DSpace at VNU: Palaeo-hydrogeological control on groundwater As levels in Red River delta, Vietnam

Applied Geochemistry 23 (2008) 3116–3126

Contents lists available at ScienceDirect

Applied Geochemistry
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 / a p g e o ch e m

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

Depart­ment of Envi­ron­men­tal Engi­neer­ing, Tech­ni­cal Uni­ver­sity of Den­mark (TUD), 2800 Kgs. Lyn­gby, Den­mark
National Geo­log­ic­ al Sur­vey of Den­mark and Green­land (GEUS), Den­mark
c
Research Cen­tre for Envi­ron­men­tal Tech­nol­ogy and Sus­tain­able Devel­op­ment (CET­ASD), Ha­noi Uni­ver­sity of Sci­ence, Viet Nam
d
Viet­nam North­ern Hydro­geo­log­i­cal and Engi­neer­ing Geo­log­i­cal Divi­sion (NHEGD), Viet Nam
e
Ha­noi Uni­ver­sity of Min­ing and Geol­ogy (HUMG), Viet Nam
f
Insti­tute for Nuclear Sci­ence and Tech­nol­ogy, Viet Nam
b

a r t i c l e

i n f o

Article histry:

Available online 4 July 2008


a b s t r a c t
To study the geo­log­i­cal con­trol on ground­wa­ter As con­cen­tra­tions in Red River delta,
depth-spe­cific ground­wa­ter sam­pling and geo­phys­i­cal log­ging in 11 mon­it­ or­ing wells
was con­ducted along a 45 km tran­sect across the south­ern and cen­tral part of the delta,
and the lit­er­a­ture on the Red River delta’s Qua­ter­nary geo­log­i­cal devel­op­ment was
reviewed. The water sam­ples (n = 30) were ana­lyzed for As, major ions, Fe2+, H2S, NH4,
CH4, d18O and dD, and the geo­phys­ic­ al log suite included nat­u­ral gamma-ray, for­ma­
tion and fluid elec­tri­cal con­duc­tiv­ity. The SW part of the tran­sect inter­sects depos­its of
grey estu­a­rine clays and del­taic sands in a 15–20 km wide and 50–60 m deep Holo­cene
incised val­ley. The NE part of the tran­sect con­sists of 60–120 m of Pleis­to­cene yel­low­ish
allu­vial depos­its under­neath 10–30 m of estu­a­rine clay over­lain by a 10–20 m veneer
of Holo­cene sed­i­ments. The dis­tri­bu­tion of d18O-val­ues (range ¡12.2‰ to ¡6.3‰) and
hydrau­lic head in the sam­ple wells indi­cate that the estu­a­rine clay units divide the flow
sys­tem into an upper Holo­cene aqui­fer and a lower Pleis­to­cene aqui­fer. The ground­wa­
ter sam­ples were all anoxic, and con­tained Fe2+ (0.03–2.0 mM), Mn (0.7–320 lM), SO4
(<2.1 lM–0.75 mM), H2S (<0.1–7.0 lM), NH4 (0.03–4.4 mM), and CH4 (0.08–14.5 mM).
Gen­er­ally, higher con­cen­tra­tions of NH4 and CH4 and low con­cen­tra­tions of SO4 were
found in the SW part of the tran­sect, dom­i­nated by Holo­cene depos­its, while the oppo­
site was the case for the NE part of the tran­sect. The dis­tri­bu­tion of the ground­wa­ter As
con­cen­tra­tion (<0.013–11.7 lM; median 0.12 lM (9 lg/L)) is related to the dis­tri­bu­tion of
NH4, CH4 and SO4. Low con­cen­tra­tions of As (60.32 lM) were found in the Pleis­to­cene
aqui­fer, while the high­est As con­cen­tra­tions were found in the Holo­cene aqui­fer. PHRE­
EQC-2 spe­ci­a­tion cal­cu­la­tions indi­cated that Fe2+ and H2S con­cen­tra­tions are con­trolled
by equi­lib­rium for dis­or­dered mack­i­naw­ite and pre­cip­i­ta­tion of sid­er­ite. An ele­vated
ground­wa­ter salin­ity (Cl range 0.19–65.1 mM) was observed in both aqui­fers, and dom­i­
nated in the deep aqui­fer. A neg­a­tive cor­re­la­tion between aque­ous As and an esti­mate of
reduced SO4 was observed, indi­cat­ing that Fe sul­phide pre­cip­i­ta­tion poses a sec­ond­ary

con­trol on the ground­wa­ter As con­cen­tra­tion.
© 2008 Else­vier Ltd. All rights reserved.

* Cor­re­spond­ing author.
E-mail address: (S. Jessen).
0883-2927/$ - see front matter © 2008 Else­vier Ltd. All rights reserved.
doi:10.1016/j.apgeochem.2008.06.015




S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

1. Intro­duc­tion
Ele­vated con­cen­tra­tions of ge­og
­ en­ic As in ground­wa­ter
poses a threat to the health of tens of mil­lions of people
living in the large delta areas of South­east Asia. In the Red
River delta, Viet­nam, an esti­mated 11 mil­lion people are
at risk (Berg et al., 2001). Although sev­eral pro­cesses lead­
ing to the release of As have been pro­posed, the reduc­tion
of As-con­tain­ing Fe-oxides with nat­u­ral organic mat­ter
is gen­er­ally con­sid­ered the most impor­tant mobi­li­za­tion
mech­a­nism (McAr­thur et al., 2001; Akai et al., 2004;
Islam et al., 2004; Post­ma et al., 2007). On a local scale,
the ground­wa­ter As con­cen­tra­tion often shows a patchy
dis­tri­bu­tion (Har­vey et al., 2005; Char­let and Polya, 2006)
prob­a­bly deter­mined by the local hydro­ge­ol­ogy and/or
vari­a­tions in abstrac­tion depth (Smed­ley and Kinni­burgh,
2002; Har­vey et al., 2005). How­ever, on a larger scale,

regional sur­veys in the Ben­gal and Me­kong del­tas indi­
cate the exis­tence of areas in which new wells are likely
to pro­duce low-As water (McAr­thur et al., 2001; Berg et
al., 2007). Regional sur­veys are costly and time con­sum­
ing and a pre­dic­tion of the ground­wa­ter As con­tent based
on exist­ing data would there­fore be pref­er­ab
­ le. Recently
maps that pre­dict the As con­cen­tra­tion in the ground­wa­ter
based on geo­log­i­cal and sur­face-soil param­e­ters have been
val­id
­ ated with rea­son­able suc­cess against sur­vey data­sets
from the Ben­gal, Me­kong, Red River, Myan­mar and Suma­
tra delta areas (Polya et al., 2005; Hoss­ain et al., 2007;
Berg et al., 2007; Win­kel et al., 2008; Rodri­guez-Lado et
al., 2008) and also global ground­wa­ter As pre­dic­tion mod­
el­ling has been con­ducted (Amini et al., 2008). For the
Red River delta, a val­i­da­tion and refine­ment of pre­dic­tion
maps should be pos­si­ble, because thousands of ground­
wa­ter sam­ples have been ana­lyzed in two recent regional
sur­veys by UNICEF (Bad­loe et al., 2004) and the Swiss Fed­
eral Insti­tute of Aquatic Sci­ence and Tech­nol­ogy, EA­WAG,
in coop­er­a­tion with CET­ASD (Unpub­lished data, Michael

3117

Berg, pers. comm.). The EA­WAG/CET­ASD sur­vey shows a
high ground­wa­ter As con­cen­tra­tion in a 20 km wide band
along the NW–SE bound­ary of the delta plain, par­al­lel to
the position of the pal­ae­o-Red River main chan­nel (Fig.
1A), while ground­wa­ters in the cen­tral and north­ern delta

plain gen­er­ally have low lev­els of As. These results are
con­sis­tent with the results of the UNICEF sur­vey, though
the lat­ter are reported per prov­ince. In the Ben­gal delta, a
high ground­wa­ter As con­cen­tra­tion is found in Holo­cene
aqui­fers, while Pleis­to­cene aqui­fers have a low-As level
(Rav­ens­croft et al., 2001, 2005). The South­east Asian del­tas
all derive their sed­i­ments ulti­mately from the Hima­la­yas
(Stan­ger, 2005; Char­let and Polya, 2006; Guil­lot and Char­
let, 2007), and depo­si­tion takes place in Ceno­zoic sub­si­
dence basins under the influ­ence of Qua­ter­nary eu­stat­ic
sea level changes (Tan­a­be et al., 2006). The sub­si­dence
rate in the Ben­gal delta (Good­bred and Ku­ehl, 2000) has
been much higher than that in the Red River delta, and in
the onshore Red River delta the Qua­ter­nary sequence is
only up to 200 m in total (Ma­thers and Zal­asiewicz, 1999).
Dur­ing the mid-Holo­cene trans­gres­sion the sea cov­ered
a large part of the Red River delta (Tan­a­be et al., 2006).
This trans­gres­sion, com­bined with the mod­est total thick­
ness of the Qua­ter­nary aqui­fers, influ­ences the pres­ent day
ground­wa­ter chem­is­try in the Red River delta.
In this study ground­wa­ter sam­ples have been col­
lected from mon­i­tor­ing wells along a 45 km tran­sect run­
ning per­pen­dic­u­lar to the Red River, from the south­ern
delta bound­ary to the cen­tral part of the delta (Fig. 1A
and B). The aims of the pres­ent study are (i) to pro­vide
a first assess­ment of the over­all dis­tri­bu­tion of Pleis­to­
cene and Holo­cene sed­i­ments in the Red River delta, (ii)
to inves­ti­gate if the ground­wa­ter As lev­els in the Red
River delta are related to geo­log­i­cal age of the aqui­fer
sed­i­ments, and (iii) to inves­ti­gate the con­trols on the

ground­wa­ter As con­cen­tra­tion in the brack­ish-marine
sed­i­ments of the Red River delta.

