Origin of arsenic in Late Pleistocene to Holocene sediments in the Nawalparasi district (Terai, Nepal)

Full text

ORIGINAL ARTICLE
Origin of arsenic in Late Pleistocene to Holocene sediments
in the Nawalparasi district (Terai, Nepal)
Ste
´phane Guillot
1
•Marion Garc¸on
1
•Beth Weinman
2
•Ananta Gajurel
3
•
Delphine Tisserand
1
•Christian France-Lanord
4
•Alex van Geen
5
•
Sudipta Chakraborty
6
•Pascale Huyghe
1
•Bishal Nath Upreti
3
•Laurent Charlet
1
Received: 9 July 2014 / Accepted: 5 March 2015
ÓSpringer-Verlag Berlin Heidelberg 2015
Abstract A sedimentological and geochemical study was
carried out to explore the origin of arsenic contamination in
sediments in Nawalparasi district, in the western Terai of
Nepal. The investigation tools include major, trace and rare
earth element analyses of core sediments, as well as
14
C
datings, and O, C isotopic analyses on mollusk shells. The
results show that black schists from the Lesser Himalaya
highly contributed to the detrital input in Parasi during the
Pleistocene–Holocene transition because of focused ero-
sion related to rapid uplift and high rainfall along the Main
Central Thrust zone. In addition, aquifer silts, sands, and
most of the brown clays underwent a certain degree of
chemical weathering and physical reworking, and show
possible inputs from the Siwaliks during the Late Holo-
cene. A possible correlation between late Quaternary cli-
mate regimes and the concentration of arsenic in sediments
is suspected, with arsenic preferentially concentrated dur-
ing the drier periods of the last 25 kyr BP. The process of
arsenic eluviations in sandy and silty sediments can explain
the lower arsenic concentrations in sediments during humid
periods. During the drier periods, seasonal precipitation
was smaller and temperature was lower, leading to wet
(less evaporative) soils in swampy environments. This
environment favoured the development of aquatic plants
and bacteria growing within in the moist land areas, en-
hancing the strong weathering of initially suspended load
particles (micas and clays), which were preferentially de-
posited in quiet hydraulic environments. These sorting and
weathering processes presumably allowed the arsenic to be
concentrated in the finest sediment fraction.
Keywords Sediments Geochemistry Arsenic 
Climate Terai Nepal
Introduction
Arsenic contamination in groundwater is a growing crisis
in parts of South Asia, including Nepal, Bangladesh, India
and Pakistan where inhabitants living on alluvial plains of
the Himalayan foreland basins depend heavily on ground-
water for their daily needs. In this region, arsenic has a
geogenic source and its elevated concentrations are con-
sidered to be due to natural weathering of the Himalayan
belt (Acharyya et al. 2000; Gurung et al. 2005; Guillot and
Charlet 2007). In fact, the sediments carried by the Ganga–
Brahmaputra river system mainly contributed to the filling
of the Himalayan foreland basin and of the Bengal fan,
which is considered to be one of the largest modern fluvial
deltas of the world (France-Lanord et al. 1993; Garzanti
et al. 2004). The amount of sediments carried by the
Ganga–Brahmaputra rivers from the Himalayan belt to the
delta is estimated at 1800 tonnes/km
2
with suspended loads
estimated between 540 and 1175 millions tonnes/year
&Ste
´phane Guillot
stephane.guillot@ujf-grenoble.fr
1
ISTerre, University of Grenoble Alpes, CNRS,
38031 Grenoble, France
2
Earth and Environmental Sciences, California State
University, Fresno, USA
3
Institute of Science and Technology, Tribhuvan University,
Kirtipur, Kathmandu, Nepal
4
Centre de Recherches Pe
´trographiques et Ge
´ochimiques,
CNRS, 54501 Vandoeuvre le
`s Nancy, France
5
Lamont-Doherty Earth Observatory of Columbia University,
Palisades, New York, USA
6
Department of Chemistry, Kanchrapara College,
Kanchrapara 743145, West Bengal, India
123
Environ Earth Sci
DOI 10.1007/s12665-015-4277-y

(Milliman and Sivitsli 1992). The aim of this study is to
understand whether the aquifer sediments, having different
arsenic concentrations, derived from the same initial source
of the Himalaya. The importance of this work defines how
pre-depositional processes, such as variable source terrains
and/or climate, can affect aquifer deposits and their geo-
chemical variability.
In an attempt to identify how sediment sources and
weathering processes can affect local distributions of
arsenic, core sediment samples from the Nawalparasi
District (Terai, Nepal) were collected by boring and were
analysed for major, trace and rare earth elements (REE). In
addition, O and C isotopes were performed on mollusk
shells that were collected during aquifer sampling. Finally,
we contextualize our geochemical findings here to other
relevant sedimentological and geochronological works,
where we reconstruct the history of the aquifer using our
own and other published
14
C datings.
Geological setting
Himalayan geology and hydrology of the Terai plain
Extensively exposed in the Narayani basin (Fig. 1) are all
four major Himalayan tectonic units: (1) the Tethys Hi-
malaya, delimited at the base by the South Tibetan Detach-
ment system (STDS); (2) the Higher Himalayan Crystallines
(HHC) delimited at the base by the Main Central Thrust I
(MCT I); (3) the Lesser Himalaya (LH) divided into upper
and lower Lesser Himalaya, is delimited at the base by the
Main Boundary Thrust (MBT); and (4) the Siwaliks, de-
limitated at its base by the Main Frontal Thrust (MFT) and the
Quaternary foreland basin. These units span a wide range of
different metamorphic, sedimentary, and igneous origins,
making it possible for their differential erosion to account for
some of the groundwater arsenic heterogeneity we see in the
foreland and delta (i.e. van Geen et al. 2008; Gurung et al.
2005; Shah 2008). In the region of provenance of the studied
basin, the Tethys Himalaya includes 10 km of various meta-
sedimentary rocks (limestones, calc-schists, shales, quart-
zites) ranging from Cambrian to Jurassic (Colchen et al.
1986). Metamorphism increases upward from amphibolite-
facies to anchizonal conditions in the Mesozoic sequence.
There is the Manaslu leucogranite emplaced within the
Tethyan rocks (Guillot et al. 1995). The Higher Himalayan
Crystallines are a metamorphic stack, including from base to
top (Colchen et al. 1986) 2–10 km thick paragneisses
(metapelites and metapsammites); *3 km-thick gneisses
with calcsilicate minerals (diopside and amphibole)
and *300–500 m-thick orthogneisses representing meta-
morphosed Lower Paleozoic granites. The Lesser Himalaya
consists of mostly unfossiliferous metasediments with a
lower part consisting of very low to low-grade quartzites and
phyllites (Kuncha Group). The upper part includes some
dolomitic meta-carbonates alternating with dominant black
schists, aluminium rich schists and quartzites. Amphibolites
occur in both of these groupings (Colchen et al. 1986).
The Siwaliks represent the Cenozoic foreland basin of
the Himalayan belt with local thickness of 6 km in Nepal
(Mugnier et al. 1999; Huyghe et al. 2005). They are di-
vided into three units having a typical coarsening-upward
succession. The lower unit consists of fluvial channel
sandstones alternating with calcareous paleosols; the mid-
dle unit consists of very thick channel sandstones with
minor paleosols, and the upper unit mainly hosts con-
glomerates of gravelly braided river deposits.
The southern plains of Nepal, known as the ‘Terai,’ are
remarkably flat, lie between 60 and 360 m above sea level,
and stretch from east to west, in front of the Main Frontal
Thrust (Fig. 1). Geologically, the Terai plain is an active
foreland basin consisting of Quaternary sediments that in-
clude molasse sediments along with gravel, sand, silt and
clay. According to a sedimentation rate of 2 mm/year for
the recent period, the first 50 m of sediment comprising the
shallow aquifers in the region should roughly represent the
last *100,000 years, recording the Late Pleistocene to
Holocene periods. Many rivers feeding sediments into the
Terai flow south from the Himalayan range (Fig. 1). Minor
rivers emanate from the nearby Siwalik Hills and deposit
fine and gravelly sediments in the form of a fan system
(Shukla and Bora 2009; Upreti 1999). In between the fans,
fine sediments and organic material have been depositing
in inter-fluvial lowlands in which present-day wetlands and
swamps are located (Shah 2008).
Although the Terai constitutes less than 20 % of the
Nepal’s surface, it contains over half of the total arable
land and is home to about 50 % of the total Nepalese
population, i.e. 30 millions of inhabitants. Groundwater is
the main source of water for drinking and irrigation in the
Terai area. Over 90 % of the Terai population draws
groundwater from tube wells for drinking, household use,
and irrigation (http://www.gwrdb.gov.np/hydrogeological_
studies.php). An estimated 800,000 shallow (\20 m) tube
wells and over 3000 deep (20–100 m) tube wells exist in
the Terai plain (Pokhrel et al. 2009). High monsoon pre-
cipitation (2000 mm) and summer snow-fed river systems
recharge the Terai sediments giving them high potential for
groundwater resources. Shallow aquifers (\
50 m) are
generally unconfined or semi-confined, whereas the deep
aquifers ([50 m) are mostly confined by clay layers (Gu-
rung et al. 2005). Regional groundwater flows southward
and the shallow aquifers are highly sensitive to precipita-
tion. However, in the Nawalparasi district, only 10 % of
the average rainfall of 2000 mm/year infiltrates and
recharges the groundwater systems (UNDP 1989).
Environ Earth Sci
123

