci a tion was done using PHRE EQC-2
(Park hurst and Ap pe lo, 1999) with the inclu sion of the
ther mo dy namic data base for the As spe cies pro vided by
Lang muir et al. (2006). For Fe sul phide equi lib rium cal cu la-
tions the FeS(ppt) phase defined in the data base for which
FeS M Fe2+ + S2¡, K = 10¡16.833 was used.
3. Results
3.1. Geo log i cal set ting
The cross sec tion shown in Fig. 2 is based on a sequence
strati graph i cal inter pre ta tion of the avail able lith o log i cal
and geo phys i cal logs, con sis tent with the work pub lished by
Su sumu Tan a be, his co-work ers and oth ers (Tran et al., 1991,
2002; Ma thers and Zal asiewicz, 1999; Lam and Boyd, 2003;
Tan a be et al., 2003a, b, 2006; Hori et al., 2004; Hane buth et
al., 2006; Li et al., 2006; Funa bi ki et al., 2007). Neo gene bed-
rock forms the base of the Qua ter nary depos its, which thick-
ens north ward from about 30 m to 150 m. The Qua ter nary
sequence con sists of Pleis to cene allu vial sand and gravel
S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126 3119
depos its, over lain by estu a rine clays and Holo cene del taic
sands. On the top there are depos its of silt and clay in Holo-
cene marine ter races and, near the mod ern chan nels such as
the Red River, over bank depos its, typ i cally form ing a 5–10 m
thick con fin ing clay layer. The cross sec tion also dis plays 14C sed i ment dates from the DT hole (Tan a be et al., 2003a),
and the HH120 hole and two addi tional loca tions (Lam and
Boyd, 2003; Funa bi ki et al., 2007). Fault loca tions from Tran
et al. (2002) are indi cated on Figs. 1B and 2. The dashed line
in Fig. 2, at ele va tion around ¡50 m in the SW part of the
cross sec tion and around ¡10 m in the NE part of the cross
sec tion, indi cates the pre sumed topo graphic sur face dur ing
the Late Pleis to cene sea level low, shown in Fig. 3 by the sea
level curve for the Late Pleis to cene–Holo cene. Fig. 3 shows,
that dur ing the mid dle of the Weich sel gla ci a tion, 80–30 ka
before pres ent (BP), the sea level oscil lated at around 80 m
below the pres ent sea level (Kit az a wa, 2007), and 25–19 ka
BP, the sea level dropped to 120 m below the pres ent sea
level, as global cool ing at the time caused sea wa ter to accu-
mu late in ter res trial ice-sheets (Yo koy ama et al., 2000; Lam-
beck et al., 2002; Tan a be et al., 2006). Dur ing the sea level
low, the pal ae o-Red River (Fig. 1A) eroded a val ley into the
Pleis to cene sed i ments in the SW part of the cross sec tion. In
Fig. 1A, the incised val ley along the NW–SE delta bound ary
is out lined by the shore line 9 ka BP (Tan a be et al., 2006).
From 6 to 4 ka BP the sea level stood 2–4 m above pres ent sea
level (Boyd and Lam, 2004; Tan a be et al., 2006) evi dent in
the cross sec tion as the marine ter races (Figs. 1A and 2).
Three of the geo phys i cal bore hole logs recorded dur-
ing the sur vey are depicted in Fig. 2, rep re sent ing pri mar-
ily Holo cene depos its in Q86 and Pleis to cene depos its in
Q130 and Q131. The Pleis to cene allu vial sed i ments in the
bore hole logs show rel a tively low gamma-ray lev els of
around 75 API, cor re spond ing to rel a tively coarse grained
sed i ments. The grad ual upward increase in the gamma-ray
lev els, e.g., from ele va tion ¡68 m to ¡62 m and again from
¡62 m to ¡39 m in Q130, and from ¡84 m to ¡60 m in Q131
indi cate fin ing-upward sequences of the allu vial, higher-
energy depo si tional envi ron ment. High-stand flu vial sed-
i ments have smaller-scale fin ing-upward sequences (Gani
Fig. 3. Com pi la tion of sea level curves (ele va tion in meters rel a tive to pres-
ent sea level) from Tan a be et al. (2006) (0–20 ka BP) and Kit az a wa (2007)
(20–150 ka BP).
Fig. 2. Geo log i cal cross sec tion of the sam pled tran sect (A–A9 in Fig. 1). The Late Pleis to cene topo graphic sur face is indi cated by a black dashed line. Geo-
phys i cal logs show, to the left, the nat u ral gamma-ray, and to the right, the for ma tion con duc tiv ity.
3120 S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126
and Alam, 2004). The two estu a rine sed i ment units (Holo-
cene in the SW, and Pleis to cene in the NE part of the tran-
sect) are clay-rich, con sis tent with the ele vated gamma-
ray lev els of up to 140–150 API. The gamma-ray lev els in
the Holo cene sand depos its are rel a tively high, 110–120
API, con firm ing the silty-clayey con di tions described in
the lith o log i cal logs. The lithol ogy of the sed i ments will be
described fur ther in Sec tion 4.1.
3.2. Cross sec tion hydrol ogy
Fig. 4A shows the dis tri bu tion of d18O in the water
sam ples col lected by the depth-spe cific sam pling in the
11 mon i tor ing wells (Fig. 1B). The value of d18O ranges
from ¡12.2 to ¡6.3‰, and rel a tively 18O-rich ground wa-
ter is gen er ally found in the Holo cene depos its while the
ground wa ter in the Pleis to cene depos its is more depleted
in 18O with d18O-val ues lower than ¡8.6‰. The ver ti cal
dis tri bu tion of hydrau lic head (not shown) gen er ally indi-
cated a down ward migra tion of water. In the NE part of
the tran sect (dis tance 20–50 km), a ver ti cal hydrau lic head
dif fer ence of 2.4–2.7 m water col umn was observed across
the estu a rine depos its, while a less pro nounced ver ti cal
hydrau lic head dif fer ence (<0.8 m) was observed in the
SW part of the tran sect (dis tance 0–20 km). The hydrau lic
head dif fer ences trans late into ver ti cal gra di ents of +8 to
¡146‰, where the neg a tive val ues infer down ward flow.