Fig. 1. (A) The Red River delta, pres­ent day sit­ua
­ ­tion and pal­ae­og­e­og­ra­phy. A–A9 indi­cates the position of the stud­ied tran­sect. (B) The position of bore­holes
for water sam­pling (Q82–Q131) and sed­i­ment descrip­tions along the tran­sect.


3118

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

2. Meth­ods

The field work was car­ried out from 27 May to 5 June
2006 and sam­pling sites included 11 of the Viet­nam­ese
National Mon­i­tor­ing Net­work wells (Fig. 1B). The wells
are nests with up to three sep­a­rate holes. Typ­i­cally, the
bore­holes are equipped with OD 120 mm PVC cas­ings and
screens; the upper one or two screens are 6 m long while
the deep­est screen is 8–10 m long.

to pre­vent As seques­tra­tion by Fe oxide pre­cip­i­ta­tion. Sam­
ples for CH4 were injected into pre-weighed evac­u­ated
glass vials equipped with a pierce­able sep­tum (ex­e­tain­ers;
Lab­co, ord.co. 819W), through a syringe nee­dle mounted
on the sam­pling tube. The ex­e­tain­ers were stored upsidedown to trap the CH4 in the head­space vol­ume of 2–3 mL.
Sam­ples for anions, NH4 and CH4 were put on ice in the
field and fro­zen later on the day of sam­pling. Sam­ples for
all other param­e­ters were acid­i­fied by add­ing 2 vol% of a

7 M HNO3 solu­tion, then put on ice and stored refrig­er­ated.
Detec­tion lim­its for Fe2+, H2S and PO4 were ca. 0.1 lM.

2.2. Bore­hole log­ging and water table mea­sure­ments

2.4. Lab­o­ra­tory anal­y­sis

Geo­phys­i­cal log­ging was car­ried out in the deep­est
bore­hole at each mon­i­tor­ing nest. The log suite included
nat­u­ral gamma-ray, for­ma­tion elec­tri­cal con­duc­tiv­ity
(focused induc­tion), and fluid tem­per­at­ ure and con­duc­tiv­
ity (Rob­ert­son Geo­log­ging). The gamma-ray log is a proxy
for sed­i­men­tary clay con­tent, while the for­ma­tion con­
duc­tiv­ity log is a proxy for the salin­ity of the for­ma­tion
pore water. To trans­late for­ma­tion con­duc­tiv­ity log-val­ues
into esti­mates of pore water salin­ity, the for­ma­tion fac­tor
Ff = rw/rf was used, where rw and rf are the con­duc­tiv­it­ ies
of the water sam­ple (mea­sured by sam­pling) and of the
for­ma­tion (mea­sured by log­ging), respec­tively. Thus, after
esti­mat­ing the for­ma­tion fac­tor for a given sed­i­men­tary
unit in the tran­sect, the pore water salin­ity expressed as
rw can be cal­cu­lated from the for­ma­tion con­duc­tiv­ity log
response.
To assess the aqui­fer hydro­dy­nam­ics the rel­a­tive
static water lev­els in the bore­holes at each loca­tion were
recorded and cor­rected for den­sity vari­a­tions using fluid
con­duc­tiv­ity mea­sure­ments to obtain com­pa­ra­ble heads.

Cat­ions were ana­lyzed by flame atomic absorp­tion
spec­tro­pho­tom­e­try on a Shi­ma­dzu AAS 6800 instru­ment.

Aque­ous As was deter­mined on the same instru­ment using
a HVG hydride gen­er­a­tor and a graph­ite fur­nace. Anions
were ana­lyzed by ion chro­ma­tog­ra­phy using a Shi­ma­dzu
LC20AD/HIC-20ASu­per. Ammo­nium was deter­mined by
spec­tro­pho­tom­e­try using nitro­prus­side. Meth­ane head
space con­cen­tra­tions were deter­mined by gas chro­ma­tog­
ra­phy using a Shi­ma­dzu GC-14A with a 1 m packed col­umn
(3% SP1500, Car­bo­pack B) and a FID detec­tor. The aque­ous
CH4 con­cen­tra­tion was cal­cu­lated using Henry’s law. Detec­
tion lim­its 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 sta­ble iso­tope ratio of O (18O/16O) and H (2H/1H)
of the water rel­a­tive to the VSMOW stan­dard was ana­
lyzed using a Mi­cro­Mass spec­trom­e­ter (Iso­Prime, GV
Instru­ments, UK) equipped with an Eu­ro­vec­tor ele­men­tal
ana­lyzer (Eu­roEA 3000, Italy) and the Mass­lynx Pro­gram
(GV Instru­ments, UK) for data pro­cess­ing. The results are
expressed in ‰ units using the d-nota­tion with a stan­dard
devi­a­tion not larger than ±0.20‰ (for d18O), as cal­cu­lated
from a min­i­mum of five rep­li­cate injec­tions into the ele­
men­tal ana­lyzer.

2.1. Field cam­paign

2.3. Water sam­pling and field anal­y­sis
To ensure that the water sam­ples orig­i­nated from the
screened inter­vals of the bore­holes the sam­ples were
col­lected by a low-flow-rate Whale-pump (0.5–2 L/min)
posi­tioned at the top of the screen, and below a high flowrate Grund­fos MP1-pump (typ­ic­ ally, 10–20 L/min). In two
of the sam­pled screens, two sam­ples (as opposed to one

sam­ple) were col­lected from sep­a­rate in-flow zones iden­
ti­fied by the geo­phys­i­cal log­ging. Before sam­pling, 3–12
bore­hole vol­umes were flushed. Dis­solved O2, pH, tem­per­
a­ture and elec­tri­cal con­duc­tiv­ity was mea­sured by WTW
elec­trodes in a flow cell and mon­it­ ored through the flush­
ing to ensure sta­ble val­ues before sam­pling. Fer­rous iron
(Fe2+), PO4 and H2S were mea­sured on a HACH DR/2010
spec­tro­pho­tom­e­ter in the field, using, respec­tively, the
Fer­ro­zine (Stoo­key, 1970), molyb­date blue and meth­y­lene
blue meth­ods (Cline, 1967). The Fe2+ con­cen­tra­tions mea­
sured in the field closely matched total Fe mea­sured in the
lab­or­ a­tory by flame atomic absorp­tion spec­tro­pho­tom­e­try
(slope 1.00; r2 = 0.99). Alka­lin­ity was mea­sured in the field
by Gran-titra­tion (Stumm and Mor­gan, 1981). Sam­ples
were col­lected by poly­eth­yl­ene tub­ing and syrin­ges, and
fil­tered through 0.20 lm cel­lu­lose ace­tate syrin­ges-fil­ters
(Sar­to­rius Min­is­art) to poly­eth­yl­ene vials. Prior to sam­
pling, syrin­ges and fil­ters were pre-flushed with N2-gas

2.5. Spe­ci­a­tion cal­cu­la­tions
Aque­ous spe­ci­a­tion was done using PHRE­EQC-2
(Park­hurst and Ap­pe­lo, 1999) with the inclu­sion of the
ther­mo­dy­namic data­base for the As spe­cies pro­vided by
Lang­muir et al. (2006). For Fe sul­phide equi­lib­rium cal­cu­la­
tions the FeS(ppt) phase defined in the data­base for which
FeS M Fe2+ + S2¡, K = 10¡16.833 was used.
3. Results
3.1. Geo­log­ic­ al set­ting
The cross sec­tion shown in Fig. 2 is based on a sequence
strati­graph­i­cal inter­pre­ta­tion of the avail­able lith­o­log­i­cal

and geo­phys­i­cal logs, con­sis­tent with the work pub­lished by
Su­sumu Tan­a­be, his co-work­ers and oth­ers (Tran et al., 1991,
2002; Ma­thers and Zal­asiewicz, 1999; Lam and Boyd, 2003;
Tan­a­be et al., 2003a, b, 2006; Hori et al., 2004; Hane­buth et
al., 2006; Li et al., 2006; Funa­bi­ki et al., 2007). Neo­gene bed­
rock forms the base of the Qua­ter­nary depos­its, which thick­
ens north­ward from about 30 m to 150 m. The Qua­ter­nary
sequence con­sists of Pleis­to­cene allu­vial sand and gravel