Fig. 1 a Location of Terai
basin, Nepal, in the Ganga–
Brahmaputra watershed areas
(after Gurung et al. 2005).
bGeological map of central
Nepal (after Guillot 1999) with
the location of drilling sites of
this study (yellow box)
including the three drilling sites
studied by Gurung et al. (2005)
(three orange stars)
Environ Earth Sci
123

Study area and lithology
The study area is located in the Indo-Gangetic plain of the
Nawalparasi district in Nepal and comprises the sediments
of the inter-alluvial fan deposits between the Narayani and
Tinau Rivers (Fig. 1). The area is *25 km south from the
frontal mountain chain of the Himalaya. In between the
Tinau and Narayani, small ephemeral rivers originating
from the Siwalik frontal mountains disappear upon enter-
ing into the Indo-Gangetic plain and reappear again close
to the study area. Hence, small natural ponds and river
meanderings were observed as characteristic geomorphic
features of the area. Close to the frontal mountain chain,
the Indo-Gangetic plain consists of boulder- to gravel-sized
sediments, while the sediments of the study area are
dominantly fine-grained sediments. Five wells were
manually drilled down to a depth of about 25 m below
ground level (bgl) in the area covering an east–west tran-
sect of *1 km around the Unwach village between
27°3100100N and 27°3005400N and 83°3904000E and
83°4003000E (Fig. 1b).
Lithology of sledge core samples from the five drill
holes shows various coarse (millimetric) to fine-grained
(micrometric) sediments. We distinguish light-grey to dark
grey sands; grey, greenish-grey to brown-grey and yellow–
brown silts; and light-grey to black-grey, yellow–brown
and black clay with occasional gravel layers. To facilitate
the interpretation of the data, the studied samples have
been categorised into three major sediment facies: (1) grey-
black clays, (2) brown clays and (3) silts and sands
(Table 1). Some calcretes were also observed. Macroscopic
observations show that on average, the drilled sediments
are composed of 33 % of silts; 30 % of grey to black clays,
27 % of brown clay, 9 % of fine-grained silt and sand and
less than 1 % of calcrete. Sands, silt and clay sediments
often contained micas that were occasionally massive to
laminated, bioturbated, and/or also containing roots and
plant debris. Binocular observations show that the detrital
minerals in the silt fraction are dominated by quartz, bi-
otite, muscovite, K-feldspar, calcite and dolomite as major
phases and garnet, zircon, and monazite as heavy minerals.
Abundance of very fine-grained particles is typical of flood
plain sediments along the modern Ganga river system in-
cluding the Narayani river as inferred from present-day
deposits (Galy et al. 1999; Singh 2009). The identified
sediment facies in Parasi’s 25 m shallow subsurface were
interpreted as channel, point bar, overbank, and oxbow lake
deposits of rivers derived from the Siwalik foothills
(Fig. 2).
Muddy sediments with angular to sub-angular gravels
appeared at around 24 m depths are interpreted as channel
deposit (Fig. 2). Three distinct yellow–brown, brown to
yellow–grey paleosol horizons containing calcareous
nodules are some of the more obvious and remarkable
features in the aquifer. The lowermost, *3 m thick pale-
osol at *21 m depth is extensively distributed in the sur-
veyed area. In studying the wash-borings, we found the
paleosols often containing calcareous nodules, or ‘kan-
kars’, which are principally comprised of microcrystalline
calcite with traces of quartz grains (Fig. 2). To compare the
lateral extent of our observed facies, we incorporate
sedimentological results from other studies, such as the
work done by Gurung et al. (2005), located 2–3 km NNE
to NE of our study area (Fig. 1b). Inasmuch, data coming
from nearby sedimentological environments will be inte-
grated in our discussion, to put into context and better
assess the general interest of our findings.
Methods
Major-element data were obtained at the University of
Grenoble using a Perkin-Elmer 3000 DV ICP-OES and
following a procedure similar to that described by Cotten
et al. (1995). The running conditions for the ICP are given
in Chauvel et al. (2011) and calibration of the signal is
performed using a blank and five different dilutions of a
mixed solution of pure elements prepared to mimic the
major-element composition of rock samples.
Trace-element concentrations were obtained using an
Agilent 7500ce ICP-MS. About 100 mg of rock powder
were precisely weighed and dissolved in steel Paar bombs
with two rock standards (BIR-1 and UB-N). Dissolution
was performed in a mixture of high-purity HF and HClO
4
for a minimum of 7 days. After complete dissolution, the
solution was evaporated and the residue was treated in
concentrated HNO
3
and evaporated before dilution in about
40 mL of 7 molL
-1
HNO
3
. A precisely weighed aliquot of
this ‘‘mother’’ solution, corresponding to 8 mg of the
starting powder, was mixed with a spike containing five
elements (Be, Ge, In, Tm, Bi) and diluted in 2 % HNO
3
with traces of HF to reach a dilution factor of about 5000
(only 2500 for BIR-1 and UB-N). This forms the final
solution that is introduced into the ICP-MS.
Measuring the arsenic content of the sedimentary sam-
ples requires a specific treatment because the As concen-
trations of many samples are far higher than those in the
standard used to calibrate the ICP-MS signal. We use a
doping technique in which a known amount of pure arsenic
standard solution is added to an international rock standard
final solution before analysis on the ICP-MS. Most ele-
ments were measured without using the collision cell (the
no-gas mode), but data for eight elements (Sc, Ti, V, Cr,
Co, Ni, Cu and Zn) were acquired using a helium flux
through the collision cell. A rock standard was used to
calibrate the signal and was run every four samples through
Environ Earth Sci
123