The dis tri bu tion of d18O and ver ti cal hydrau lic heads indi-
cate that the estu a rine clay depos its (Fig. 2) divide the
sequence into an upper and a lower aqui fer. No attempt
was made to deduce any hor i zon tal gra di ents along the
tran sect, as the ele va tion of ref er ence points was not ver-
i fied.
3.3. Ground wa ter types
The dis tri bu tion of the ground wa ter Cl con cen tra tion
is shown in Fig. 4B. The Cl con cen tra tion ranges from 0.19
to 65 mM, and is gen er ally higher in the lower aqui fer. In
Fig. 4C each water sam ple is rep re sented by a Stiff dia-
gram. The fig ure also includes exam ples of Stiff dia grams
for intru sion (saline water dis plac ing fresh wa ter), fresh-
en ing (fresh wa ter dis plac ing saline water), fresh wa ter and
a 10% oce anic sea wa ter. Fresh- or fresh en ing waters of,
respec tively, Ca–HCO3 or Na–HCO3 water types, are dom-
i nant in the Holo cene aqui fer. Fresh wa ter is also found in
Fig. 4. (A) The dis tri bu tion of d18O in the tran sect. The d18O-val ues in the Pleis to cene aqui fer are depleted rel a tive to those in the Holo cene aqui fer. (B)
The dis tri bu tion of Cl in the ground wa ter. The area of the dots in this and con sec u tive fig ures vary lin e arly with the param e ter value (square root scal ing).
(C) Water types in the sam pled tran sect, as indi cated by Stiff dia grams. For com par i son four exam ples of Stiff dia grams are shown below dia gram C to
exem plify the three dif fer ent ground wa ter types and 10% oce anic sea wa ter.
S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126 3121
the Pleis to cene aqui fer in Q87, for which the bore hole logs
(not shown) describe a rel a tively sandy sequence, allow-
ing fresh wa ter recharge to the lower aqui fer. The elec tri-
cal con duc tiv ity and the Cl con cen tra tion of the fresh wa-
ter are, respec tively, 60–120 mS/m and 0.19–3.03 mM (Cl
median: 1.73). Intru sive waters are dom i nant in the Pleis-
to cene aqui fer and are of a Na–Cl water type, which, due to
cat ion exchange for Ca2+, are depleted in Na+ rel a tive to the
Na/Cl ratio of sea wa ter. The elec tri cal con duc tiv ity of the
intru sive water is as high as 600 mS/m, or 11% that of oce-
anic sea wa ter. Com pared to this, the geo phys i cal logs (Fig.
2) sug gest a for ma tion con duc tiv ity of up to 400 mS/m,
which indi cates a pore water salin ity of 600–1000 mS/m
(Ff = 1.5–2.5), cor re spond ing to 11–19% of the salin ity of
oce anic sea wa ter. Because pres ent day tide-induced salin-
ity intru sion in sur face waters extend only 20–30 km
inland from the coast (Vu, 1996), the high salin ity found in
the tran sect (Fig. 4B) must be due to the pres ence of stag-
nant saline water in the estu a rine clay units or, espe cially
for the upper aqui fer, from the mid-Holo cene marine sed-
i ments. The for ma tion con duc tiv ity is high est in the mid-
dle of the estu a rine clay units (Fig. 2) and low ers towards
Fig. 5. The dis tri bu tion of ground wa ter redox com po nents in the tran sect. ‘BD’ indi cates a con cen tra tion below the limit of detec tion.
3122 S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126
the top and bot tom of the units, which could indi cate that
flush ing of the stag nant saline water is con trolled by dif-
fu sion.
3.4. Ground wa ter redox chem is try
The dis tri bu tion of the redox-sen si tive param e ters,
Fe2+, SO4, H2S, NH4 and CH4, is shown in Fig. 5. All sam ples
have a detect able Fe2+ con cen tra tion (Fig. 5A) imply ing
the pres ence of reduced con di tions. Con sis tently, the dis-
solved O2 con cen tra tion was always below the 0.016 mM
detec tion limit The high est Fe2+ con cen tra tions of 0.50–
0.95 mM (Q86), 2.0 mM (Q88) and 0.56 mM (Q131) are
found in the Pleis to cene aqui fer. The Mn con cen tra tion
(not shown) ranged from 0.7 to 320 lM and was above the
WHO guide line of 7.3 lM (0.4 mg/L) in 70% of the sam ples.
The SO4 con cen tra tion (Fig. 5B) is up to 0.75 mM, being
high est in the NE part of the tran sect, where SO4 is found
both in the upper aqui fer and in the Pleis to cene aqui fer. In
the SW end of the tran sect, SO4 is found in a con cen tra tion of
up to 0.05 mM, mainly in shal low wells and at depth in Q84.
Sul phide was detected in the ground wa ter (Fig. 5C)
in con cen tra tions of up to 7.0 lM, indi cat ing ongo ing
SO4 reduc tion. The dis tri bu tion of sul phide appears scat-
tered, though the high est lev els are pres ent in the Holo-
cene aqui fer.
High NH4 con cen tra tions (Fig. 5D) of typ i cally 1–2 mM,
and up to 4.4 mM, are found in the SW part of the tran sect,
except for many of the shal low screens, which have con cen-
tra tions below 0.2 mM. In the NE part of the tran sect, the
NH4 con cen tra tion is below 0.7 mM.
The CH4 con cen tra tion (Fig. 5E) shows a dis tri bu tion
sim i lar to that of NH4. In the SW part of the tran sect the
CH4 con cen tra tion typ i cally ranges from 1 to 3 mM, except
from some screens placed in the Holo cene aqui fer. A very
high CH4 con cen tra tion of 14.5 mM was found in the deep-
est screen of Q85. Lower lev els of CH4 are found in the NE
part of the tran sect.