S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

3119

Fig. 2. Geo­log­i­cal cross sec­tion of the sam­pled tran­sect (A–A9 in Fig. 1). The Late Pleis­to­cene topo­graphic sur­face is indi­cated by a black dashed line. Geo­
phys­i­cal logs show, to the left, the nat­ur­ al gamma-ray, and to the right, the for­ma­tion con­duc­tiv­ity.

depos­its, over­lain by estu­a­rine clays and Holo­cene del­taic
sands. On the top there are depos­its of silt and clay in Holo­
cene marine ter­races and, near the mod­ern chan­nels such as
the Red River, over­bank depos­its, typ­i­cally form­ing a 5–10 m
thick con­fin­ing clay layer. The cross sec­tion also dis­plays
14
C sed­i­ment dates from the DT hole (Tan­a­be et al., 2003a),
and the HH120 hole and two addi­tional loca­tions (Lam and
Boyd, 2003; Funa­bi­ki et al., 2007). Fault loca­tions from Tran
et al. (2002) are indi­cated on Figs. 1B and 2. The dashed line
in Fig. 2, at ele­va­tion around ¡50 m in the SW part of the

cross sec­tion and around ¡10 m in the NE part of the cross
sec­tion, indi­cates the pre­sumed topo­graphic sur­face dur­ing
the Late Pleis­to­cene sea level low, shown in Fig. 3 by the sea
level curve for the Late Pleis­to­cene–Holo­cene. Fig. 3 shows,
that dur­ing the mid­dle of the Weich­sel gla­ci­a­tion, 80–30 ka
before pres­ent (BP), the sea level oscil­lated at around 80 m
below the pres­ent sea level (Kit­az­a­wa, 2007), and 25–19 ka
BP, the sea level dropped to 120 m below the pres­ent sea
level, as global cool­ing at the time caused sea­wa­ter to accu­
mu­late in ter­res­trial ice-sheets (Yo­koy­ama et al., 2000; Lam­
beck et al., 2002; Tan­a­be et al., 2006). Dur­ing the sea level
low, the pal­ae­o-Red River (Fig. 1A) eroded a val­ley into the
Pleis­to­cene sed­i­ments in the SW part of the cross sec­tion. In
Fig. 1A, the incised val­ley along the NW–SE delta bound­ary
is out­lined by the shore line 9 ka BP (Tan­a­be et al., 2006).
From 6 to 4 ka BP the sea level stood 2–4 m above pres­ent sea
level (Boyd and Lam, 2004; Tan­a­be et al., 2006) evi­dent in
the cross sec­tion as the marine ter­races (Figs. 1A and 2).

Three of the geo­phys­i­cal bore­hole logs recorded dur­
ing the sur­vey are depicted in Fig. 2, rep­re­sent­ing pri­mar­
ily Holo­cene depos­its in Q86 and Pleis­to­cene depos­its in
Q130 and Q131. The Pleis­to­cene allu­vial sed­i­ments in the
bore­hole logs show rel­a­tively low gamma-ray lev­els of
around 75 API, cor­re­spond­ing to rel­a­tively coarse grained
sed­i­ments. The grad­ual upward increase in the gamma-ray
lev­els, e.g., from ele­va­tion ¡68 m to ¡62 m and again from
¡62 m to ¡39 m in Q130, and from ¡84 m to ¡60 m in Q131
indi­cate fin­ing-upward sequences of the allu­vial, higherenergy depo­si­tional envi­ron­ment. High-stand flu­vial sed­
i­ments have smaller-scale fin­ing-upward sequences (Gani

Fig. 3. Com­pi­la­tion of sea level curves (ele­va­tion in meters rel­a­tive to pres­
ent sea level) from Tan­a­be et al. (2006) (0–20 ka BP) and Kit­az­a­wa (2007)
(20–150 ka BP).


3120

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

and Alam, 2004). The two estu­a­rine sed­im
­ ent units (Holo­
cene in the SW, and Pleis­to­cene in the NE part of the tran­
sect) are clay-rich, con­sis­tent with the ele­vated gammaray lev­els of up to 140–150 API. The gamma-ray lev­els in
the Holo­cene sand depos­its are rel­at­ ively high, 110–120
API, con­firm­ing the silty-clayey con­di­tions described in
the lith­o­log­i­cal logs. The lithol­ogy of the sed­i­ments will be
described fur­ther in Sec­tion 4.1.
3.2. Cross sec­tion hydrol­ogy
Fig. 4A shows the dis­tri­bu­tion of d18O in the water
sam­ples col­lected by the depth-spe­cific sam­pling in the
11 mon­i­tor­ing wells (Fig. 1B). The value of d18O ranges
from ¡12.2 to ¡6.3‰, and rel­at­ ively 18O-rich ground­wa­
ter is gen­er­ally found in the Holo­cene depos­its while the
ground­wa­ter in the Pleis­to­cene depos­its is more depleted
in 18O with d18O-val­ues lower than ¡8.6‰. The ver­ti­cal
dis­tri­bu­tion of hydrau­lic head (not shown) gen­er­ally indi­
cated a down­ward migra­tion of water. In the NE part of
the tran­sect (dis­tance 20–50 km), a ver­ti­cal hydrau­lic head
dif­fer­ence of 2.4–2.7 m water col­umn was observed across

the estu­ar­ ine depos­its, while a less pro­nounced ver­ti­cal

hydrau­lic head dif­fer­ence (<0.8 m) was observed in the
SW part of the tran­sect (dis­tance 0–20 km). The hydrau­lic
head dif­fer­ences trans­late into ver­ti­cal gra­di­ents of +8 to
¡146‰, where the neg­a­tive val­ues infer down­ward flow.
The dis­tri­bu­tion of d18O and ver­ti­cal hydrau­lic heads indi­
cate that the estu­a­rine clay depos­its (Fig. 2) divide the
sequence into an upper and a lower aqui­fer. No attempt
was made to deduce any hor­i­zon­tal gra­di­ents along the
tran­sect, as the ele­va­tion of ref­er­ence points was not ver­
i­fied.
3.3. Ground­wa­ter types
The dis­tri­bu­tion of the ground­wa­ter Cl con­cen­tra­tion
is shown in Fig. 4B. The Cl con­cen­tra­tion ranges from 0.19
to 65 mM, and is gen­er­ally higher in the lower aqui­fer. In
Fig. 4C each water sam­ple is rep­re­sented by a Stiff dia­
gram. The fig­ure also includes exam­ples of Stiff dia­grams
for intru­sion (saline water dis­plac­ing fresh­wa­ter), fresh­
en­ing (fresh­wa­ter dis­plac­ing saline water), fresh­wa­ter and
a 10% oce­anic sea­wa­ter. Fresh- or fresh­en­ing waters of,
respec­tively, Ca–HCO3 or Na–HCO3 water types, are dom­
i­nant in the Holo­cene aqui­fer. Fresh­wa­ter is also found in

Fig. 4. (A) The dis­tri­bu­tion of d18O in the tran­sect. The d18O-val­ues in the Pleis­to­cene aqui­fer are depleted rel­a­tive to those in the Holo­cene aqui­fer. (B)
The dis­tri­bu­tion of Cl in the ground­wa­ter. The area of the dots in this and con­sec­u­tive fig­ures vary lin­e­arly with the param­e­ter value (square root scal­ing).
(C) Water types in the sam­pled tran­sect, as indi­cated by Stiff dia­grams. For com­par­is­ on four exam­ples of Stiff dia­grams are shown below dia­gram C to
exem­plify the three dif­fer­ent ground­wa­ter types and 10% oce­anic sea­wa­ter.

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

the Pleis­to­cene aqui­fer in Q87, for which the bore­hole logs
(not shown) describe a rel­a­tively sandy sequence, allow­
ing fresh­wa­ter recharge to the lower aqui­fer. The elec­tri­
cal con­duc­tiv­ity and the Cl con­cen­tra­tion of the fresh­wa­
ter are, respec­tively, 60–120 mS/m and 0.19–3.03 mM (Cl
median: 1.73). Intru­sive waters are dom­in
­ ant in the Pleis­
to­cene aqui­fer and are of a Na–Cl water type, which, due to
cat­ion exchange for Ca2+, are depleted in Na+ rel­at­ ive to the
Na/Cl ratio of sea­wa­ter. The elec­tri­cal con­duc­tiv­ity of the
intru­sive water is as high as 600 mS/m, or 11% that of oce­
anic sea­wa­ter. Com­pared to this, the geo­phys­i­cal logs (Fig.