Table 1 Geochemical composition in major elements (wt %) of the Nawalparasi borehole sediments and rocks coming from Himalaya (HHC
higher himalayan crystallines, LH lesser himalaya, Siw siwaliks)
Samples Depth Lithology SiO
2
TiO
2
Al
2
O
3
Fe
2
O
3
MnO MgO CaO Na
2
OK
2
O LOI Total
PNS1 8 Silt 81.08 0.41 8.49 2.82 0.09 0.92 2.79 0.67 2.00 0.20 99.48
PNS1 15 Silt 87.07 0.26 3.87 2.32 0.13 0.38 2.84 bdl 1.10 1.20 99.16
PNS1 15 Brown clay 60.46 0.53 10.95 3.96 0.16 2.02 11.19 0.56 2.78 7.30 99.91
PNS1 18 Brown clay 57.14 0.66 14.10 5.42 0.19 1.94 11.33 1.45 3.30 3.80 99.35
PNS 2/10 3 Brown clay 56.03 0.53 18.04 5.81 0.05 1.88 3.06 0.24 3.97 10.01 99.63
PNS 2/20 6 Brown clay 58.88 0.50 19.49 7.66 0.05 1.45 0.42 0.18 3.96 7.19 99.78
PNS 2/25 8 Silt 95.17 0.22 2.43 1.40 0.04 0.05 0.26 0.00 0.79 0.00 100.36
PNS 2/28 8 Grey clay 61.67 0.47 12.77 4.64 0.06 1.70 5.29 1.29 2.74 8.88 99.51
PNS 2/30 9 Silt 79.01 0.57 8.54 3.49 0.17 1.00 2.04 0.00 2.38 1.20 98.39
PNS 2/35 11 Grey clay 64.74 0.62 12.38 4.67 0.41 2.05 5.41 0.00 3.04 5.30 98.63
PNS 2/36 11 Grey clay 64.04 0.54 12.75 4.78 0.13 1.61 3.62 1.24 2.77 7.65 99.14
PNS 2/39 12 Silt 75.20 0.27 9.40 1.94 0.08 0.63 5.78 1.08 2.27 3.20 99.84
PNS 2/50 15 Calcrete 63.29 0.76 15.33 5.54 0.06 2.42 3.85 1.09 3.87 3.60 99.80
PNS 2/52 16 Brown clay 41.82 0.47 11.96 5.01 0.11 1.74 17.00 1.21 3.03 16.48 98.84
PNS 2/57 17 Silt 68.41 0.39 9.32 2.34 0.06 1.17 7.65 1.68 2.24 5.12 98.37
PNS 2/60 18 Silt 62.56 0.47 11.04 2.94 0.05 1.51 10.53 1.90 2.73 6.15 99.88
PNS 2/62 19 Grey clay 51.64 0.55 19.34 6.79 0.22 2.48 2.74 0.34 4.22 10.98 99.30
PNS 2/63 19 Grey clay 53.31 0.54 18.15 7.00 0.03 2.43 0.85 0.58 4.01 12.43 99.33
PNS 2/67 20 Black clay 56.42 0.54 18.04 6.06 0.05 2.32 1.81 0.47 3.89 9.84 99.44
PNS 2/69 21 Brown clay 58.41 0.67 14.91 5.40 0.08 2.55 7.03 0.00 3.21 6.70 98.95
PNS 2/75 23 Brown clay 49.16 0.53 14.85 5.39 0.10 1.96 10.17 0.26 3.24 12.80 98.47
PNS 2/80 24 Black clay 63.35 0.87 18.07 5.50 0.07 2.89 1.32 0.00 4.51 3.10 99.68
PNS 3/5 2 Silt 78.85 0.50 7.75 3.39 0.11 0.90 1.48 bdl 2.20 4.50 99.70
PNS 3/10 3 Brown clay 69.58 0.57 13.51 4.46 0.04 1.20 0.22 0.44 3.21 5.85 99.07
PNS 3/15 5 Silt 74.29 0.70 11.99 4.70 0.11 1.25 0.37 bdl 3.15 2.70 99.27
PNS 3/18 5 Brown clay 66.11 0.44 14.64 6.19 0.04 1.22 0.24 0.42 3.09 6.57 98.96
PNS 3/28 8 Grey clay 63.40 0.59 17.45 5.48 0.04 1.88 0.38 0.47 4.11 4.73 98.53
PNS 3/32 10 Grey clay 52.55 0.62 21.14 6.35 0.03 2.33 0.73 0.22 4.33 11.06 99.37
PNS 3/37 11 Grey clay 58.20 0.52 16.08 6.02 0.13 2.65 3.03 0.55 3.83 8.47 99.48
PNS 3/40 12 Silt 71.71 0.30 9.90 2.17 0.05 1.14 4.25 1.84 2.22 3.85 97.43
PNS 3/45 14 Brown clay 58.05 0.51 11.31 3.98 0.10 2.05 12.55 0.86 2.69 7.40 99.50
PNS 3/46 14 Grey clay 43.79 0.47 11.90 4.72 0.08 1.93 16.43 1.36 3.15 15.15 98.97
PNS 3/47 14 Brown clay 62.79 0.70 14.40 4.46 0.08 2.23 6.83 1.23 3.59 3.30 99.61
PNS 3/55 17 Silt 62.26 0.43 11.87 3.21 0.05 1.54 7.73 1.95 3.26 6.05 98.34
PNS 3/60 18 Silt 71.05 0.42 9.25 2.57 0.04 1.22 6.01 1.49 2.54 4.53 99.12
PNS 3/62 19 Grey clay 62.65 0.48 15.41 5.51 0.03 1.69 0.62 0.62 3.53 7.80 98.34
PNS 3/70 21 Brown clay 77.95 0.43 8.87 3.10 0.05 0.84 0.48 0.59 2.08 3.91 98.31
PNS 3/85 26 Grey clay 69.28 0.51 12.82 3.73 0.03 1.35 1.01 0.50 2.96 5.70 97.87
PNS 4/5 2 Grey clay 66.19 0.87 17.47 5.76 0.08 2.26 0.48 bdl 4.31 2.20 99.62
PNS 4/12 4 Grey clay 58.39 1.11 17.81 6.94 0.14 1.08 4.83 4.13 3.12 2.30 99.85
PNS 4/17 5 Brown clay 70.72 0.56 13.37 4.28 0.03 1.23 0.23 0.52 3.27 4.57 98.78
PNS 4/35 11 Black clay 59.56 0.52 15.19 5.30 0.04 1.87 1.20 1.00 3.37 10.45 98.51
PNS 4/45 14 Brown clay 47.48 0.48 12.55 4.63 0.08 2.08 13.62 1.56 3.24 13.38 99.10
PNS 4/53 16 Silt 68.61 0.46 9.27 2.71 0.04 1.32 7.33 1.43 2.50 5.04 98.71
PNS 4/60 18 Brown clay 71.41 0.39 9.24 3.67 0.07 1.02 2.83 0.44 2.13 6.70 97.90
PNS 4/63 19 Silt 75.61 0.54 9.35 3.92 0.11 1.13 2.80 bdl 2.27 3.10 98.84
PNS 4/67 20 Grey clay 61.83 0.83 16.57 6.85 0.02 3.42 2.03 bdl 3.53 4.60 99.68
PNS 4/69 21 Grey clay 61.24 0.54 15.21 6.05 0.03 1.61 0.88 0.67 3.68 9.49 99.40
Environ Earth Sci
123

Table 1 continued
Samples Depth Lithology SiO
2
TiO
2
Al
2
O
3
Fe
2
O
3
MnO MgO CaO Na
2
OK
2
O LOI Total
PNS 4/72 22 Silt 73.40 0.53 9.55 4.40 0.26 1.15 4.31 0.16 2.27 3.70 99.74
PNS 4/74 22 Grey clay 50.00 0.45 16.73 6.67 0.07 2.43 5.58 0.49 3.50 13.29 99.21
PNS 5/10 3 Silt 77.73 0.61 10.58 4.40 0.07 1.18 0.53 bdl 2.77 2.40 100.28
PNS 5/16 5 Brown clay 71.94 0.28 10.92 4.16 0.06 1.00 1.17 0.50 2.72 5.93 98.68
PNS 5/19 6 Brown clay 82.06 0.41 7.56 2.79 0.02 0.69 bdl 0.44 1.97 2.65 98.58
PNS 5/25 8 Silt 93.42 0.18 2.18 1.01 0.01 0.22 bdl 0.12 0.74 0.74 98.62
PNS 5/30 9 Silt 87.70 0.31 3.62 2.05 0.12 0.32 2.53 bdl 1.20 2.00 99.86
PNS 5/33 10 Silt 94.60 0.28 2.04 1.57 0.00 0.08 0.50 bdl 0.77 0.10 99.95
PNS 5/35 11 Brown clay 68.75 0.28 3.87 3.14 0.34 0.46 14.88 bdl 1.16 6.90 99.79
PNS 5/36 11 Silt 71.00 0.11 3.35 2.26 0.11 0.42 10.77 0.19 0.91 9.78 98.89
PNS 5/40 12 Black clay 55.07 0.55 14.96 6.16 0.05 2.04 2.82 0.89 3.34 13.56 99.43
PNS 5/42 13 Calcrete 45.78 0.28 8.02 2.89 0.14 2.58 23.66 0.74 1.88 14.20 100.14
PNS 5/45 14 Silt 71.22 0.34 8.12 2.25 0.05 0.93 5.93 1.20 1.99 6.38 98.40
TE11 HHC Gneiss 63.79 0.82 17.25 6.52 0.02 5.08 0.28 2.58 3.01 1.17 100.64
TE12 HHC Gneiss 68.72 0.74 14.27 5.60 0.04 5.02 0.74 1.72 2.51 1.41 100.89
TE13 HHC Marble 67.59 0.63 15.59 4.42 0.05 3.30 0.84 4.38 1.62 1.34 99.79
TE19 HHC Gneiss 76.56 0.63 10.69 4.27 0.05 1.71 0.76 1.27 2.96 1.25 100.29
TE20 HHC Gneiss 74.20 0.61 10.48 4.71 0.24 1.53 6.34 0.92 0.49 0.72 100.43
TE21 HHC Calc-gneiss 51.55 0.33 9.00 3.39 0.05 7.42 9.99 0.64 3.10 13.86 99.43
TE23 HHC Calc-gneiss 38.85 0.33 9.08 3.50 0.07 10.81 13.34 0.13 3.28 21.37 100.86
TE24 HHC Amphibolite 67.68 0.52 13.35 3.20 0.02 2.90 0.72 1.03 3.27 8.24 100.97
TE26a HHC Quartzite 69.20 0.55 15.12 2.37 0.01 2.86 0.75 0.55 4.64 3.82 100.45
TE26b HHC Gneiss 47.56 0.33 8.73 1.36 0.04 9.35 10.65 1.18 2.52 17.91 99.76
TE27 HHC Amphibolite 49.18 1.54 14.48 10.91 0.18 6.84 8.32 4.76 0.08 4.45 100.92
TE28 HHC Schist 56.61 1.44 11.72 13.21 0.14 6.09 5.07 3.77 0.85 1.84 100.92
TE29 HHC Quartzite 95.83 0.04 0.81 0.96 0.00 0.71 0.04 bdl bdl 0.67 99.06
TE10 LH Black schist 69.99 0.51 15.28 3.35 0.01 1.42 0.06 1.28 4.18 3.61 99.79
TE2 LH Black schist 62.59 0.55 18.14 5.81 0.03 2.62 0.14 0.73 5.16 4.76 100.64
TE30 LH Black schist 58.11 0.61 21.88 5.72 0.07 2.03 0.40 1.41 5.88 3.19 99.43
TE35 LH Black schist 80.43 0.24 7.94 3.97 0.01 2.20 0.16 bdl 2.78 2.18 100.01
TE1 LH Marble 11.57 0.02 0.42 1.51 0.08 18.35 26.88 bdl 0.14 41.49 100.48
TE18 LH Schist 69.11 0.48 14.76 4.85 0.02 2.51 0.36 2.12 4.44 1.94 100.70
TE3 LH Marble 0.93 0.01 0.24 0.15 0.00 0.30 54.85 bdl 0.07 43.29 99.84
TE31 LH Serpentinite 45.75 1.86 15.97 10.05 0.15 15.48 1.54 1.96 bdl 6.84 99.76
TE32 LH Marble 7.49 0.04 0.99 0.72 0.07 19.40 27.80 bdl 0.31 42.34 99.22
TE33 LH Calc-schist 58.95 0.01 0.20 0.16 0.01 9.07 12.50 bdl 0.06 17.99 98.94
TE34 LH Calc-schist 54.16 0.41 10.79 4.18 0.06 6.39 7.32 0.20 4.21 12.94 100.75
TE36 LH Marble 10.26 0.10 2.58 1.02 0.06 19.08 25.94 bdl 1.07 40.48 100.61
TE37 LH Schist 71.05 0.70 14.07 5.55 0.05 0.96 0.13 0.65 2.84 4.40 100.42
TE38 LH Schist 88.40 0.54 4.33 3.86 0.01 0.21 0.05 bdl 0.76 2.19 100.41
TE39 LH Schist 60.08 0.59 19.17 7.27 0.03 1.94 0.03 bdl 5.60 5.43 100.25
TE4 LH Marble 25.69 0.01 0.26 0.18 0.03 16.12 22.86 bdl 0.07 34.16 99.40
TE40 LH Schist 78.47 1.43 6.00 7.27 0.16 0.57 1.70 bdl 0.56 3.14 99.50
TE48 LH Marble 2.67 0.01 0.21 0.17 0.00 21.27 30.06 bdl 0.05 44.94 99.39
TE50 LH Schist 40.36 0.77 12.60 9.50 0.38 2.21 15.33 0.20 2.35 16.01 99.92
TE7 LH Marble 19.20 0.07 0.20 0.02 17.72 25.04 bdl bdl 37.12 99.38
TE8 LH Dolomite 0.60 0.01 0.53 1.06 0.11 21.21 30.30 bdl 0.07 45.73 99.61
TE9a LH Schist 64.85 0.62 17.45 6.11 0.04 2.88 0.41 0.73 3.53 3.48 100.16
Environ Earth Sci
123