3.5. The dis tri bu tion of As in ground wa ter
The dis tri bu tion of total As in the ground wa ter is
shown in Fig. 6. In the SW part of the tran sect the As con-
cen tra tion ranges from below the 0.013 lM detec tion limit
(1 lg/L) to nearly 12 lM (900 lg/L) in Q128. The As con cen-
tra tion shows a large spa tial var i a tion, although the high-
est val ues are found close to the Red River. In the NE end of
the tran sect the As con cen tra tion is low, the high est being
0.32 lM.
4. Dis cus sion
4.1. Lithol ogy of the sed i ments
Dur ing the rapid eu stat ic sea level rise around 15 ka BP
(Fig. 3) the incised val ley became filled with suc ces sively
estu a rine dark-grey silt and clay, over lain by del taic dark-
grey, in places described as green ish, silty, clayey sand
and fine sand. In the NE part of the cross sec tion in Fig.
2, Pleis to cene depos its com prise most of the Qua ter nary
sequence, while Holo cene depos its only form a thin super-
fi cial aqui fer of up to 10 m in thick ness. This shal low Holo-
cene aqui fer is char ac ter ised by silty and clayey sed i ments
depos ited at a low sed i men ta tion rate (Lam and Boyd,
2003; Funa bi ki et al., 2007). The Pleis to cene estu a rine
unit, with a top ele va tion of around ¡10 m (Fig. 2), con-
sists of lat er itic, grey-yel low or spot ted silt and clay. Other
stud ies describe the sequence bound ary for the Holo cene
sed i ments as a sub aer i ally weath ered, lat er itic marker hori-
zon, dated to the time of the sea level low (Ma thers and
Zal asiewicz, 1999; Tran et al., 2002; Funa bi ki et al., 2007)
or ear lier in Pleis to cene (Hane buth et al., 2006). The Pleis-
to cene aqui fer con sists of thick depos its of sand, gran ules
and peb bles, which in parts of the more detailed lith o log i-
cal descrip tions of bore holes LK20 and LK25 (Figs. 1B and
2) are described also as grey-yel low and brown col oured,
in both bore holes to ele va tions as deep as ¡105 m. The sed-
i ments under neath the incised val ley in the SW part of the
cross sec tion (ele va tion ¡50 to ¡70 m), how ever, are typ i-
cally described as grey.
4.2. Redox envi ron ment and sed i ment age
The con cen tra tions of NH4 and CH4 in and beneath the
Holo cene incised val ley in the SW part of the tran sect are
mark edly higher than in the NE end of the tran sect (Fig. 5D
and E). The dis tri bu tion of NH4 and CH4 over depth sug-
gests that they are gen er ated by the deg ra da tion of sed-
i men tary organic mat ter within the aqui fer, rather than
being derived from organic C infil trat ing from the sur face.
Fig. 6. The dis tri bu tion of the ground wa ter As con cen tra tion in the tran sect. The dot-size of Q128 was set to cor re spond to 4 lM in order to improve res o-
lu tion in the plot. ‘BD’ indi cates a con cen tra tion below the limit of detec tion.
S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126 3123
The high con cen tra tions of NH4 and CH4 in the SW part
of the tran sect indi cate that either the abun dance or the
reac tiv ity of the sed i men tary organic mat ter is higher than
in the Pleis to cene depos its in the NE end of the tran sect.
While a reduced reac tiv ity of the organic mat ter would
be expected in the older sed i ments (c.f., Row land et al.,
2007), the organic mat ter con tent is also likely to be lower
because the sed i ments were depos ited in a higher-energy,
allu vial regime (Ma thers and Zal asiewicz, 1999).
In Fig. 7, the ion activ ity prod ucts (IAPs) for dis or dered
mack i naw ite (FeS) and sid er ite (FeCO3) are plot ted vs. depth.
The dashed lines in Fig. 7 indi cate the equi lib rium con stants
for these two phases. The sam ples plot close to the equi lib-
rium line for dis or dered mack i naw ite and gen er ally one to
two log units to the right of the equi lib rium line for sid er ite,
thus mak ing the pre cip i ta tion of sid er ite fea si ble (Post ma et
al., 2007). The results in Fig. 7 indi cate that the pre cip i ta tion
of Fe sulp hides and sid er ite con trols the Fe2+ con cen tra tion
in the tran sect. Both sid er ite and amor phous FeS, the lat ter
being a pre cur sor for pyrite, has been detected in Ben gal
aqui fer sed i ments (Ahmed et al., 2004; Akai et al., 2004;
Seng upta et al., 2004; Low ers et al., 2007).
4.3. As in the ground wa ter
The dis tri bu tion of ground wa ter As in the tran sect is
related to the dis tri bu tion of redox spe cies, espe cially SO4,
NH4 and CH4. A low-As con cen tra tion is found in the NE
part of the tran sect in both the Holo cene and the Pleis to-
cene aqui fer (Fig. 6), while a high As con cen tra tion is found
in the more reduced Holo cene val ley fill. The high est con-
cen tra tion of ground wa ter As is found in the pre sum ably
youn gest aqui fer sed i ments close to the Red River (Fig. 6).
The lateral As dis tri bu tion observed in the rather lim ited
data set is, how ever, sup ported by the regional As dis tri bu-
tion in the delta (Michael Berg, pers. comm.; Bad loe et al.,
2004). Thus, it appears that ground wa ter As lev els in the
Red River delta are linked to the age of the sed i ments.
A hydro dy namic con trol on the As dis tri bu tion is
inferred for Q87 (dis tance 12.4 km) in which dis solved
As is found at depth where fresh wa ter migrates from the
Holo cene to the Pleis to cene aqui fer.
An excep tion to the above described rela tion ship
between sed i ment age, hydro dy nam ics and As dis tri bu-
tion, is the high con cen tra tion of As found in bore hole Q85,
61 m below sur face (Fig. 6). This sam ple has an extremely
high CH4 con cen tra tion of 14.5 mM (232 mg/L) (Fig. 5E) indi-
cat ing that excep tional hy drog eo chem i cal pro cesses occur.
In an attempt to char ac ter ize the organic mat ter sources
respon si ble for As release to ground wa ters in the Ben gal
delta, Row land et al. (2006) found bio de grad able nat u-
ral petro leum-derived hydro car bons in their sed i ments.