3121

2) sug­gest a for­ma­tion con­duc­tiv­ity of up to 400 mS/m,
which indi­cates a pore water salin­ity of 600–1000 mS/m
(Ff = 1.5–2.5), cor­re­spond­ing to 11–19% of the salin­ity of
oce­anic sea­wa­ter. Because pres­ent day tide-induced salin­
ity intru­sion in sur­face waters extend only 20–30 km
inland from the coast (Vu, 1996), the high salin­ity found in
the tran­sect (Fig. 4B) must be due to the pres­ence of stag­
nant saline water in the estu­a­rine clay units or, espe­cially
for the upper aqui­fer, from the mid-Holo­cene marine sed­
i­ments. The ­for­ma­tion con­duc­tiv­ity is high­est in the mid­
dle of the estu­a­rine clay units (Fig. 2) and low­ers towards

Fig. 5. The dis­tri­bu­tion of ground­wa­ter redox com­po­nents in the tran­sect. ‘BD’ indi­cates a con­cen­tra­tion below the limit of detec­tion.


3122

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

the top and bot­tom of the units, which could indi­cate that
flush­ing of the stag­nant saline water is con­trolled by dif­
fu­sion.

tra­tion shows a large spa­tial var­i­a­tion, although the high­
est val­ues are found close to the Red River. In the NE end of
the tran­sect the As con­cen­tra­tion is low, the high­est being
0.32 lM.

3.4. Ground­wa­ter redox chem­is­try
4. Dis­cus­sion
The dis­tri­bu­tion of the redox-sen­si­tive param­e­ters,
Fe2+, SO4, H2S, NH4 and CH4, is shown in Fig. 5. All sam­ples
have a detect­able Fe2+ con­cen­tra­tion (Fig. 5A) imply­ing
the ­pres­ence of reduced con­di­tions. Con­sis­tently, the dis­
solved O2 con­cen­tra­tion was always below the 0.016 mM
detec­tion limit The high­est Fe2+ con­cen­tra­tions of 0.50–
0.95 mM (Q86), 2.0 mM (Q88) and 0.56 mM (Q131) are
found in the Pleis­to­cene aqui­fer. The Mn con­cen­tra­tion
(not shown) ranged from 0.7 to 320 lM and was above the
WHO guide­line of 7.3 lM (0.4 mg/L) in 70% of the sam­ples.
The SO4 con­cen­tra­tion (Fig. 5B) is up to 0.75 mM, being

high­est in the NE part of the tran­sect, where SO4 is found
both in the upper aqui­fer and in the Pleis­to­cene aqui­fer. In
the SW end of the tran­sect, SO4 is found in a con­cen­tra­tion of
up to 0.05 mM, mainly in shal­low wells and at depth in Q84.
Sul­phide was detected in the ground­wa­ter (Fig. 5C)
in con­cen­tra­tions of up to 7.0 lM, indi­cat­ing ongo­ing
SO4 reduc­tion. The dis­tri­bu­tion of sul­phide appears scat­
tered, though the high­est lev­els are pres­ent in the Holo­
cene aqui­fer.
High NH4 con­cen­tra­tions (Fig. 5D) of typ­ic­ ally 1–2 mM,
and up to 4.4 mM, are found in the SW part of the tran­sect,
except for many of the shal­low screens, which have con­cen­
tra­tions below 0.2 mM. In the NE part of the tran­sect, the
NH4 con­cen­tra­tion is below 0.7 mM.
The CH4 con­cen­tra­tion (Fig. 5E) shows a dis­tri­bu­tion
sim­il­ ar to that of NH4. In the SW part of the tran­sect the
CH4 con­cen­tra­tion typ­i­cally ranges from 1 to 3 mM, except
from some screens placed in the Holo­cene aqui­fer. A very
high CH4 con­cen­tra­tion of 14.5 mM was found in the deep­
est screen of Q85. Lower lev­els of CH4 are found in the NE
part of the tran­sect.
3.5. The dis­tri­bu­tion of As in ground­wa­ter
The dis­tri­bu­tion of total As in the ground­wa­ter is
shown in Fig. 6. In the SW part of the tran­sect the As con­
cen­tra­tion ranges from below the 0.013 lM detec­tion limit
(1 lg/L) to nearly 12 lM (900 lg/L) in Q128. The As con­cen­

4.1. Lithol­ogy of the sed­im
­ ents
Dur­ing the rapid eu­stat­ic sea level rise around 15 ka BP

(Fig. 3) the incised val­ley became filled with ­suc­ces­sively
estu­a­rine dark-grey silt and clay, over­lain by del­taic ­­darkgrey, in places described as green­ish, silty, clayey sand
and fine sand. In the NE part of the cross sec­tion in Fig.
2, Pleis­to­cene depos­its com­prise most of the Qua­ter­nary
sequence, while Holo­cene depos­its only form a thin super­
fi­cial aqui­fer of up to 10 m in thick­ness. This shal­low Holo­
cene aqui­fer is char­ac­ter­ised by silty and clayey sed­i­ments
depos­ited at a low sed­i­men­ta­tion rate (Lam and Boyd,
2003; Funa­bi­ki et al., 2007). The Pleis­to­cene estu­a­rine
unit, with a top ele­va­tion of around ¡10 m (Fig. 2), con­
sists of lat­er­itic, grey-yel­low or spot­ted silt and clay. Other
stud­ies describe the sequence bound­ary for the Holo­cene
sed­i­ments as a sub­aer­i­ally weath­ered, lat­er­itic marker hori­
zon, dated to the time of the sea level low (Ma­thers and
Zal­asiewicz, 1999; Tran et al., 2002; Funa­bi­ki et al., 2007)
or ear­lier in Pleis­to­cene (Hane­buth et al., 2006). The Pleis­
to­cene aqui­fer con­sists of thick depos­its of sand, gran­ules
and peb­bles, which in parts of the more detailed lith­o­log­i­
cal descrip­tions of bore­holes LK20 and LK25 (Figs. 1B and
2) are described also as grey-yel­low and brown col­oured,
in both bore­holes to ele­va­tions as deep as ¡105 m. The sed­
i­ments under­neath the incised val­ley in the SW part of the
cross sec­tion (ele­va­tion ¡50 to ¡70 m), how­ever, are typ­i­
cally described as grey.
4.2. Redox envi­ron­ment and sed­i­ment age
The con­cen­tra­tions of NH4 and CH4 in and beneath the
Holo­cene incised val­ley in the SW part of the tran­sect are
mark­edly higher than in the NE end of the tran­sect (Fig. 5D
and E). The dis­tri­bu­tion of NH4 and CH4 over depth sug­
gests that they are gen­er­ated by the deg­ra­da­tion of sed­

i­men­tary organic mat­ter within the aqui­fer, rather than
being derived from organic C infil­trat­ing from the sur­face.

Fig. 6. The dis­tri­bu­tion of the ground­wa­ter As con­cen­tra­tion in the tran­sect. The dot-size of Q128 was set to cor­re­spond to 4 lM in order to improve res­o­
lu­tion in the plot. ‘BD’ indi­cates a con­cen­tra­tion below the limit of detec­tion.




S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

Fig. 7. The ion activ­ity prod­ucts (IAPs) for sid­er­ite (FeCO3) and dis­or­dered
mack­i­naw­ite (FeS) vs. depth. PHRE­EQC-2 was used for the spe­ci­a­tion cal­
cu­la­tions.

The high con­cen­tra­tions of NH4 and CH4 in the SW part
of the tran­sect indi­cate that either the abun­dance or the
reac­tiv­ity of the sed­i­men­tary organic mat­ter is higher than
in the Pleis­to­cene depos­its in the NE end of the tran­sect.
While a reduced reac­tiv­ity of the organic mat­ter would
be expected in the older sed­i­ments (c.f., Row­land et al.,
2007), the organic mat­ter con­tent is also likely to be lower
because the sed­i­ments were depos­ited in a higher-energy,
allu­vial regime (Ma­thers and Zal­asiewicz, 1999).
In Fig. 7, the ion activ­ity prod­ucts (IAPs) for dis­or­dered
mack­i­naw­ite (FeS) and sid­er­ite (FeCO3) are plot­ted vs. depth.
The dashed lines in Fig. 7 indi­cate the equi­lib­rium con­stants
for these two phases. The sam­ples plot close to the equi­lib­
rium line for dis­or­dered mack­i­naw­ite and gen­er­ally one to
two log units to the right of the equi­lib­rium line for sid­er­ite,

thus mak­ing the pre­cip­i­ta­tion of sid­er­ite fea­si­ble (Post­ma et
al., 2007). The results in Fig. 7 indi­cate that the pre­cip­i­ta­tion
of Fe sulp­hides and sid­er­ite con­trols the Fe2+ con­cen­tra­tion
in the tran­sect. Both sid­er­ite and amor­phous FeS, the lat­ter
being a pre­cur­sor for pyrite, has been detected in Ben­gal
aqui­fer sed­i­ments (Ahmed et al., 2004; Akai et al., 2004;
Seng­upta et al., 2004; Low­ers et al., 2007).
4.3. As in the ground­wa­ter
The dis­tri­bu­tion of ground­wa­ter As in the tran­sect is
related to the dis­tri­bu­tion of redox spe­cies, espe­cially SO4,
NH4 and CH4. A low-As con­cen­tra­tion is found in the NE
part of the tran­sect in both the Holo­cene and the Pleis­to­
cene aqui­fer (Fig. 6), while a high As con­cen­tra­tion is found
in the more reduced Holo­cene val­ley fill. The high­est con­
cen­tra­tion of ground­wa­ter As is found in the pre­sum­ably
youn­gest aqui­fer sed­i­ments close to the Red River (Fig. 6).
The lateral As dis­tri­bu­tion observed in the rather lim­ited
data­set is, how­ever, sup­ported by the regional As dis­tri­bu­