the entire sequence. In addition, at the start of each daily
sequence, pure solutions of Ba, Ce, Pr and Nd were run to
evaluate and correct oxide interferences on the middle and
heavy REE and the interference of double-charged Nd on
Ge. All calculations to transform peak signals into con-
centrations were done offline. For each element, the aver-
age peak signal of the chosen rock standard was used as the
calibrating value for all calculations. Before calculation of
the element concentrations, we perform three corrections:
(1) an oxide interference correction based on the oxide/
metal ratio determined with the pure element solutions, (2)
a machine drift correction based on the five-element spike
with a mass-based interpolation and (3) a blank subtraction.
To perform the machine drift correction for both modes
(with or without helium), the beryllium and germanium
peaks are measured in both modes whereas other spikes are
measured only in the no-gas mode. The overall quality of
the chemical procedure is controlled as follows: (1) sys-
tematic evaluation of the difference between the procedural
blank and the running solution blank, (2) the entire che-
mical and analytical procedure is checked by system-
atically duplicating at least one sample per batch, and (3)
the machine stability is controlled using multiple runs of
randomly selected rock solutions.
Shells of freshwater mollusks were found in the clay,
silty-clay/clayey-silt, and the fine sand facies (mollusk
fossils in Fig. 2). The individual shells were analysed for
stable isotopes (C and O) in the laboratory of Centre de
Recherche Pe
´trographiques et Ge
´ochimiques, Nancy,
France. The mollusk shells were cleaned in distilled water
with ultrasound to remove adhered sediments and organic
debris. Then, they were reacted under vacuum
with &100 % phosphoric acid at 25 ±0.1 °C (McCrea
1950). Isotopic analyses of the released CO
2
were per-
formed in a modified mass spectrometer, model VG 602.
Carbon and oxygen isotopic ratios of aragonite are ex-
pressed as d
13
C
Ara
and d
18
O
Ara
relative to Pee Dee
Belemnite (PDB). Following common practice for arago-
nite shell analyses (e.g. Dettman et al. 1999;Le
´cuyer et al.
2004), we did not apply any correction specific for arago-
nite and used the standard correction for calcite. Ten to
fifteen repeated analyses of calcite international standards
over the analytical period are: NBS-18 (d
13
C=
-5.02 ±0.04 %and d
18
O=-23.06 ±0.13 %), NBS-19
(d
13
C=2.00 ±0.04 %and d
18
O=-2.15 ±0.13 %),
IAEA-CO-1 (d
13
C=2.47 ±0.04 %and d
18
O=
-2.46 ±0.09 %), IAEA-CO-8 (td
13
C=-5.74 ±0.07 %
and d
18
O=-22.79 ±0.1 %). Overall reproducibility
is ±0.1 %.
Sediment depositional ages were determined when
possible by
14
C on organic matter (Table 2). Retrieved
sediment samples were typically organic-poor, with rare
small pieces of preserved terrestrial detritus and shell
fragments seen perhaps once in each drill hole. One sample
of woody fragments from 7.3 m depth from the last drill
hole (Site 5) was retrieved from Nepal and rinsed with
distilled water, dried overnight at 60 °C (to 3.5540 g) and
sent to Daniel Weinand at the University of Tennessee’s
Center for Archaeometry and Geochronology (UTAG) for
liquid scintillation counting
14
C dating.
Results
14
C Ages
The wood fragment from *7 m depth in borehole 5 gives
a
14
C age between 3354 and 2928 year BC using an av-
erage delta-13C value of -26.5 for the fractionation cor-
rection (UTAG sample ID 07-013).The final calibrated age
was determined using IntCal04/OxCal4.0 for reservoir
correction.
Geochemistry
Average geochemical compositions (major and trace ele-
ments) of 61 sediments of the Nawalparasi boreholes are
reported in Table 2. We also report, when available, the
composition of the potential sources using 41 rock samples
collected from the Higher Himalayan Crystallines (HHC),
the Lesser Himalaya s.l. (LH), the black schists from the
Lesser Himalaya and the Siwaliks (Siw) along the Kali
Table 1 continued
Samples Depth Lithology SiO
2
TiO
2
Al
2
O
3
Fe
2
O
3
MnO MgO CaO Na
2
OK
2
O LOI Total
TE9b LH Calc-schist 40.52 0.02 0.47 0.26 0.05 12.77 17.76 bdl 0.16 27.33 99.37
TE41 SIW Calcrete 35.82 0.31 6.13 4.22 0.15 2.46 25.48 0.30 1.50 23.83 100.28
TE42 SIW Grey clay 76.94 0.43 6.91 2.33 0.03 1.54 3.96 0.66 1.61 5.40 99.90
TE44 SIW Calcrete 39.69 0.43 9.48 3.94 0.11 2.79 20.50 0.31 2.84 20.11 100.32
TE45 SIW Calcrete 44.87 0.48 9.88 3.83 0.13 3.48 16.62 0.57 2.65 17.87 100.50
TE47 SIW Silt 80.84 0.38 5.85 2.49 0.03 0.85 2.81 0.45 1.62 3.60 98.97
Black schists from the upper Lesser Himalaya are in bold
Environ Earth Sci
123

Gandaki, Seti and Narayani rivers (Tables 1,2). System-
atically, the silt and sand fractions are enriched in SiO
2
(between 60 and 90 %wt) relative to the clay fractions
(between 40 and 70 %wt). Less mobile elements such as
Al
2
O
3
,Fe
2
O
3
, MgO, TiO
2
are systematically enriched in
the clay fractions for a given SiO
2
content (Fig. 3). Nev-
ertheless, the general negative trend between those ele-
ments and the SiO
2
content is partly due to a dilution effect
by quartz in the silt and sand fractions. It is noticeable that
the grey clays are the most enriched in immobile elements.
Na
2
O and MnO have similar patterns, depleted in the clay
fraction (Fig. 3), and this may be related to a leaching
effect in anoxic conditions where Mn(IV) is reduced to
Mn(II). Calcium is mostly concentrated in the brown clays
and associated with calcrete.
Altogether, 38 samples from the boreholes and 41
samples from the Himalaya were analysed for REEs
(Table 3). As before, they were also categorised into the
same three sediment-type groupings: grey clay, brown
clay, silt-sand. All the groupings of sediments have similar
chondrite-normalised REE patterns (Fig. 4) with overlap-
ping abundances. The average REE concentrations show an
increase from sand/silt fraction to clay fraction. On aver-
age, REE concentrations in the sands and silts are *30 %
Fig. 2 Interpreted stratigraphic
logs of the five drilling bores
with sampling depth and OSL
ages and
14
C dating*
Environ Earth Sci
123