Onshore hydro car bon seep age from Ter tiary out crops in
the Red River delta plain has been reported (Tray nor and
Sla den, 1997; Pet er sen et al., 2001, 2005), indi cat ing that
hydro car bons could be a local organic mat ter source in
deep parts of the Red River delta’s ground wa ter envi ron-
ment.
The ele vated ground wa ter salin ity observed in the
stud ied tran sect (Fig. 4B), indi cates that in the Red
River delta saline ground wa ter occurs far inland from
the coast and ren ders some of the ground wa ter that is
low in As to be unsuited for domes tic use. In the Ben-
gal delta, mod ern intru sion causes a high ground wa ter
salin ity mainly in the coastal region, while saline water
far from the coast is restricted to local pock ets (Ahmed
et al., 2004).
4.4. Pal ae o-hydrol ogy in the Red River delta
The course of the Red River dur ing the sea level low,
and at pres ent, appears con trolled by the Nam Dinh and
Chay fault struc tures (Fig. 1B) (Tran et al., 2002) and, sim-
i larly, other sur face water chan nels in the study area are
aligned along fault struc tures. The cen tre of sub si dence in
the delta has shifted towards the SW from a pre-Holo cene
loca tion north of the Vinh Ninh fault (Tran et al., 2002). To
the north, the tran sect extends into the study area of Lam
and Boyd (2003) (Fig. 1A), who observed a rel a tively high
ele va tion (>¡36 m) of the Pleis to cene marker hori zon in
the north-east ern delta plain. The few avail able sed i ment 14C dates from below the marker hori zon (Figs. 1B and 2)
sig nifi cantly pre-date the Holo cene (Lam and Boyd, 2003).
These obser va tions sug gest a rel a tively sta ble position of
the main chan nel in the south ern part of the delta dur ing
Late Pleis to cene–Holo cene. Tec tonic char ac ter is tics appear
to have pre vented the Red River from mean der ing over the
delta plain dur ing the last part of the sea level low, and
has thereby per mit ted age ing of the depos its in the cen tral
and north ern delta region.
Rav ens croft et al. (2005) pro posed for their Ben gal
delta study area, that the 120 m deep inci sion of the pal-
ae o-riv ers dur ing the sea level low caused steep hydrau lic
gra di ents along with the devel op ment of a thick unsat u-
rated zone in the Pleis to cene sed i ments. Steep gra di ents
will increase the ground wa ter flow veloc ity and thereby
Fig. 7. The ion activ ity prod ucts (IAPs) for sid er ite (FeCO3) and dis or dered
mack i naw ite (FeS) vs. depth. PHRE EQC-2 was used for the spe ci a tion cal-
cu la tions.
3124 S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126
pro mote the flush ing of labile As from the sed i ment. An
enhanced flux of oxy gen ated water and the unsat u rated
con di tions will also pro mote oxi da tion of sed i men tary
organic mat ter and immo bi lize As in re crys tal lized, more
sta bile Fe-oxides. There fore, in the Ben gal delta the oxi-
dized Pleis to cene sed i ments has a smaller As con tent (and
a smaller oxa late extract able pro por tion of the total As)
than the Holo cene sed i ments (Ahmed et al., 2004; Swartz
et al., 2004; Stol len werk et al., 2007). How ever, in some
cases a high As con tent in Pleis to cene sed i ments has also
been reported (Shah, 2008). Because of the sed i men to log i-
cal sim i lar i ties, a com pa ra ble pal ae o-hydro log i cal con trol
on the As con tent of ground wa ter in the Red River delta
might be expected. The deep est ele va tion of the bot tom
of the incised val ley in the pres ent tran sect is only about
¡54 m, and the lower chan nel slope dur ing the sea level
low, when the shore line moved east ward to the shelf
edge south of Ha inan island (inset in Fig. 1A), makes it
dif cult to argue that the incised val ley in the tran sect
was much deeper than 50–60 m. Con di tions required for a
thick unsat u rated zone to develop are a com bi na tion of a
low recharge and a high trans mis siv ity of the sed i ments.
The thick coarse grained depos its in the NE part of the
tran sect have a high trans mis siv ity, and these depos its
have a regional exten sion in the cen tral part of the delta
(Ma thers and Zal asiewicz, 1999). Low mete oric recharge
is likely under the cli matic arid con di tions which pre-
vailed in parts of the cold gla cial phases (Liew et al., 1998;
Rost, 2000; Zheng and Li, 2000; Jian et al., 2001). There-
fore, in the Red River delta, ground wa ter flow may have
occurred towards the high trans mis siv ity regions, rather
than towards incised pal ae o-chan nels, giv ing rise to deep
atmo spheric oxi diz ing con di tions in the cen tral part of
the delta. This may have caused the observed deep yel low-
ish col our ing of the Pleis to cene depos its in the NE part
of the tran sect (see Sec tion 4.1). Alter na tively, the allu vial
depos its have been oxi dized from the start, in a sed i men-
to log i cal regime fea si ble for co-depo si tion of only lim ited
and low-reac tiv ity organic mat ter, in that way pre serv ing
the oxi dized col our ing.
4.5. Seques tra tion of As
Fig. 7 sug gests ongo ing pre cip i ta tion of iron sulp hides
which in estu a rine, sul phi dic envi ron ments has a large As
sorp tion capac ity (Bo stick et al., 2004). The con cen tra tion
of sea wa ter derived SO4 in a sam ple is cal cu lated as the Cl
con cen tra tion in the sam ple mul ti plied by the SO4/Cl ratio
in oce anic sea wa ter (29.5 mM/566 mM = 0.052). The result
of this cal cu la tion is shown in Fig. 8, where the grey and
the black dots indi cate, respec tively, the sea wa ter derived
SO4 and the mea sured ground wa ter SO4 con cen tra tions.
Except for the sam ple from the shal low screen in Q131, all
sam ples have a defi cit of SO4 (Fig. 8), which must be due to
the reduc tion of SO4 to sul phide.