3123

tion in the delta (Michael Berg, pers. comm.; Bad­loe et al.,
2004). Thus, it appears that ground­wa­ter As lev­els in the
Red River delta are linked to the age of the sed­i­ments.
A hydro­dy­namic con­trol on the As dis­tri­bu­tion is
inferred for Q87 (dis­tance 12.4 km) in which dis­solved
As is found at depth where fresh­wa­ter migrates from the
Holo­cene to the Pleis­to­cene aqui­fer.
An excep­tion to the above described rela­tion­ship
between sed­i­ment age, hydro­dy­nam­ics and As dis­tri­bu­

tion, is the high con­cen­tra­tion of As found in bore­hole Q85,
61 m below sur­face (Fig. 6). This sam­ple has an extremely
high CH4 con­cen­tra­tion of 14.5 mM (232 mg/L) (Fig. 5E) indi­
cat­ing that excep­tional hy­drog­eo­chem­i­cal pro­cesses occur.
In an attempt to char­ac­ter­ize the organic mat­ter sources
respon­si­ble for As release to ground­wa­ters in the Ben­gal
delta, Row­land et al. (2006) found bio­de­grad­able nat­u­
ral petro­leum-derived hydro­car­bons in their sed­i­ments.
Onshore hydro­car­bon seep­age from Ter­tiary out­crops in
the Red River delta plain has been reported (Tray­nor and
Sla­den, 1997; Pet­er­sen et al., 2001, 2005), indi­cat­ing that
hydro­car­bons could be a local organic mat­ter source in
deep parts of the Red River delta’s ground­wa­ter envi­ron­
ment.
The ele­vated ground­wa­ter salin­ity observed in the
stud­ied tran­sect (Fig. 4B), indi­cates that in the Red
River delta saline ground­wa­ter occurs far inland from
the coast and ren­ders some of the ground­wa­ter that is
low in As to be unsuited for domes­tic use. In the Ben­
gal delta, mod­ern intru­sion causes a high ground­wa­ter
salin­ity mainly in the coastal region, while saline water
far from the coast is restricted to local pock­ets (Ahmed
et al., 2004).
4.4. Pal­ae­o-hydrol­ogy in the Red River delta
The course of the Red River dur­ing the sea level low,
and at pres­ent, appears con­trolled by the Nam Dinh and
Chay fault struc­tures (Fig. 1B) (Tran et al., 2002) and, sim­
i­larly, other sur­face water chan­nels in the study area are
aligned along fault struc­tures. The cen­tre of sub­si­dence in
the delta has shifted towards the SW from a pre-Holo­cene

loca­tion north of the Vinh Ninh fault (Tran et al., 2002). To
the north, the tran­sect extends into the study area of Lam
and Boyd (2003) (Fig. 1A), who observed a rel­a­tively high
ele­va­tion (>¡36 m) of the Pleis­to­cene marker hori­zon in
the north-east­ern delta plain. The few avail­able sed­i­ment
14
C dates from below the marker hori­zon (Figs. 1B and 2)
sig­nif­i­cantly pre-date the Holo­cene (Lam and Boyd, 2003).
These obser­va­tions sug­gest a rel­a­tively sta­ble position of
the main chan­nel in the south­ern part of the delta dur­ing
Late Pleis­to­cene–Holo­cene. Tec­tonic char­ac­ter­is­tics appear
to have pre­vented the Red River from mean­der­ing over the
delta plain dur­ing the last part of the sea level low, and
has thereby per­mit­ted age­ing of the depos­its in the cen­tral
and north­ern delta region.
Rav­ens­croft et al. (2005) pro­posed for their Ben­gal
delta study area, that the 120 m deep inci­sion of the pal­
ae­o-riv­ers dur­ing the sea level low caused steep hydrau­lic
gra­di­ents along with the devel­op­ment of a thick unsat­u­
rated zone in the Pleis­to­cene sed­i­ments. Steep gra­di­ents
will increase the ground­wa­ter flow veloc­ity and thereby


3124

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

pro­mote the flush­ing of labile As from the sed­i­ment. An
enhanced flux of oxy­gen­ated water and the unsat­u­rated
con­di­tions will also pro­mote oxi­da­tion of sed­i­men­tary

organic mat­ter and immo­bi­lize As in re­crys­tal­lized, more
sta­bile Fe-oxides. There­fore, in the Ben­gal delta the oxi­
dized Pleis­to­cene sed­i­ments has a smaller As con­tent (and
a smaller oxa­late extract­able pro­por­tion of the total As)
than the Holo­cene sed­i­ments (Ahmed et al., 2004; Swartz
et al., 2004; Stol­len­werk et al., 2007). How­ever, in some
cases a high As con­tent in Pleis­to­cene sed­i­ments has also
been reported (Shah, 2008). Because of the sed­i­men­to­log­i­
cal sim­i­lar­i­ties, a com­pa­ra­ble pal­ae­o-hydro­log­i­cal con­trol
on the As con­tent of ground­wa­ter in the Red River delta
might be expected. The deep­est ele­va­tion of the bot­tom
of the incised val­ley in the pres­ent tran­sect is only about
¡54 m, and the lower chan­nel slope dur­ing the sea level
low, when the shore line moved east­ward to the shelf
edge south of Ha­inan island (inset in Fig. 1A), makes it
dif
­fi­cult to argue that the incised val­ley in the tran­sect
was much deeper than 50–60 m. Con­di­tions required for a
thick unsat­ur­ ated zone to develop are a com­bi­na­tion of a
low recharge and a high trans­mis­siv­ity of the sed­i­ments.
The thick coarse grained depos­its in the NE part of the
tran­sect have a high trans­mis­siv­ity, and these depos­its
have a regional exten­sion in the cen­tral part of the delta
(Ma­thers and Zal­asiewicz, 1999). Low mete­oric recharge
is likely under the cli­matic arid con­di­tions which pre­
vailed in parts of the cold gla­cial phases (Liew et al., 1998;
Rost, 2000; Zheng and Li, 2000; Jian et al., 2001). There­
fore, in the Red River delta, ground­wa­ter flow may have
occurred towards the high trans­mis­siv­ity regions, rather
than towards incised pal­ae­o-chan­nels, giv­ing rise to deep

atmo­spheric oxi­diz­ing con­di­tions in the cen­tral part of
the delta. This may have caused the observed deep yel­low­
ish col­our­ing of the Pleis­to­cene depos­its in the NE part
of the tran­sect (see Sec­tion 4.1). Alter­na­tively, the allu­vial
depos­its have been oxi­dized from the start, in a sed­i­men­
to­log­i­cal regime fea­si­ble for co-depo­si­tion of only lim­ited
and low-reac­tiv­ity organic mat­ter, in that way pre­serv­ing
the oxi­dized col­our­ing.
4.5. Seques­tra­tion of As
Fig. 7 sug­gests ongo­ing pre­cip­i­ta­tion of iron sulp­hides
which in estu­a­rine, sul­phi­dic envi­ron­ments has a large As

Fig. 9. The mea­sured con­cen­tra­tion of ground­wa­ter As plot­ted against
the cal­cu­lated con­cen­tra­tion of miss­ing SO4 (see text). Note the log­a­rith­
mic axis used for As.

sorp­tion capac­ity (Bo­stick et al., 2004). The con­cen­tra­tion
of sea­wa­ter derived SO4 in a sam­ple is cal­cu­lated as the Cl
con­cen­tra­tion in the sam­ple mul­ti­plied by the SO4/Cl ratio
in oce­anic sea­wa­ter (29.5 mM/566 mM = 0.052). The result
of this cal­cu­la­tion is shown in Fig. 8, where the grey and
the black dots indi­cate, respec­tively, the sea­wa­ter derived
SO4 and the mea­sured ground­wa­ter SO4 con­cen­tra­tions.
Except for the sam­ple from the shal­low screen in Q131, all
sam­ples have a def­i­cit of SO4 (Fig. 8), which must be due to
the reduc­tion of SO4 to sul­phide.
Fig. 9 shows the con­cen­tra­tion of As vs. ‘miss­ing SO4’,
the lat­ter cal­cu­lated as the dif­fer­ence between the mea­
sured SO4 and sea­wa­ter derived SO4. The As con­cen­tra­tion
decreases when the amount of miss­ing SO4 increases, sug­

gest­ing As seques­tra­tion in Fe sulp­hides (Fig. 9). This trend
is not fol­lowed by the sam­ple from Q85 (61 mbs) which
had an excep­tion­ally high CH4 con­cen­tra­tion (Fig. 5E) and
appears to be an out­lier (see Sec­tion 4.3). In Fig. 9, sam­ples
that plot close to the 0.013 lM detec­tion limit for As and
which are inde­pen­dent of miss­ing SO4 (open dia­monds)
are from the Pleis­to­cene aqui­fer in the NE part of the tran­

Fig. 8. The dis­tri­bu­tion of the cal­cu­lated sea­wa­ter derived SO4 con­cen­tra­tion (grey) and the mea­sured (pres­ent) con­cen­tra­tion of SO4 (black) (Seaw. = sea­
wa­ter).