Table 2 Geochemical composition in trace elements of the Nawalparasi borehole sediments and rocks coming from Himalaya (HHC higher
himalayan crystallines, LH lesser himalaya, Siw siwaliks)
Samples As
(ppm)
Ce
(ppm)
Co
(ppm)
Cr
(ppm)
Cs
(ppm)
Cu
(ppm)
Ni
(ppm)
Pb
(ppm)
Rb
(ppm)
Sr
(ppm)
Th
(ppm)
U
(ppm)
Zr
(ppm)
C org
(%)
PNS 2/10 8.41 101 16 92 12.2 28.8 40.3 34.3 213 45.4 22.8 3.47 164 0.14
PNS 2/20 15 99.9 16.9 99.8 12.4 29.4 42.7 32.6 199 39.7 22.9 3.59 178 0.06
PNS 2/28 17.2 65.6 12.9 77.2 8.65 22.8 37.5 21.2 134 124 14.3 3.81 240 0.41
PNS 2/36 10.2 78 13.3 81.1 8.85 24.8 37.8 24.3 148 108 16.5 3.83 216 0.34
PNS 2/49 3.09 76.5 5.11 30.4 4.9 5.79 10.6 20.6 96.9 183 16.6 3.04 215 0.05
PNS 2/52 3.97 49.9 10.4 53.6 9.24 27.2 28 23.2 139 210 11.8 2.08 100 0.15
PNS 2/57 3.04 81.5 6.16 41.5 4.53 10.2 25.5 20.7 106 241 19.1 3.42 297 0.03
PNS 2/60 2.85 69.8 6.63 38.5 5.82 7.62 14.7 20.4 124 311 16.3 3.39 204 0.04
PNS 2/62 9.87 97 17.5 104 13.8 35 48 32.4 231 86.7 23.4 3.83 132 0.51
PNS 2/63 16.8 87.5 16.8 102 13.7 37 49.4 29.2 216 101 20.4 3.51 144 0.78
PNS 2/67 12.3 86.5 15.8 97.7 11.9 45.3 46.4 29.1 221 107 21.4 3.48 170 0.47
PNS 2/75 10.1 88.7 13.5 77.2 9.96 23.1 32.1 31.8 176 87.9 18.7 3.41 153 0.04
PNS 3/10 7.38 95.7 11.9 69.9 7.89 23.7 27.7 27.1 160 37.9 19.1 3.18 277 0.04
PNS 3/18 10.4 91.9 15 81 9.26 24.2 35.5 26.8 170 38.7 19.8 3.46 242 0.11
PNS 3/28 4.92 107 15.7 93.6 11.5 28.5 40.3 32.9 206 48.3 22.4 3.22 209 0.15
PNS 3/32 6.55 105 18.6 114 14.6 40.2 55.9 35.3 233 54.9 25.4 4.13 143 0.98
PNS 3/37 10.6 93.3 15.1 96.4 9.4 33.2 46 21.7 169 56.2 20.3 3.73 223 1.08
PNS 3/40 2.15 67.3 4.96 32.4 5.64 6.53 11.9 21.7 115 186 14.8 2.6 177 0.05
PNS 3/46 4.17 62 11.8 62.1 11.1 29.5 29.4 24.9 180 277 14.3 2.81 133 0.16
PNS 3/55 2.6 57.1 8.41 51.6 8.34 10.8 21.2 25.1 171 280 12.5 2.42 219 0.09
PNS 3/60 3.27 53.4 6.33 38.5 6.1 8.19 15.1 19.1 127 222 11.6 2.43 289 0.07
PNS 3/62 13.4 90.5 12.6 88.4 10.3 27.6 38.1 25 170 72.9 19.4 5.12 274 0.55
PNS 3/70 6.55 89.4 8.78 50.6 4.88 13.4 23.2 16.4 102 41.3 17.2 3.27 489 0.05
PNS 3/85 4.43 86.6 10.9 68.1 8.3 20.3 29.5 21 158 51.2 18.1 2.73 266 0.15
PNS 4/17 5.86 97.9 10.4 71.1 7.69 19.8 27.3 24.1 158 39.7 19.1 3.28 299 0.15
PNS 4/35 10.8 85.5 12.9 112 10.6 39.9 44.1 23.7 162 90.8 17.8 7.02 195 1.86
PNS 4/45 3.8 74.8 11.7 60.9 10.4 34.4 30.5 25.3 173 271 15.6 3.05 179 0.16
PNS 4/53 3.22 56 7.47 43.7 6.42 6.78 17.1 18.1 129 245 12.4 2.47 215 0.05
PNS 4/60 11.9 77.9 11.2 58.7 5.88 18.4 27.6 20.6 117 71 15.8 3.41 333 0.08
PNS 4/69 27.7 88.9 14 93.4 11.3 31 43.6 26 190 88 19.4 8.06 206 0.73
PNS 4/74 8.56 86.1 16.9 98.7 11.3 34.4 47.6 28.3 190 127 19.2 2.87 138 0.45
PNS 5/16 8.11 78.8 10.3 55.8 6.24 18.5 23.7 23.2 129 35 15.7 2.56 242 0.04
PNS 5/19 3.53 68.8 7.2 39.1 3.94 13 16.9 18.7 87.7 26.1 13.8 2.79 418 0.05
PNS 5/25 1.74 31.5 2.59 14.9 1.32 3.79 5.63 6.95 32.2 10.7 6.03 1.38 283 0.05
PNS 5/36 9.92 76.5 13 78.4 8.73 24.4 37.5 24 145 108 16.2 3.79 217 0.03
PNS 5/40 11 78.5 14.8 112 10.6 49.5 49.2 26.8 171 86.8 17.5 6.04 176 2.73
PNS 5/42 14.5 48.3 10.9 44.7 7.15 20.8 27.6 20.5 107 304 10.7 5.22 130 0.11
PNS 5/45 5.06 60.3 6.09 32.4 5.24 9.37 13.5 19.4 101 134 12.6 2.63 257 0.10
TE11 0.00 5.11 bdl 4.87 2.30 1.50 129.50 bdl 7.04 16.71 128.60 25.76 175.00 nd
TE12 0.00 7.01 12.82 4.78 2.99 1.32 109.60 bdl 18.56 16.23 124.20 26.57 231.30 nd
TE13 0.00 5.99 11.65 3.99 2.14 1.18 71.43 bdl 9.49 14.63 80.01 20.86 187.70 nd
TE19 0.00 4.41 13.16 4.81 2.67 1.21 117.40 bdl 3.31 14.70 69.65 26.58 306.60 nd
TE20 3.15 1.13 bdl 5.93 3.32 1.54 25.61 0.42 5.64 13.19 82.69 35.87 243.20 nd
TE21 0.00 4.78 13.89 3.21 1.83 0.86 124.10 bdl 2.38 10.76 50.97 18.40 134.70 nd
TE23 0.00 4.03 bdl 3.44 1.92 0.81 122.60 0.24 3.12 12.12 46.59 19.99 120.60 nd
TE24 0.00 11.75 24.24 3.44 1.78 1.11 151.20 0.97 2.24 12.65 127.00 18.94 115.90 nd
TE26a 0.00 9.27 7.28 5.59 3.43 1.29 193.90 0.50 3.72 18.68 168.80 37.07 170.40 nd
Environ Earth Sci
123