Fig. 9 shows the con cen tra tion of As vs. ‘miss ing SO4’,
the lat ter cal cu lated as the dif fer ence between the mea-
sured SO4 and sea wa ter derived SO4. The As con cen tra tion
decreases when the amount of miss ing SO4 increases, sug-
gest ing As seques tra tion in Fe sulp hides (Fig. 9). This trend
is not fol lowed by the sam ple from Q85 (61 mbs) which
had an excep tion ally high CH4 con cen tra tion (Fig. 5E) and
appears to be an out lier (see Sec tion 4.3). In Fig. 9, sam ples
that plot close to the 0.013 lM detec tion limit for As and
which are inde pen dent of miss ing SO4 (open dia monds)
are from the Pleis to cene aqui fer in the NE part of the tran-
Fig. 8. The dis tri bu tion of the cal cu lated sea wa ter derived SO4 con cen tra tion (grey) and the mea sured (pres ent) con cen tra tion of SO4 (black) (Seaw. = sea-
wa ter).
Fig. 9. The mea sured con cen tra tion of ground wa ter As plot ted against
the cal cu lated con cen tra tion of miss ing SO4 (see text). Note the log a rith-
mic axis used for As.
S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126 3125
sect and from shal low screens, rep re sent ing super fi cial
ground wa ter which may have a lower As con cen tra tion
(c.f., Post ma et al., 2007) (For the dis tri bu tion of ground-
wa ter As see Fig. 6). It should be noted, that because most
of the sea wa ter derived SO4 has in fact been reduced, the
rela tion ship between As and ‘sea wa ter derived SO4’ is
sim i lar to that shown in Fig. 9 for As and ‘miss ing SO4’,
infer ring that a con trol on ground wa ter As might exist
even with out the pre cip i ta tion of Fe sulp hides. Also, the
Stiff dia grams (Fig. 4C) for the Holo cene aqui fer indi cate
fresh wa ter sig na tures and Cl (and SO4) in these sam ples
may be derived from sea spray from the SE mon soon. The
Cl con cen tra tions in the fresh wa ter in the tran sect is gen-
er ally an order of mag ni tude higher than in fresh ground-
wa ter from upstream Ha noi (<0.2 mM; Lar sen et al., 2008).
Nev er the less, the sat u ra tion for dis or dered mack i naw ite
(Fig. 7) com bined with the drop in ground wa ter As con-
cen tra tions with the increas ing amount of miss ing SO4
(Fig. 9) indi cate that As seques tra tion in Fe sulp hides may
con trol ground wa ter As con cen tra tions. This sug gests that
a dis tinc tion between As con trol in brack ish-marine influ-
enced geo log i cal set tings and fresh wa ter set tings is made
in future stud ies.
The super sat u ra tion of the ground wa ter for sid er ite
(Fig. 7) indi cates that sid er ite could also be a sink for As.
Seng upta et al. (2004) iden ti fied As in sid er ite con cre tions
in aqui fers of the Ben gal delta. How ever, water chem is try
data sug gest that sid er ite pre cip i ta tion removes pro por-
tion ally more Fe2+ than As, i.e., that the As/Fe ratio of the
pre cip i tate is smaller than the As/Fe ratio of the solu tion
(Del e mos et al., 2006; Post ma et al., 2007).
5. Con clu sions
A sequence strati graphic inter pre ta tion of bore hole
descrip tions and geo phys i cal logs was con ducted. The SW
part of the tran sect con sists of grey estu a rine clays and del-
taic sands, depos ited in a Holo cene incised val ley. The NW
part of the tran sect is dom i nated by Pleis to cene depos its
of yel low ish allu vial grav els and sand under neath depos-
its of estu a rine clay, over lain by a thin Holo cene sed i ment
veneer. A review of the lit er a ture indi cates that the Holo-
cene incised val ley, inter sected by the stud ied tran sect,
com prises the dom i nant part of the Qua ter nary sequence
along the south ern bound ary of the Red River delta, and
that the Pleis to cene allu vial sed i ments, found in the NE
part of the tran sect, are dom i nant in the cen tral delta
plain.
The ground wa ter in the stud ied tran sect is anoxic, and
gen er ally con tains Fe2+, sul phide, NH4 and CH4. Rel a tive to
the lev els of SO4, NH4 and CH4 observed in the NE part of
the tran sect, higher con cen tra tions of NH4 and CH4, and
low con cen tra tions of SO4, are found in the SW part of the
tran sect, indi cat ing that the ground wa ter redox con di tions
are related to the geo log i cal age of the sed i ments. Sim i-
larly, higher ground wa ter As con cen tra tions are found in
the SW part of the tran sect (up to 11.7 lM or nearly 900 lg/
L), com pared to the NE part of the tran sect (60.32 lM).
The con cen tra tions of Fe2+ and H2S are con trolled by
the pre cip i ta tion of dis or dered mack i naw ite and sid er-
ite, as indi cated by PHRE EQC-2 spe ci a tion cal cu la tions. A
neg a tive cor re la tion between the ground wa ter As con cen-
tra tions and a Cl-based esti mate of reduced SO4 indi cate,
when com bined with the observed equi lib rium con di tion
for dis or dered mack i naw ite, that Fe sulp hides are a sink
for As in the tran sect.
Water type sig na tures and the for ma tion elec tri cal con-
duc tiv ity logs indi cate that the SO4 orig i nates from sea wa-
ter intrud ing dur ing the mid-Holo cene trans gres sion of the
delta (i.e., from the marine ter races cov er ing the delta area)
and entrapped sea wa ter in estu a rine sed i ment units.
Acknowl edge ments
This study has been con ducted with a grant from DAN-
IDA. We are espe cially grate ful to Michael Berg for shar-
ing with us the results from the EA WAG / CETASD sur vey
at an early stage; our sam pling wells were selected based
on his data. We thank Per Jen sen (GEUS) for skill fully
under tak ing the geo phys i cal log ging and Tor ben Dolin
(TUD) for assist ing with the art work. We also thank two
anon y mous review ers for com ments which improved the
man u script.