S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

sect and from shal­low screens, rep­re­sent­ing super­fi­cial
ground­wa­ter which may have a lower As con­cen­tra­tion
(c.f., Post­ma et al., 2007) (For the dis­tri­bu­tion of ground­
wa­ter As see Fig. 6). It should be noted, that because most
of the sea­wa­ter derived SO4 has in fact been reduced, the
rela­tion­ship between As and ‘sea­wa­ter derived SO4’ is
sim­i­lar to that shown in Fig. 9 for As and ‘miss­ing SO4’,
infer­ring that a con­trol on ground­wa­ter As might exist
even with­out the pre­cip­it­ a­tion of Fe sulp­hides. Also, the
Stiff dia­grams (Fig. 4C) for the Holo­cene aqui­fer indi­cate
­fresh­wa­ter sig­na­tures and Cl (and SO4) in these sam­ples
may be derived from sea spray from the SE mon­soon. The
Cl con­cen­tra­tions in the fresh­wa­ter in the tran­sect is gen­
er­ally an order of mag­ni­tude higher than in fresh ground­

wa­ter from upstream Ha­noi (<0.2 mM; Lar­sen et al., 2008).
Nev­er­the­less, the sat­u­ra­tion for dis­or­dered mack­in
­ aw­ite
(Fig. 7) com­bined with the drop in ground­wa­ter As con­
cen­tra­tions with the increas­ing amount of miss­ing SO4
(Fig. 9) indi­cate that As seques­tra­tion in Fe sulp­hides may
con­trol ground­wa­ter As con­cen­tra­tions. This sug­gests that
a dis­tinc­tion between As con­trol in brack­ish-marine influ­
enced geo­log­ic­ al set­tings and fresh­wa­ter set­tings is made
in future stud­ies.
The super­sat­u­ra­tion of the ground­wa­ter for sid­er­ite
(Fig. 7) indi­cates that sid­er­ite could also be a sink for As.
Seng­upta et al. (2004) iden­ti­fied As in sid­er­ite con­cre­tions
in aqui­fers of the Ben­gal delta. How­ever, water chem­is­try
data sug­gest that sid­er­ite pre­cip­it­ a­tion removes pro­por­
tion­ally more Fe2+ than As, i.e., that the As/Fe ratio of the
pre­cip­it­ ate is smaller than the As/Fe ratio of the solu­tion
(Del­e­mos et al., 2006; Post­ma et al., 2007).
5. Con­clu­sions
A sequence strati­graphic inter­pre­ta­tion of bore­hole
descrip­tions and geo­phys­i­cal logs was con­ducted. The SW
part of the tran­sect con­sists of grey estu­ar­ ine clays and del­
taic sands, depos­ited in a Holo­cene incised val­ley. The NW
part of the tran­sect is dom­in
­ ated by Pleis­to­cene depos­its
of yel­low­ish allu­vial grav­els and sand under­neath depos­
its of estu­a­rine clay, over­lain by a thin Holo­cene sed­im
­ ent
veneer. A review of the lit­er­a­ture indi­cates that the Holo­
cene incised val­ley, inter­sected by the stud­ied tran­sect,

com­prises the dom­i­nant part of the Qua­ter­nary sequence
along the south­ern bound­ary of the Red River delta, and
that the Pleis­to­cene allu­vial sed­i­ments, found in the NE
part of the tran­sect, are dom­in
­ ant in the cen­tral delta
plain.
The ground­wa­ter in the stud­ied tran­sect is anoxic, and
gen­er­ally con­tains Fe2+, sul­phide, NH4 and CH4. Rel­a­tive to
the lev­els of SO4, NH4 and CH4 observed in the NE part of
the tran­sect, higher con­cen­tra­tions of NH4 and CH4, and
low con­cen­tra­tions of SO4, are found in the SW part of the
tran­sect, indi­cat­ing that the ground­wa­ter redox con­di­tions
are related to the geo­log­ic­ al age of the sed­i­ments. Sim­i­
larly, higher ground­wa­ter As con­cen­tra­tions are found in
the SW part of the tran­sect (up to 11.7 lM or nearly 900 lg/
L), com­pared to the NE part of the tran­sect (60.32 lM).
The con­cen­tra­tions of Fe2+ and H2S are con­trolled by
the pre­cip­i­ta­tion of dis­or­dered mack­in
­ aw­ite and sid­er­
ite, as indi­cated by PHRE­EQC-2 spe­ci­at­ ion cal­cu­la­tions. A

3125

neg­a­tive cor­re­la­tion between the ground­wa­ter As con­cen­
tra­tions and a Cl-based esti­mate of reduced SO4 indi­cate,
when com­bined with the observed equi­lib­rium con­di­tion
for dis­or­dered mack­i­naw­ite, that Fe sulp­hides are a sink
for As in the tran­sect.
Water type sig­na­tures and the for­ma­tion elec­tri­cal con­
duc­tiv­ity logs indi­cate that the SO4 orig­i­nates from sea­wa­

ter intrud­ing dur­ing the mid-Holo­cene trans­gres­sion of the
delta (i.e., from the marine ter­races cov­er­ing the delta area)
and entrapped sea­wa­ter in estu­a­rine sed­i­ment units.
Acknowl­edge­ments
This study has been con­ducted with a grant from DAN­
IDA. We are espe­cially grate­ful to Michael Berg for shar­
ing with us the results from the EA­WAG / CETASD sur­vey
at an early stage; our sam­pling wells were selected based
on his data. We thank Per Jen­sen (GEUS) for skill­fully
under­tak­ing the geo­phys­i­cal log­ging and Tor­ben Dolin
(TUD) for assist­ing with the art­work. We also thank two
anon­y­mous review­ers for com­ments which improved the
man­u­script.
Ref­er­ences
Ahmed, K.M., Bhat­tach­arya, P., Ha­san, M.A., Akh­ter, S.H., Alam, S.M.M.,
Bhuy­ian, M.A.H., Imam, M.B., Khan, A.A., Sracek, O., 2004. Arsenic
enrich­ment in ground­wa­ter of the allu­vial aqui­fers in Ban­gla­desh: an
over­view. Appl. Geo­chem. 19, 181–200.
Akai, J., Iz­umi, K., Fuku­ha­ra, H., Ma­su­da, H., Nak­ano, S., Yo­shim­ura, T.,
Ohfuji, H., Ana­war, H.M., Ku­ru­mi Akai, K., 2004. Min­er­al­og­i­cal and
geo­mi­cro­bi­o­log­i­cal inves­ti­ga­tions on ground­wa­ter arsenic enrich­
ment in Ban­gla­desh. Appl. Geo­chem. 19, 215–230.
Amini, M., Ab­bas­pour, K.C., Berg, M., Win­kel, L., Hug, S.J., Hoe­hn, E.,
Yang, H., John­son, C.A., 2008. Sta­tis­ti­cal mod­el­ling of global ge­o­gen­ic
arsenic con­tam­i­na­tion in ground­wa­ter. Envi­ron. Sci. Tech­nol. 42.
Bad­loe, C., Ngu­yen, T.P.T., Ngu­yen, Q.H., 2004. Random sur­vey of arsenic
con­tam­i­na­tion in tube­well water of 12 prov­inces in Viet­nam and ini­
tially human health arsenic risk assess­ment through food chain. In:
Proc. Third Sci­en­tific Conf. Ha­noi Uni­ver­sity of Sci­ence, Mul­ti­dis­ci­plin­
ary Sci­en­tific Ses­sion “Envi­ron­men­tal Sci­ence – Tech­nol­ogy and Sus­