lower compared to the clay fraction (Fig. 4). Brown and
grey clays have more or less the same REE concentrations
(200 ppm), but the grey clays show less dispersion. A di-
lution effect by quartz can be invoked to explain the lower
REE concentrations in the silty fraction because quartz is
relatively abundant in these sediments. This effect cannot
however, account for the difference of concentration ob-
served between the brown and grey clays because they
share similar SiO
2
content (Fig. 3). The higher abundance
of CaO in the brown clay (5–15 wt %) relative to the grey
clay (CaO \5 wt %) could partly explain the lower REE
content (i.e. dilution effect by carbonate). Another possi-
bility is that the source for the clays is different: Higher
Himalayan Crystalline or Lesser Himalaya for the grey
clays, Siwaliks for the brown clays. All the sediments
analysed in this study show fractionated normalised REE
patterns with average La
N
/Yb
N
ratios of 9.95, 9.74 and 9.62
for the grey clays, brown clays and sand/silt, respectively
(Fig. 4). Weathering processes can explain the different
REE concentrations in the sediments (e.g. Potter et al.
2005; Singh 2009). The (Eu/Eu*)
N
ratio is very consistent
from one sample to another, with the europium anomaly
being strongly negative, varying between 0.17 and 0.23.
For comparison, we report the average REE pattern of
sediments from different parts of the mainstream Ganga
river, after the confluence of all of its tributaries and before
entering the Ganga plain (Singh 2009). It is noticeable that
REE patterns of the Ganga sediments are quite similar to
Table 2 continued
Samples As
(ppm)
Ce
(ppm)
Co
(ppm)
Cr
(ppm)
Cs
(ppm)
Cu
(ppm)
Ni
(ppm)
Pb
(ppm)
Rb
(ppm)
Sr
(ppm)
Th
(ppm)
U
(ppm)
Zr
(ppm)
C org
(%)
TE26b 0.00 4.47 71.53 1.67 0.96 0.35 112.80 0.71 2.57 10.49 43.71 9.62 130.00 nd
TE27 2.23 bdl 171.50 4.03 2.13 1.46 0.48 1.80 1.34 3.32 245.60 21.69 130.90 nd
TE28 0.00 7.11 26.71 3.85 2.01 1.40 38.01 0.84 1.49 3.72 292.10 20.37 128.50 nd
TE29 0.00 bdl bdl 0.39 0.21 0.19 0.33 0.31 0.48 2.00 4.43 2.14 54.36 nd
TE10 18.32 6.54 52.21 3.74 2.49 0.86 168.60 1.13 3.41 14.88 79.63 24.91 187.80 nd
TE2 18.21 11.99 10.55 4.50 2.57 1.27 276.80 1.32 5.01 17.78 99.09 26.73 138.00 nd
TE30 14.01 10.16 6.32 5.68 3.04 1.64 289.20 0.27 7.65 20.28 81.44 31.35 139.80 nd
TE35 17.60 3.04 4.06 3.16 1.67 0.60 92.21 1.35 2.04 8.40 23.43 17.33 142.20 nd
TE1 6.75 bdl 14.76 0.44 0.22 0.16 4.56 0.54 bdl 0.51 13.17 2.79 8.25 nd
TE18 0.00 7.79 bdl 4.31 2.34 1.37 223.40 1.06 3.68 27.71 39.56 24.98 260.20 nd
TE3 0.00 0.19 bdl 0.07 0.05 0.02 3.72 bdl bdl 0.19 1.27 0.62 2.05 nd
TE31 0.00 0.11 bdl 4.25 2.22 1.76 0.84 1.22 1.25 3.18 272.40 22.80 135.00 nd
TE32 1.85 0.33 bdl 0.72 0.41 0.15 7.61 0.26 0.25 1.75 6.86 4.27 26.80 nd
TE33 0.00 bdl bdl 0.06 0.03 0.01 1.77 bdl bdl 0.18 bdl bdl 2.03 nd
TE34 3.64 9.73 bdl 3.93 2.31 0.93 157.10 1.59 2.93 13.73 49.35 23.24 176.00 nd
TE36 0.00 1.49 bdl 1.58 0.78 0.50 42.29 0.71 0.75 3.04 18.86 9.70 33.45 nd
TE37 7.85 7.44 25.64 6.67 3.67 1.69 131.80 0.43 3.64 17.43 79.83 39.12 276.70 nd
TE38 14.67 1.51 17.07 8.55 3.30 3.32 32.64 1.75 1.19 6.30 43.19 34.77 186.50 nd
TE39 5.94 15.16 27.85 6.02 3.50 1.32 262.10 2.23 5.24 21.03 95.91 36.07 158.20 nd
TE4 3.57 bdl bdl 0.16 0.10 0.03 2.06 0.24 bdl 0.30 1.80 0.96 2.81 nd
E40 1.90 1.76 25.05 8.73 4.37 2.28 33.00 0.39 2.14 16.02 99.06 45.82 511.30 nd
TE48 0.00 0.25 bdl 0.18 0.10 0.04 2.47 bdl bdl 0.38 1.65 1.10 5.68 nd
TE50 5.86 4.50 27.70 4.22 2.26 1.28 121.10 0.23 2.31 10.77 110.80 22.75 153.10 nd
TE7 0.00 bdl bdl 0.07 0.07 0.01 0.54 bdl bdl 1.72 2.04 0.50 5.23 nd
TE8 0.00 0.14 bdl 0.28 0.15 0.08 3.20 1.08 bdl 0.70 5.18 1.65 3.79 nd
TE9a 0.00 8.80 51.13 4.89 2.58 1.30 196.20 0.20 bdl 18.53 55.71 27.07 136.60 nd
TE9b 5.73 0.11 bdl 0.16 0.09 0.04 4.91 2.17 bdl 0.44 2.97 0.86 4.05 nd
TE41 7.17 3.62 10.42 3.09 1.72 0.75 63.58 1.32 1.84 8.69 38.28 20.04 130.10 nd
TE42 0.00 2.36 14.59 3.49 1.91 0.85 62.19 0.68 1.70 11.73 36.45 19.38 254.30 nd
TE44 10.33 6.18 16.66 4.04 2.22 1.00 123.30 1.36 2.74 12.76 56.62 23.90 106.70 nd
TE45 8.79 5.39 18.61 4.28 2.40 1.05 116.90 1.52 2.82 14.34 59.13 24.87 164.20 nd
TE47 0.00 2.55 6.96 3.15 1.64 0.93 64.70 0.69 1.71 11.13 32.87 16.70 203.50 nd
bdl below detection limit, nd not determined. Black schists from the upper Lesser Himalaya are in bold
Environ Earth Sci
123

the pattern of our samples, particularly for the clay fraction.
However, the Ganga sediments are on average two times
more enriched in HREE than the studied samples. In
contrast, the average Siwaliks pattern is similar to the
pattern of our three groups, and its average REE concen-
tration is equal to the average concentration of the studied
samples.
The Upper Continental Crust (UCC) (Taylor and
McLennan 1985) normalised REE patterns of the studied
samples are reported in Fig. 5. As already noticed by Singh
(2009) for the Ganga river sediments, the significant fea-
ture of the shale normalised pattern is the depletion of Eu
in the three groups of sediments, and the similarity of the
REE patterns of the three groups with the Higher Hi-
malayan Crystalline and Lesser Himalaya is highlighted.
Moreover, in both REE normalised diagrams (Figs. 4,5),
the grey clays fractions representing the more evolved
sediments are equally enriched compared to the Higher
Himalayan Crystallines and Lesser Himalayan rocks.
Arsenic distribution
The distribution of arsenic is not uniform among sediment
types. In general, arsenic concentrations are higher in the
fine-grained fraction (grey clay and brown clay) than in
coarser materials (fine sand and silt). This could be related
to the quartz dilution effect. The average arsenic concen-
tration in the analysed samples is 8 ppm. The average
arsenic concentration in the grey clays is 11 ppm, 8 ppm in
the brown clays and 4 ppm in the silt/sand fraction
(Table 2). The highest arsenic concentration (27.7 ppm) is
observed in the borehole 4 at a depth of 20.7 m, in a metre-
thick, dark grey layer (Fig. 2). A general positive trend
appears between arsenic concentration, Fe
2
O
3
,K
2
O and
Al
2
O
3
contents (Fig. 6). The average total iron as Fe
2
O
3
is
4.33 wt %, with higher concentrations in the clay layers
(5.71 wt %) and lower concentrations in the silt/sand lay-
ers (2.76 wt %). The arsenic concentration partly increases
with the total C organic content. In contrast, there is no
relationship between arsenic content and calcium content.
We also reported the average arsenic content of the po-
tential sources. The black schists from the Lesser Himalaya
have the highest content (14–19 ppm), the micaschists and
gneisses of the Higher Himalayan Crystallines have the
lowest content (0–3 ppm) while the rest of the Lesser Hi-
malaya (0–7 ppm) and the Siwaliks (0–10 ppm) have in-
termediate arsenic content. In the discussion, we will show
that weathering processes affecting the Lesser Himalaya
Fig. 3 Variation diagram of major elements against SiO
2
(wt %) for
all the groups of sediment studied. Average composition of the source
rocks are from Galy and France-Lanord (2001) including our data
presented in Table 1.THB trans-himalayan batholith, HHC higher
himalayan crystallines, LH Lesser Himalaya
Environ Earth Sci
123