Ref er ences
Ahmed, K.M., Bhat tach arya, P., Ha san, M.A., Akh ter, S.H., Alam, S.M.M., Bhuy ian, M.A.H., Imam, M.B., Khan, A.A., Sracek, O., 2004. Arsenic enrich ment in ground wa ter of the allu vial aqui fers in Ban gla desh: an over view. Appl. Geo chem. 19, 181–200.
Akai, J., Iz umi, K., Fuku ha ra, H., Ma su da, H., Nak ano, S., Yo shim ura, T., Ohfuji, H., Ana war, H.M., Ku ru mi Akai, K., 2004. Min er al og i cal and geo mi cro bi o log i cal inves ti ga tions on ground wa ter arsenic enrich-ment in Ban gla desh. Appl. Geo chem. 19, 215–230.
Amini, M., Ab bas pour, K.C., Berg, M., Win kel, L., Hug, S.J., Hoe hn, E., Yang, H., John son, C.A., 2008. Sta tis ti cal mod el ling of global ge o gen ic arsenic con tam i na tion in ground wa ter. Envi ron. Sci. Tech nol. 42.
Bad loe, C., Ngu yen, T.P.T., Ngu yen, Q.H., 2004. Random sur vey of arsenic con tam i na tion in tube well water of 12 prov inces in Viet nam and ini-tially human health arsenic risk assess ment through food chain. In: Proc. Third Sci en tific Conf. Ha noi Uni ver sity of Sci ence, Mul ti dis ci plin-ary Sci en tific Ses sion “Envi ron men tal Sci ence – Tech nol ogy and Sus-tain able Devel op ment”, 16 Novem ber, CET ASD, Ha noi, Viet nam.
Berg, M., Tran, H.C., Ngu yen, T.C., Viet, P.H., Scher ten leib, R., Giger, W., 2001. Arsenic con tam i na tion of ground wa ter and drink ing water in Viet nam: a human health threat. Envi ron. Sci. Tech nol. 35, 2621–2626.
Berg, M., Sten gel, C., Trang, P.T.K., Viet, P.H., Samp son, M.L., Leng, M., Sam-reth, S., Fred er icks, D., 2007. Mag ni tude of arsenic pol lu tion in the Me kong and Red River del tas – Cam bo dia and Viet nam. Sci. Total Envi-ron. 372, 413–425.
Bo stick, B.C., Chen, C., Fen dorf, S., 2004. Arse nite reten tion mech a nisms within estu a rine sed i ments of Pes ca der o, CA. Envi ron. Sci. Tech nol. 38, 3299–3304.
Boyd, W., Lam, D.D., 2004. Holo cene ele vated sea lev els on the north coast of Viet nam. Aust. Ge ogr. Stud. 42, 77–88.
Char let, L., Polya, D.A., 2006. Arsenic in shal low, reduc ing ground wa ters in south ern Asia: an envi ron men tal health disas ter. Ele ments 2, 91–96.
Cline, J.D., 1967. Spec tro pho to met ric deter mi na tion of hydro gen sul fide in nat u ral waters. Lim nol. Oce a nogr. 14, 454–458.
Del e mos, J.L., Bo stick, B.C., Ren shaw, C.E., Stü rup, S., Feng, X., 2006. Land-fill-stim u lated iron reduc tion and arsenic release at the Coak ley Super fund Site (NH). Envi ron. Sci. Tech nol. 40, 67–73.
Funa bi ki, A., Har uy ama, S., Ngu yen, V.Q., Viet, P.H., Dinh, H.T., 2007. Holo-cene delta plain devel op ment in the Song Hong (Red River) delta, Viet-nam. J. Asian Earth Sci. 30, 518–529.
Gani, M.R., Alam, M.M., 2004. Flu vial facies ar chi tech ture in small-scale river sys tems in the Upper Dupi Tila For ma tion, north east Ben gal Basin, Ban gla desh. J. Asian Earth Sci. 24, 225–236.
Good bred, S.L., Ku ehl, S.A., 2000. The sig nifi cance of large sed i ment sup-ply, active tec to nism, and eu sta sy on mar gin sequence devel op ment: late qua ter nary stra tig ra phy and evo lu tion of the Gan ges–Brah mapu-tra delta. Sed i ment. Geol. 133, 227–248.
Guil lot, S., Char let, L., 2007. Ben gal arsenic, an archive of Hima laya orog-eny and pa leo hy drol o gy. J. Envi ron. Sci. Health, A 42, 1785–1794.
3126 S. Jes sen et al. / Applied Geochemistry 23 (2008) 3116–3126
Hane buth, T.J.J., Sa i to, Y., Tan a be, S., Quang, L.V., Quang, T.N., 2006. Sea lev els dur ing late marine iso tope stage 3 (or older?) reported from the Red River delta (north ern Viet nam) and adja cent regions. Qua-tern. Int. (145/146), 119–134.
Har vey, C.F., Swartz, C.H., Bad ruzz aman, A.B.M., Keon-Blute, N., Yu, W., Ali, M.A., Jay, J., Bec kie, R., Nie dan, V., Brab an der, D., Oates, P.M., Ashf-aque, K.N., Islam, S., He mond, H.F., Ahmed, M.F., 2005. Ground wa ter arsenic con tam i na tion on the Gan ges Delta: bio geo chem is try, hydrol-ogy, human per tur ba tions, and human suffering on a large scale. C. R. Geo sci. 337, 285–296.
Hori, K., Tan a be, S., Sa i to, Y., Har uy ama, S., Viet, N., Ki tam ura, A., 2004. Delta ini ti a tion and Holo cene sea-level change: exam ple from the Song Hong (Red River) delta, Viet nam. Sed i ment. Geol. 164, 237–249.
Hoss ain, F., Hill, J., Bag tzog lou, A.C., 2007. Geo sta tis ti cal ly based man age-ment of arsenic con tam i nated ground water in shal low wells of Ban-gla desh. Water Re sour. Man age. 21, 1245–1261.