tain­able Devel­op­ment”, 16 Novem­ber, CET­ASD, Ha­noi, Viet­nam.
Berg, M., Tran, H.C., Ngu­yen, T.C., Viet, P.H., Scher­ten­leib, R., Giger, W., 2001.
Arsenic con­tam­i­na­tion of ground­wa­ter and drink­ing water in Viet­nam:
a human health threat. Envi­ron. Sci. Tech­nol. 35, 2621–2626.
Berg, M., Sten­gel, C., Trang, P.T.K., Viet, P.H., Samp­son, M.L., Leng, M., Sam­
reth, S., Fred­er­icks, D., 2007. Mag­ni­tude of arsenic pol­lu­tion in the
Me­kong and Red River del­tas – Cam­bo­dia and Viet­nam. Sci. Total Envi­
ron. 372, 413–425.
Bo­stick, B.C., Chen, C., Fen­dorf, S., 2004. Arse­nite reten­tion mech­a­nisms
within estu­ar­ ine sed­i­ments of Pes­ca­der­o, CA. Envi­ron. Sci. Tech­nol.
38, 3299–3304.
Boyd, W., Lam, D.D., 2004. Holo­cene ele­vated sea lev­els on the north coast
of Viet­nam. Aust. Ge­ogr. Stud. 42, 77–88.
Char­let, L., Polya, D.A., 2006. Arsenic in shal­low, reduc­ing ground­wa­ters in
south­ern Asia: an envi­ron­men­tal health disas­ter. Ele­ments 2, 91–96.
Cline, J.D., 1967. Spec­tro­pho­to­met­ric deter­mi­na­tion of hydro­gen sul­fide
in nat­u­ral waters. Lim­nol. Oce­a­nogr. 14, 454–458.
Del­e­mos, J.L., Bo­stick, B.C., Ren­shaw, C.E., Stü­rup, S., Feng, X., 2006. Land­
fill-stim­u­lated iron reduc­tion and arsenic release at the Coak­ley
Super­fund Site (NH). Envi­ron. Sci. Tech­nol. 40, 67–73.
Funa­bi­ki, A., Har­uy­ama, S., Ngu­yen, V.Q., Viet, P.H., Dinh, H.T., 2007. Holo­
cene delta plain devel­op­ment in the Song Hong (Red River) delta, Viet­
nam. J. Asian Earth Sci. 30, 518–529.
Gani, M.R., Alam, M.M., 2004. Flu­vial facies ar­chi­tech­ture in small-scale
river sys­tems in the Upper Dupi Tila For­ma­tion, north­east Ben­gal
Basin, Ban­gla­desh. J. Asian Earth Sci. 24, 225–236.
Good­bred, S.L., Ku­ehl, S.A., 2000. The sig­nif­i­cance of large sed­i­ment sup­
ply, active tec­to­nism, and eu­sta­sy on mar­gin sequence devel­op­ment:
late qua­ter­nary stra­tig­ra­phy and evo­lu­tion of the Gan­ges–Brah­mapu­
tra delta. Sed­i­ment. Geol. 133, 227–248.

Guil­lot, S., Char­let, L., 2007. Ben­gal arsenic, an archive of Hima­laya orog­
eny and pa­leo­hy­drol­o­gy. J. Envi­ron. Sci. Health, A 42, 1785–1794.


3126

S. Jes­sen et al. / Applied Geochemistry 23 (2008) 3116–3126

Hane­buth, T.J.J., Sa­i­to, Y., Tan­ab
­ e, S., Quang, L.V., Quang, T.N., 2006. Sea
lev­els dur­ing late marine iso­tope stage 3 (or older?) reported from
the Red River delta (north­ern Viet­nam) and adja­cent regions. Qua­
tern. Int. (145/146), 119–134.
Har­vey, C.F., Swartz, C.H., Bad­ruzz­aman, A.B.M., Keon-Blute, N., Yu, W.,
Ali, M.A., Jay, J., Bec­kie, R., Nie­dan, V., Brab­an­der, D., Oates, P.M., Ashf­
aque, K.N., Islam, S., He­mond, H.F., Ahmed, M.F., 2005. Ground­wa­ter
arsenic con­tam­i­na­tion on the Gan­ges Delta: bio­geo­chem­is­try, hydrol­
ogy, human per­tur­ba­tions, and human suffering on a large scale. C. R.
Geo­sci. 337, 285–296.
Hori, K., Tan­a­be, S., Sa­i­to, Y., Har­uy­ama, S., Viet, N., Ki­tam­ura, A., 2004. Delta
ini­ti­a­tion and Holo­cene sea-level change: exam­ple from the Song Hong
(Red River) delta, Viet­nam. Sed­i­ment. Geol. 164, 237–249.
Hoss­ain, F., Hill, J., Bag­tzog­lou, A.C., 2007. Geo­sta­tis­ti­cal­ly based man­age­
ment of arsenic con­tam­i­nated ground water in shal­low wells of Ban­
gla­desh. Water Re­sour. Man­age. 21, 1245–1261.
Islam, F.S., Ga­ult, A.G., Booth­man, C., Polya, D.A., Char­nock, J.M., Chat­ter­
jee, D., Lloyd, J.R., 2004. Role of metal-reduc­ing bac­te­ria in arsenic
release from Ben­gal delta sed­i­ments. Nature 430, 68–71.
Jian, Z., Hu­ang, B., Ku­hnt, W., Lin, H.-L., 2001. Late Qua­ter­nary upwell­ing
inten­sity and East Asian mon­soon forc­ing in the South China Sea. Qua­

tern. Res. 55, 336–370.
Kit­az­a­wa, T., 2007. Pleis­to­cene mac­ro­tid­al tide-dom­i­nated estu­ary–delta
suc­ces­sion, along the Dong Nai River, south­ern Viet­nam. Sed­i­ment.
Geol. 194, 115–140.
Lam, D.D., Boyd, W.E., 2003. Holo­cene cos­tal stra­tig­ra­phy and the sed­i­
men­tary devel­op­ment of the Hai Phong area of the Bac Bo plain (Red
River delta), Viet­nam. Aust. Ge­ogr. 34, 177–194.
Lam­beck, K., Yo­koy­ama, Y., Pur­cell, T., 2002. Into and out of the last gla­cial
max­i­mum: sea-level change dur­ing oxy­gen iso­tope stages 3 and 2.
Qua­tern. Sci. Rev. 21, 343–360.
Lang­muir, D.L., Ma­hon­ey, J., Row­son, J., 2006. Sol­u­bil­ity prod­ucts of amor­
phous fer­ric arse­nate and crys­tal­line sco­ro­dite (Fe­AsO4·2H2O) and
their appli­ca­tion to arsenic behav­ior in bur­ied mine tail­ings. Geo­
chim. Cos­mo­chim. Acta 70, 2942–2956.
Lar­sen, F., Pham, N.Q., Dang, N.D., Post­ma, D., Jes­sen, S., Pham, V.H., Ngu­
yen, T.B., Trieu, H.D., Tran, L.T., Ngu­yen, H., Cham­bon, J., Ngu­yen. H.V.,
Ha, D. H., Hue, N.T, Duc, M.T., Ref­sg­aard, J.C., 2008. Con­trol­ling geo­
log­i­cal and hydro­geo­log­i­cal pro­cesses in an arsenic con­tam­i­nated
aqui­fer on the Red River flood plain, Viet­nam. Appl. Geo­chem. 23(11),
3099-3115.
Li, Z., Sa­i­to, Y., Mat­sum­ot­o, E., Wang, Y., Tan­a­be, S., Quang, L.V., 2006. Cli­
mate change and human impact on the Song Hong (Red River) delta,
Viet­nam, dur­ing the Holo­cene. Qua­tern. Int. 144, 4–28.
Liew, P.M., Kuo, C.M., Hu­ang, S.Y., Tseng, M.H., 1998. Veg­e­ta­tion change
and ter­res­trial car­bon stor­age in east­ern Asia dur­ing the last gla­cial
max­i­mum as indi­cated by a new pol­len record from cen­tral Tai­wan.
Global Planet. Change (16/17), 85–94.
Low­ers, H.A., Bre­it, G.N., Fos­ter, A.L., Whit­ney, J., You­nt, J., Ud­din, Md.N.,
Mu­neem, Ad.A., 2007. Arsenic incor­po­ra­tion into authi­genic pyrite,
Ben­gal Basin sed­i­ment, Ban­gla­desh. Geo­chim. Cos­mo­chim. Acta 71,