Table 3 Geochemical composition in REE elements of the Nawalparasi borehole sediments and rocks coming from Himalaya (HHC higher
himalayan crystallines, LH lesser himalaya, Siw Siwaliks)
Samples La
(ppm)
Ce
(ppm)
Pr
(ppm)
Nd
(ppm)
Sm
(ppm)
Eu
(ppm)
Gd
(ppm)
Tb
(ppm)
Dy
(ppm)
Ho
(ppm)
Er
(ppm)
Tm
(ppm)
Yb
(ppm)
Lu
(ppm)
PNS 2/10 53.20 101.00 12.10 44.40 8.61 1.59 6.94 1.10 6.27 1.26 3.61 nd 3.31 0.50
PNS 2/20 48.60 99.90 11.40 42.10 8.30 1.51 6.77 1.07 6.17 1.24 3.54 nd 3.26 0.49
PNS 2/28 32.20 65.60 7.40 28.40 5.69 1.04 4.73 0.79 4.46 0.91 2.64 nd 2.54 0.39
PNS 2/36 38.90 78.00 9.40 33.30 6.55 1.20 5.74 0.92 5.05 1.08 3.20 nd 2.81 0.43
PNS 2/49 37.60 76.50 8.80 32.00 6.35 1.00 5.21 0.85 5.06 1.00 2.92 nd 2.77 0.43
PNS 2/52 25.30 49.90 5.83 21.50 4.13 0.72 3.59 0.57 3.14 0.67 1.82 nd 1.64 0.26
PNS 2/57 42.20 81.50 9.67 35.30 6.60 1.09 5.96 0.94 5.35 1.06 3.03 nd 2.80 0.42
PNS 2/60 37.40 69.80 8.68 31.40 6.28 1.18 5.26 0.84 4.70 0.94 2.64 nd 2.40 0.36
PNS 2/62 49.20 97.00 11.40 41.20 8.19 1.52 6.62 1.06 6.10 1.21 3.57 nd 3.18 0.47
PNS 2/63 42.30 87.50 9.78 35.60 7.03 1.26 5.67 0.91 5.18 1.03 2.94 nd 2.78 0.42
PNS 2/67 45.30 86.50 10.40 38.60 7.34 1.35 5.96 0.97 5.34 1.11 3.04 nd 2.83 0.44
PNS 2/75 45.00 88.70 10.40 38.00 7.31 1.36 6.11 0.98 5.47 1.13 3.24 nd 2.92 0.44
PNS 3/10 48.30 95.70 11.30 40.70 7.67 1.39 6.16 1.02 5.85 1.23 3.49 nd 3.26 0.49
PNS 3/18 45.60 91.90 10.80 39.10 7.51 1.38 6.31 1.02 5.93 1.21 3.44 nd 3.23 0.49
PNS 3/28 52.50 107.00 12.10 44.30 8.59 1.61 7.06 1.14 6.58 1.31 3.75 nd 3.43 0.51
PNS 3/32 51.70 105.00 11.70 42.50 8.32 1.53 6.62 1.08 6.03 1.21 3.49 nd 3.20 0.49
PNS 3/37 45.20 93.30 10.60 38.60 7.48 1.45 6.23 1.00 5.89 1.18 3.35 nd 3.12 0.48
PNS 3/40 34.00 67.30 8.00 29.00 5.62 0.99 4.68 0.75 4.09 0.82 2.34 nd 2.20 0.33
PNS 3/46 31.90 62.00 7.32 27.20 5.26 0.98 4.66 0.75 4.16 0.89 2.42 nd 2.31 0.35
PNS 3/55 29.40 57.10 6.66 23.30 4.38 0.97 3.89 0.62 3.65 0.75 2.04 nd 1.95 0.30
PNS 3/60 27.10 53.40 6.28 22.70 4.43 0.86 3.69 0.60 3.41 0.70 2.06 nd 1.93 0.29
PNS 3/62 42.70 90.50 9.82 36.40 6.97 1.26 5.72 0.93 5.41 1.09 3.15 nd 3.00 0.45
PNS 3/70 44.00 89.40 10.10 36.80 6.93 1.16 5.49 0.90 5.37 1.11 3.30 nd 3.15 0.49
PNS 3/85 43.40 86.60 10.20 37.20 7.05 1.25 5.58 0.89 5.10 1.05 2.99 nd 2.78 0.43
PNS 4/17 49.20 97.90 11.50 42.00 7.99 1.43 6.54 1.07 6.04 1.26 3.62 nd 3.36 0.51
PNS 4/35 38.90 85.50 9.30 34.90 7.02 1.33 6.00 0.95 5.63 1.14 3.27 nd 3.02 0.46
PNS 4/45 36.70 74.80 8.50 30.70 6.05 1.12 5.17 0.84 4.87 0.99 2.78 nd 2.59 0.40
PNS 4/53 27.90 56.00 6.60 24.20 4.81 0.88 4.02 0.66 3.61 0.74 2.09 nd 1.89 0.29
PNS 4/60 39.20 77.90 9.08 33.20 6.28 1.13 5.19 0.84 4.76 1.00 2.92 nd 2.74 0.42
PNS 4/69 44.00 88.90 10.20 36.80 7.14 1.34 5.81 0.94 5.39 1.07 3.01 nd 2.80 0.43
PNS 4/74 43.10 86.10 9.87 36.20 7.10 1.30 5.74 0.91 5.28 1.07 2.98 nd 2.80 0.42
PNS 5/16 39.80 78.80 9.14 33.20 6.33 1.17 5.18 0.82 4.91 1.02 2.88 nd 2.68 0.42
PNS 5/19 34.40 68.80 8.13 29.50 5.62 0.97 4.72 0.75 4.45 0.93 2.76 nd 2.55 0.40
PNS 5/25 15.80 31.50 3.72 13.60 2.59 0.43 2.09 0.36 2.05 0.44 1.33 nd 1.31 0.20
PNS 5/36 38.50 76.50 8.96 33.30 6.58 1.17 5.45 0.90 5.05 1.04 2.95 nd 2.78 0.41
PNS 5/40 39.30 78.50 9.24 34.10 6.84 1.28 5.77 0.96 5.48 1.11 3.24 nd 2.96 0.45
PNS 5/42 24.20 48.30 5.70 20.70 4.18 0.76 3.69 0.59 3.47 0.72 2.07 nd 1.93 0.29
PNS 5/45 30.40 60.30 7.12 25.80 5.11 0.91 4.28 0.68 3.83 0.78 2.27 nd 2.11 0.32
TE11 41.05 79.72 8.75 34.92 7.08 1.50 6.03 0.90 4.87 0.87 2.30 0.32 1.96 0.29
TE12 43.20 85.34 9.38 37.37 7.37 1.32 6.06 0.88 4.78 0.95 2.99 0.49 3.47 0.56
TE13 34.77 68.14 7.51 30.03 6.06 1.18 5.07 0.74 3.99 0.75 2.14 0.33 2.28 0.36
TE19 36.14 71.35 7.89 31.57 6.35 1.21 5.55 0.85 4.81 0.92 2.67 0.41 2.81 0.43
TE20 41.41 76.92 8.30 32.81 6.59 1.54 5.98 0.97 5.93 1.17 3.32 0.49 3.33 0.51
TE26a 41.84 74.51 9.20 34.73 6.76 1.29 5.77 0.91 5.59 1.14 3.43 0.53 3.61 0.56
TE26b 27.50 50.55 5.15 18.91 2.96 0.35 2.08 0.29 1.67 0.32 0.96 0.15 1.07 0.18
TE27 15.96 33.49 4.11 18.44 4.46 1.46 4.44 0.68 4.03 0.77 2.13 0.31 1.96 0.30
TE28 19.04 38.10 4.55 19.91 4.45 1.40 4.29 0.65 3.85 0.72 2.01 0.30 1.91 0.30
Environ Earth Sci
123