Islam, F.S., Ga ult, A.G., Booth man, C., Polya, D.A., Char nock, J.M., Chat ter-jee, D., Lloyd, J.R., 2004. Role of metal-reduc ing bac te ria in arsenic release from Ben gal delta sed i ments. Nature 430, 68–71.
Jian, Z., Hu ang, B., Ku hnt, W., Lin, H.-L., 2001. Late Qua ter nary upwell ing inten sity and East Asian mon soon forc ing in the South China Sea. Qua-tern. Res. 55, 336–370.
Kit az a wa, T., 2007. Pleis to cene mac ro tid al tide-dom i nated estu ary–delta suc ces sion, along the Dong Nai River, south ern Viet nam. Sed i ment. Geol. 194, 115–140.
Lam, D.D., Boyd, W.E., 2003. Holo cene cos tal stra tig ra phy and the sed i-men tary devel op ment of the Hai Phong area of the Bac Bo plain (Red River delta), Viet nam. Aust. Ge ogr. 34, 177–194.
Lam beck, K., Yo koy ama, Y., Pur cell, T., 2002. Into and out of the last gla cial max i mum: sea-level change dur ing oxy gen iso tope stages 3 and 2. Qua tern. Sci. Rev. 21, 343–360.
Lang muir, D.L., Ma hon ey, J., Row son, J., 2006. Sol u bil ity prod ucts of amor-phous fer ric arse nate and crys tal line sco ro dite (Fe AsO4·2H2O) and their appli ca tion to arsenic behav ior in bur ied mine tail ings. Geo-chim. Cos mo chim. Acta 70, 2942–2956.
Lar sen, F., Pham, N.Q., Dang, N.D., Post ma, D., Jes sen, S., Pham, V.H., Ngu-yen, T.B., Trieu, H.D., Tran, L.T., Ngu yen, H., Cham bon, J., Ngu yen. H.V., Ha, D. H., Hue, N.T, Duc, M.T., Ref sg aard, J.C., 2008. Con trol ling geo-log i cal and hydro geo log i cal pro cesses in an arsenic con tam i nated aqui fer on the Red River flood plain, Viet nam. Appl. Geo chem. 23(11), 3099-3115.
Li, Z., Sa i to, Y., Mat sum ot o, E., Wang, Y., Tan a be, S., Quang, L.V., 2006. Cli-mate change and human impact on the Song Hong (Red River) delta, Viet nam, dur ing the Holo cene. Qua tern. Int. 144, 4–28.
Liew, P.M., Kuo, C.M., Hu ang, S.Y., Tseng, M.H., 1998. Veg e ta tion change and ter res trial car bon stor age in east ern Asia dur ing the last gla cial max i mum as indi cated by a new pol len record from cen tral Tai wan. Global Planet. Change (16/17), 85–94.
Low ers, H.A., Bre it, G.N., Fos ter, A.L., Whit ney, J., You nt, J., Ud din, Md.N., Mu neem, Ad.A., 2007. Arsenic incor po ra tion into authi genic pyrite, Ben gal Basin sed i ment, Ban gla desh. Geo chim. Cos mo chim. Acta 71, 2699–2717.
Ma thers, S., Zal asiewicz, J., 1999. Holo cene sed i men tary archi tec ture of the Red River delta Viet nam. J. Coast. Res. 15, 314–325.
McAr thur, J.M., Rav ens croft, P., Safiu lla, S., Thirl wall, M.F., 2001. Arsenic in ground wa ter: test ing pol lu tion mech a nisms for sed i men tary aqui fers in Ban gla desh. Water Re sour. Res. 37, 109–117.
Park hurst, D.L., Ap pe lo, C.A.J., 1999. User’s guide to PHRE EQC (Ver sion 2) – a com puter pro gram for spe ci a tion, reac tion-path, 1D-trans port, and inverse geo chem i cal cal cu la tions. U.S Geol. Surv. Water Re sour. Invest. Rep. 99–4259.
Pet er sen, H.I., Ander sen, C., Anh, P.H., Bo je sen-Koef oed, J.A., Niel sen, L.H., Nyt oft, H.P., Rosen berg, P., Thanh, L., 2001. Petro leum potential of Oli-go cene lacus trine mud stones and coals at Dong Ho, Viet nam – an out-crop ana logue to ter res trial source rocks in the greater Song Hong Basin. J. Asian Earth Sci. 19, 135–154.
Pet er sen, H.I., Vu, T., Niel sen, L.H., Ngu yen, A.D., Nyt oft, H.P., 2005. Source rock prop er ties of lacus trine mud stones and coals (Oli go cene Dong Ho For ma tion), onshore Song Hong Basin, north ern Viet nam. J. Pet rol. Geol. 28, 19–38.
Polya, D.A., Ga ult, A.G., Di eb e, N., Feld man, P., Ro sen boom, J.W., Gil li gan, E., Fred er icks, D., Mil ton, A.H., Samp son, M., Row land, H.A.L., Lyth goe, P.R., Jones, J.C., Mid dle ton, C., Cooke, D.A., 2005. Arsenic haz ard in shal low Cam bo dian ground wa ters. Min eral. Mag. 69, 807–823.
Post ma, D., Lar sen, F., Hue, N.T.M., Duc, M.T., Viet, P.H., Nhan, P.Q., Jes sen, S., 2007. Arsenic in ground wa ter of the Red River flood plain, Viet nam: con trol ling geo chem i cal pro cesses and reac tive trans port mod el ing. Geo chim. Cos mo chim. Acta 71, 5054–5071.
Rav ens croft, P., McAr thur, J.M., Ho que, B.A., 2001. Geo chem i cal and pal-a eohy dro log i cal con trols on pol lu tion of ground wa ter by arsenic. In: Chap pell, W.R., Ab ern a thy, C.O., Cald er on, R.L. (Eds.), Fourth Int. Conf. Arsenic Expo sure and Health Effects. Else vier Sci ence, Ltd, Oxford.
Rav ens croft, P., Bur gess, W.G., Ahmed, K.M., Bur ren, M., Per rin, J., 2005. Arsenic in ground wa ter of the Ben gal Basin, Ban gla desh: dis tri bu-tion, field rela tions, and hydro geo log i cal set ting. Hy dro ge ol. J. 13, 727–751.