2699–2717.
Ma­thers, S., Zal­asiewicz, J., 1999. Holo­cene sed­i­men­tary archi­tec­ture of
the Red River delta Viet­nam. J. Coast. Res. 15, 314–325.
McAr­thur, J.M., Rav­ens­croft, P., Safiu­lla, S., Thirl­wall, M.F., 2001. Arsenic in
ground­wa­ter: test­ing pol­lu­tion mech­a­nisms for sed­i­men­tary aqui­fers
in Ban­gla­desh. Water Re­sour. Res. 37, 109–117.
Park­hurst, D.L., Ap­pe­lo, C.A.J., 1999. User’s guide to PHRE­EQC (Ver­sion
2) – a com­puter pro­gram for spe­ci­a­tion, reac­tion-path, 1D-trans­port,
and inverse geo­chem­i­cal cal­cu­la­tions. U.S Geol. Surv. Water Re­sour.
Invest. Rep. 99–4259.
Pet­er­sen, H.I., Ander­sen, C., Anh, P.H., Bo­je­sen-Koef­oed, J.A., Niel­sen, L.H.,
Nyt­oft, H.P., Rosen­berg, P., Thanh, L., 2001. Petro­leum potential of Oli­
go­cene lacus­trine mud­stones and coals at Dong Ho, Viet­nam – an out­
crop ana­logue to ter­res­trial source rocks in the greater Song Hong
Basin. J. Asian Earth Sci. 19, 135–154.
Pet­er­sen, H.I., Vu, T., Niel­sen, L.H., Ngu­yen, A.D., Nyt­oft, H.P., 2005. Source
rock prop­er­ties of lacus­trine mud­stones and coals (Oli­go­cene Dong
Ho For­ma­tion), onshore Song Hong Basin, north­ern Viet­nam. J. Pet­rol.
Geol. 28, 19–38.
Polya, D.A., Ga­ult, A.G., Di­eb­e, N., Feld­man, P., Ro­sen­boom, J.W., Gil­li­gan,
E., Fred­er­icks, D., Mil­ton, A.H., Samp­son, M., Row­land, H.A.L., Lyth­goe,
P.R., Jones, J.C., Mid­dle­ton, C., Cooke, D.A., 2005. Arsenic haz­ard in
shal­low Cam­bo­dian ground­wa­ters. Min­eral. Mag. 69, 807–823.
Post­ma, D., Lar­sen, F., Hue, N.T.M., Duc, M.T., Viet, P.H., Nhan, P.Q., Jes­sen,
S., 2007. Arsenic in ground­wa­ter of the Red River flood­plain, Viet­nam:
con­trol­ling geo­chem­i­cal pro­cesses and reac­tive trans­port mod­el­ing.
Geo­chim. Cos­mo­chim. Acta 71, 5054–5071.

Rav­ens­croft, P., McAr­thur, J.M., Ho­que, B.A., 2001. Geo­chem­i­cal and pal­
a­eohy­dro­log­i­cal con­trols on pol­lu­tion of ground­wa­ter by arsenic. In:

Chap­pell, W.R., Ab­ern­a­thy, C.O., Cald­er­on, R.L. (Eds.), Fourth Int. Conf.
Arsenic Expo­sure and Health Effects. Else­vier Sci­ence, Ltd, Oxford.
Rav­ens­croft, P., Bur­gess, W.G., Ahmed, K.M., Bur­ren, M., Per­rin, J., 2005.
Arsenic in ground­wa­ter of the Ben­gal Basin, Ban­gla­desh: dis­tri­bu­
tion, field rela­tions, and hydro­geo­log­i­cal set­ting. Hy­dro­ge­ol. J. 13,
727–751.
Rod­rí­guez Lado, L., Polya, D., Win­kel, L., Berg, M., Hegan, A., 2008. Mod­
el­ling arsenic haz­ard in Cam­bo­dia: A geo­sta­tis­ti­cal approach using
ancil­lary data. Appl. Geo­chem. 23(11), 3010-3018.
Rost, K.T., 2000. Pleis­to­cene pa­leo­en­vi­ron­men­tal changes in the high
moun­tain ranges of cen­tral China and adja­cent regions. Qua­tern. Int.
(65/66), 147–160.
Row­land, H.A.L., Polya, D.A., Lloyd, J.R., Pan­cost, R.D., 2006. Char­ac­ter­isa­
tion of organic mat­ter in a shal­low, reduc­ing, arsenic-rich aqui­fer,
West Ben­gal. Org. Geo­chem. 37, 1101–1114.
Row­land, H.A.L., Ped­er­ick, R.L., Polya, D.A., Pan­cost, R.A., van Don­gen, B.E.,
Ga­ult, A.G., Vaughan, D.J., Bry­ant, C., Ander­son, B., Lloyd, J.R., 2007.
The con­trol of organic mat­ter on mi­cro­bi­al­ly med­ia
­ ted iron reduc­tion
and arsenic release in shal­low allu­vial aqui­fers, Cam­bo­dia. Geo­bi­ol­
o­gy 5, 281–292.
Seng­upta, S., Muk­her­jee, P.K., Pal, T., Shome, S., 2004. Nature and ori­gin of
arsenic car­ri­ers in shal­low aqui­fer sed­i­ments of Ben­gal Delta, India.
Envi­ron. Geol. 45, 1071–1081.
Shah, B.A., 2008. Role of Qua­ter­nary stra­tig­ra­phy on arsenic-con­tam­
i­nated ground­wa­ter from parts of Mid­dle Ganga Plain, UP–Bi­har,
India. Envi­ron. Geol. 53, 1553–1561.
Smed­ley, P.L., Kinni­burgh, D.G., 2002. A review of the source, behav­iour and
dis­tri­bu­tion of arsenic in nat­u­ral waters. Appl. Geo­chem. 17, 517–568.
Stan­ger, G., 2005. A pal­ae­o-hydro­geo­log­i­cal model for arsenic con­tam­

i­na­tion in south­ern and south-east Asia. Envi­ron. Geo­chem. Health
27, 359–367.
Stol­len­werk, K.G., Bre­it, G.N., Welch, A.H., You­nt, J.C., Whit­ney, J.W., Fos­
ter, A.L., Ud­din, M.N., Ma­jum­der, R.K., Ahmed, N., 2007. Arsenic atten­u­
a­tion by oxi­dized aqui­fer sed­i­ments in Ban­gla­desh. Sci. Total Envi­ron.
379, 133–150.
Stoo­key, L.L., 1970. Fer­ro­zine – a new spec­tro­pho­to­met­ric reagent for
iron. Anal. Chem. 42, 779–781.
Stumm, W., Mor­gan, J.J., 1981. Aquatic Chem­is­try, sec­ond ed. Wiley &
Sons, New York.
Swartz, C.H., Blute, N.K., Bad­ruzz­man, B., Ali, A., Brab­an­der, D., Jay, J.,
Be­san­con, J., Islam, S., He­mond, H.F., Har­vey, C.F., 2004. Mobil­ity of
arsenic in a Ban­gla­desh aqui­fer: Infer­ences from geo­chem­i­cal pro­
files, leach­ing data, and min­er­al­og­i­cal char­ac­ter­iza­tion. Geo­chim. Cos­
mo­chim. Acta 68, 4539–4557.
Tan­a­be, S., Hori, K., Sa­i­to, Y., Har­uy­ama, S., Le, Q.D., Sato, Y., Hira­ide, S.,
2003a. Sed­i­men­tary facies and radio­car­bon dates of the Nam Dinh-1
core from the Song Hong (Red River) delta, Viet­nam. J. Asian Earth
Sci. 21, 503–513.
Tan­a­be, S., Hori, K., Sa­i­to, Y., Har­uy­ama, S., Van, P.V., Ki­tam­ura, A., 2003b.
Song Hong (Red River) delta evo­lu­tion related to mil­len­nium-scale
Holo­cene sea-level changes. Qua­tern. Sci. Rev. 22, 2345–2361.
Tan­a­be, S., Sa­i­to, Y., Quang, L.V., Hane­buth, T.J.J., Quang, L.N., Ki­tam­ura, A.,
2006. Holo­cene evo­lu­tion of the Song Hong (Red River) delta sys­tem,
north­ern Viet­nam. Sed­i­ment. Geol. 187, 29–61.
Tran, N., Ngo, Q.T., Do, T.V.T., Ngu­yen, D.M., Ngu­yen, V.V., 1991. Qua­ter­
nary sed­im
­ en­ta­tion of the prin­ci­pal del­tas of Viet­nam. J. South­east
Asian Earth Sci. 6, 103–110.
Tran, N., Mai, T.N., Chu, V.N., Hoek­stra, P., Weer­ing, V.Tj., van den Bergh,

J.H., Dinh, X.T., Ngu­yen, D.N., Vu, V.P., 2002. Holo­cene sed­i­men­tary
evo­lu­tion, geo­dy­namic and anthro­po­genic con­trol of the Ba­lat river
mouth for­ma­tion (Red River-delta, north­ern Viet­nam). Z. Geol. Wiss.,
Ber­lin 30, 157–172.
Tray­nor, J.J., Sla­den, C., 1997. Seep­age in Viet­nam – onshore and off­shore
exam­ples. Mar. Pet­rol. Geol. 14, 345–362.
Vu, T.C., 1996. Salin­ity intru­sion in the Red River delta. Sem­i­nar on Envi­
ron­ment and Devel­op­ment in Viet­nam, Decem­ber 6-7, Aus­tra­lian
National Uni­ver­sity. (See coo­mbs.anu.edu.au/~vern/env_dev/sem­i­
nar96.html).
Win­kel, L., Berg, M., Amini, M., Hug, S.J., John­son, C.A., 2008. Pre­dict­ing
ground­wa­ter arsenic con­tam­i­na­tion in South­east Asia from sur­face
param­e­ters. Nature Geo­sci. 1, 536–542.
Yo­koy­ama, Y., Lam­beck, K., Deck­ker, P.D., John­ston, P., Fi­field, L.K., 2000.
Tim­ing of the last gla­cial max­i­mum from observed sea-level min­ima.
Nature 406, 713–716.
Zheng, Z., Li, Q., 2000. Veg­e­ta­tion, cli­mate, and sea level in the past 55,000
years, Han­ji­ang Delta, south­east­ern China. Qua­tern. Res. 53, 330–340.