Rod rí guez Lado, L., Polya, D., Win kel, L., Berg, M., Hegan, A., 2008. Mod-el ling arsenic haz ard in Cam bo dia: A geo sta tis ti cal approach using ancil lary data. Appl. Geo chem. 23(11), 3010-3018.
Rost, K.T., 2000. Pleis to cene pa leo en vi ron men tal changes in the high moun tain ranges of cen tral China and adja cent regions. Qua tern. Int. (65/66), 147–160.
Row land, H.A.L., Polya, D.A., Lloyd, J.R., Pan cost, R.D., 2006. Char ac ter isa-tion of organic mat ter in a shal low, reduc ing, arsenic-rich aqui fer, West Ben gal. Org. Geo chem. 37, 1101–1114.
Row land, H.A.L., Ped er ick, R.L., Polya, D.A., Pan cost, R.A., van Don gen, B.E., Ga ult, A.G., Vaughan, D.J., Bry ant, C., Ander son, B., Lloyd, J.R., 2007. The con trol of organic mat ter on mi cro bi al ly med i ated iron reduc tion and arsenic release in shal low allu vial aqui fers, Cam bo dia. Geo bi ol-o gy 5, 281–292.
Seng upta, S., Muk her jee, P.K., Pal, T., Shome, S., 2004. Nature and ori gin of arsenic car ri ers in shal low aqui fer sed i ments of Ben gal Delta, India. Envi ron. Geol. 45, 1071–1081.
Shah, B.A., 2008. Role of Qua ter nary stra tig ra phy on arsenic-con tam-i nated ground wa ter from parts of Mid dle Ganga Plain, UP–Bi har, India. Envi ron. Geol. 53, 1553–1561.
Smed ley, P.L., Kinni burgh, D.G., 2002. A review of the source, behav iour and dis tri bu tion of arsenic in nat u ral waters. Appl. Geo chem. 17, 517–568.
Stan ger, G., 2005. A pal ae o-hydro geo log i cal model for arsenic con tam-i na tion in south ern and south-east Asia. Envi ron. Geo chem. Health 27, 359–367.
Stol len werk, K.G., Bre it, G.N., Welch, A.H., You nt, J.C., Whit ney, J.W., Fos-ter, A.L., Ud din, M.N., Ma jum der, R.K., Ahmed, N., 2007. Arsenic atten u-a tion by oxi dized aqui fer sed i ments in Ban gla desh. Sci. Total Envi ron. 379, 133–150.
Stoo key, L.L., 1970. Fer ro zine – a new spec tro pho to met ric reagent for iron. Anal. Chem. 42, 779–781.
Stumm, W., Mor gan, J.J., 1981. Aquatic Chem is try, sec ond ed. Wiley & Sons, New York.
Swartz, C.H., Blute, N.K., Bad ruzz man, B., Ali, A., Brab an der, D., Jay, J., Be san con, J., Islam, S., He mond, H.F., Har vey, C.F., 2004. Mobil ity of arsenic in a Ban gla desh aqui fer: Infer ences from geo chem i cal pro-files, leach ing data, and min er al og i cal char ac ter iza tion. Geo chim. Cos-mo chim. Acta 68, 4539–4557.
Tan a be, S., Hori, K., Sa i to, Y., Har uy ama, S., Le, Q.D., Sato, Y., Hira ide, S., 2003a. Sed i men tary facies and radio car bon dates of the Nam Dinh-1 core from the Song Hong (Red River) delta, Viet nam. J. Asian Earth Sci. 21, 503–513.
Tan a be, S., Hori, K., Sa i to, Y., Har uy ama, S., Van, P.V., Ki tam ura, A., 2003b. Song Hong (Red River) delta evo lu tion related to mil len nium-scale Holo cene sea-level changes. Qua tern. Sci. Rev. 22, 2345–2361.
Tan a be, S., Sa i to, Y., Quang, L.V., Hane buth, T.J.J., Quang, L.N., Ki tam ura, A., 2006. Holo cene evo lu tion of the Song Hong (Red River) delta sys tem, north ern Viet nam. Sed i ment. Geol. 187, 29–61.
Tran, N., Ngo, Q.T., Do, T.V.T., Ngu yen, D.M., Ngu yen, V.V., 1991. Qua ter-nary sed i men ta tion of the prin ci pal del tas of Viet nam. J. South east Asian Earth Sci. 6, 103–110.
Tran, N., Mai, T.N., Chu, V.N., Hoek stra, P., Weer ing, V.Tj., van den Bergh, J.H., Dinh, X.T., Ngu yen, D.N., Vu, V.P., 2002. Holo cene sed i men tary evo lu tion, geo dy namic and anthro po genic con trol of the Ba lat river mouth for ma tion (Red River-delta, north ern Viet nam). Z. Geol. Wiss., Ber lin 30, 157–172.
Tray nor, J.J., Sla den, C., 1997. Seep age in Viet nam – onshore and off shore exam ples. Mar. Pet rol. Geol. 14, 345–362.
Vu, T.C., 1996. Salin ity intru sion in the Red River delta. Sem i nar on Envi-ron ment and Devel op ment in Viet nam, Decem ber 6-7, Aus tra lian National Uni ver sity. (See coo mbs.anu.edu.au/~vern/env_dev/sem i-nar96.html).
Win kel, L., Berg, M., Amini, M., Hug, S.J., John son, C.A., 2008. Pre dict ing ground wa ter arsenic con tam i na tion in South east Asia from sur face param e ters. Nature Geo sci. 1, 536–542.
Yo koy ama, Y., Lam beck, K., Deck ker, P.D., John ston, P., Fi field, L.K., 2000. Tim ing of the last gla cial max i mum from observed sea-level min ima. Nature 406, 713–716.
Zheng, Z., Li, Q., 2000. Veg e ta tion, cli mate, and sea level in the past 55,000 years, Han ji ang Delta, south east ern China. Qua tern. Res. 53, 330–340.