black schists most likely explain the arsenic enrichment
observed in the grey clays.
Shell geochemistry
Fragments of shell of freshwater mollusks were collected
from the sledge samples at eight different depths below
ground level and belong to Gastropod sp, Bellamya sp.,
and bivalvia sp. Carbon (d
13
C) and oxygen (d
18
O) isotope
values of these shells range from -7to?2%, and -9to
-3%, respectively, in PDB scaling (Table 2). One living
freshwater mollusk of Lymnaea sp. was also sampled on 8
May 2007 from 2 m wide and 0.15 m deep running river
water of the Turia Khola at Amahawa (27°3005500N,
83°3601100E; 106 m a.m.s.l.). Temperature and pH value of
the water were, respectively, 32.9 °C and 8.33. The d
13
C
and d
18
O values of the shell are, respectively, -9.3 and
-3.8 %(Table 4).
Discussion
Stratigraphy and dating
The stratigraphy and depositional history of Quaternary
sediments in the Himalayan foreland basin have been
Table 3 continued
Samples La
(ppm)
Ce
(ppm)
Pr
(ppm)
Nd
(ppm)
Sm
(ppm)
Eu
(ppm)
Gd
(ppm)
Tb
(ppm)
Dy
(ppm)
Ho
(ppm)
Er
(ppm)
Tm
(ppm)
Yb
(ppm)
Lu
(ppm)
BU2 2.43 7.12 1.05 4.63 1.18 0.24 1.05 0.16 0.90 0.18 0.45 0.06 0.40 0.06
TE1 4.40 9.25 1.05 3.72 0.67 0.16 0.54 0.08 0.44 0.08 0.22 0.03 0.19 0.03
TE3 0.60 1.12 0.12 0.47 0.08 0.02 0.08 0.01 0.07 0.02 0.05 0.01 0.05 0.01
TE4 0.95 2.09 0.21 0.86 0.16 0.03 0.15 0.02 0.16 0.03 0.10 0.01 0.09 0.02
TE7 0.41 0.68 0.08 0.31 0.06 0.01 0.04 0.01 0.07 0.02 0.07 0.01 0.11 0.02
TE8 1.67 3.97 0.40 1.63 0.33 0.08 0.31 0.05 0.28 0.06 0.15 0.02 0.14 0.02
TE9a 46.46 87.95 9.45 36.44 6.75 1.30 5.60 0.86 4.89 0.93 2.58 0.38 2.33 0.34
TE9b 1.10 2.15 0.25 1.00 0.20 0.04 0.16 0.03 0.16 0.03 0.09 0.01 0.10 0.01
TE2 48.52 94.69 10.71 37.89 6.65 1.27 5.28 0.79 4.50 0.89 2.57 0.39 2.66 0.42
TE10 32.01 56.71 6.28 23.72 4.36 0.86 3.65 0.60 3.74 0.80 2.49 0.39 2.75 0.43
TE30 57.11 107.20 11.72 45.48 8.36 1.64 6.63 0.99 5.68 1.08 3.04 0.45 2.98 0.46
TE35 14.46 29.06 3.12 12.60 3.12 0.60 3.70 0.56 3.16 0.59 1.67 0.25 1.60 0.24
TE21 24.24 47.67 5.22 20.56 4.02 0.86 3.47 0.53 3.21 0.63 1.83 0.28 1.91 0.30
TE23 25.92 51.35 5.51 21.43 4.26 0.81 3.69 0.58 3.44 0.67 1.92 0.30 1.94 0.29
TE24 39.10 73.08 8.10 30.87 5.44 1.11 4.14 0.61 3.44 0.65 1.78 0.25 1.61 0.23
TE18 53.92 100.20 10.32 38.40 6.66 1.37 5.17 0.76 4.31 0.83 2.34 0.37 2.52 0.39
TE29 11.29 16.35 1.52 5.35 0.78 0.19 0.50 0.07 0.39 0.07 0.21 0.03 0.24 0.04
TE31 15.69 33.39 4.20 19.09 4.43 1.76 4.59 0.71 4.25 0.81 2.22 0.32 2.05 0.32
TE32 4.45 10.13 1.04 4.26 0.85 0.15 0.76 0.12 0.72 0.15 0.41 0.06 0.43 0.07
TE33 0.41 0.86 0.10 0.41 0.08 0.01 0.06 0.01 0.06 0.01 0.03 0.01 0.03 0.01
TE34 33.44 64.97 7.09 27.71 5.18 0.93 4.26 0.66 3.93 0.77 2.31 0.35 2.39 0.39
TE36 9.19 18.29 2.25 9.46 2.31 0.50 2.17 0.29 1.58 0.29 0.78 0.11 0.70 0.11
TE37 43.82 85.43 10.33 40.91 8.47 1.69 7.64 1.18 6.67 1.31 3.67 0.53 3.59 0.56
TE38 60.45 184.80 23.03 93.86 19.56 3.32 14.93 1.85 8.55 1.37 3.30 0.43 2.65 0.41
TE39 50.15 98.45 10.86 40.44 7.45 1.32 6.15 1.00 6.02 1.21 3.50 0.52 3.60 0.56
TE40 41.27 98.53 10.65 43.97 10.03 2.28 9.91 1.54 8.73 1.64 4.37 0.61 4.08 0.64
TE41 22.84 43.63 5.10 18.93 3.83 0.75 3.42 0.53 3.09 0.60 1.72 0.25 1.69 0.26
TE42 28.90 57.34 6.55 24.20 4.75 0.85 3.95 0.61 3.49 0.67 1.91 0.28 1.95 0.31
TE44 30.75 60.43 6.89 25.67 5.06 1.00 4.42 0.71 4.04 0.78 2.22 0.33 2.23 0.35
TE45 34.41 67.97 7.69 28.50 5.52 1.05 4.62 0.73 4.28 0.83 2.40 0.36 2.44 0.38
TE47 34.41 66.64 7.58 27.45 4.96 0.93 3.83 0.57 3.15 0.58 1.64 0.24 1.67 0.27
TE48 1.60 3.16 0.36 1.29 0.24 0.04 0.20 0.03 0.18 0.04 0.10 0.01 0.09 0.01
TE50 29.41 62.44 6.89 26.84 5.49 1.28 4.66 0.71 4.22 0.80 2.26 0.33 2.28 0.34
nd not determined. Black schists from the upper Lesser Himalaya are in bold
Environ Earth Sci
123

previously investigated (Sharma et al. 2004; Gurung
et al. 2005; Shah 2008; Singh 2009). In general, these
investigations show that sediment deposition has been
affected by climate change during the Pleistocene–
Holocene period (Sinha and Friend 1994; Sharma et al.
2004). Additionally, studies dealing with arsenic con-
tamination of groundwater in the Bengal fan (e.g.
McArthur et al. 2001; Ishiga et al. 2000; Charlet and
Polya 2006) have demonstrated that high arsenic con-
centrations are restricted to Holocene sediments rich in
organic matter. Here we correlate the stratigraphy of the
Terai sediments with the Bangladesh sediments and dis-
cuss how past climatic conditions can change the aquifer
lithology and arsenic concentrations.
The wood fragment from 7 m depth in borehole 5 gives
a
14
C age between 3354 and 2928 year BC (Table 4). This
age is confirmed by Gurung et al. (2005), who also
obtained a little earlier
14
C age of 3340 ±70 years BP
from a 4 m depth in ND-2 borehole, 2 km ENE from
Unwach village (Fig. 1). We also use Gurung et al.’s
(2005) findings of a 12,680
14
C BP aged black mud deposit
to assign a date to a similar continuous black mud unit
found in the Parasi transect just above 15 m (i.e., boreholes
1 and 3 ‘‘lake deposits’’ in Fig. 2). The depositional envi-
ronment for all black clay beds may be similar and these
beds can be regarded as markers. A black clay layer is also
recognized in the Ganga plain at 13,030 ±114
14
C year
BP in the Sanaki Lake, Central Ganga Plain (Sharma et al.
2004) but also in Bangladesh at around 12,000 year BP
(Acharyya et al. 2000). Overall, we interpret the shared
extent and ages of the black clays as the transition from the
end of Late Glacial Pleistocene period and the beginning of
the warmer Holocene period starting *12,000 year BP
(Zhao et al. 1995).
Fig. 4 Chondrite (McDonough and Sun 1995) normalised REE
patterns of the Nawalparasi sediments. Average composition of
potential source rocks is also reported including our data presented in
Table 3.HHL higher himalayan leucogranites (Guillot and Le Fort
1995; Ayres and Harris 1997); HHC higher himalayan crystallines
(Harris and Inger 1992); LH Lesser Himalaya (Rashid 2002). Siwaliks
(C. France-Lanord, unpublished data). The average compositions of
modern Ganga sediments (Singh 2009) are also reported
Fig. 5 Upper crust (Taylor and
McLennan 1985) normalised
REE patterns of the
Nawalparasi sediments with
average composition of
potential source rocks (same
reference as Fig. 4)
Environ Earth Sci
123

Provenance
Previous studies of the Indo-Gangetic plain, mostly based
on bedload and suspended-load composition (Galy and
France-Lanord 2001), have suggested that Higher Hi-
malayan Crystallines form the major source of the Ganga
plain sediments with the Lesser Himalaya contributing to
only 10–20 % of the sediment load. In contrast, Tripathi
et al. (2007) proposed that the REE homogeneity of the
Ganga Plain sediments shows a high degree of sediment
homogenization through several episodes of sedimentary
reworking and proposed that they most likely represent the
reworking of Siwalik sediments. Moreover, Huyghe et al.
(2005) show that the major source for the Tertiary Siwalik
sediments is the Higher Himalayan Crystallines, and thus it
remains difficult to discriminate if the Ganga plain sedi-
ments derive directly from the Higher Himalayan Crys-
tallines or if they are reworked from the Siwaliks
sediments. To depict the chemical changes recorded by the
sediments from the source region to the deposition site, it is
first necessary to evaluate element mobility. This allows us
to discriminate between the changes controlled by physical
processes, such as sorting, and changes due to chemical
weathering. For that we used the method of Fralick and
Kronberg (1997) reactualised by Lupker et al. (2012) based
on binary diagrams using elements suspected to be
Fig. 6 Variation diagram against the arsenic content. Average composition of the source rocks is from Galy and France-Lanord (2001) including
our data presented in Tables 1and 2.HHC higher himalayan crystallines, LH lesser himalaya, BS black schists
Environ Earth Sci
123

immobile. The binary graphs plot elements against SiO
2
(Fig. 3), gives a first indication about the sediment’s hy-
draulic behaviour and element mobility. Except for Na
2
O,
CaO and MnO, the other elements define a negative linear
array extending towards 100 % SiO
2
. This indicates that
these elements are more chemically immobile and only
record sorting effects as they concentrate in the finer
(clays) fractions. In contrast, Na
2
O, CaO and MnO show
depletion in the finer fractions, suggesting that their con-
centrations should be more sensitive to and predictive of
chemical weathering.
In the binary diagrams (Fig. 7), *70 % of the samples
define a linear array along a line extending towards the
origin. This demonstrates that TiO
2
–Fe
2
O
3
–Al
2
O
3
were
mostly immobile and hydraulically fractionated in a similar
way. Along these lines, the effect of hydraulic sorting is
highlighted with the highest concentration of elements in
the finer fraction (clays) and depleted in coarser fraction
(silt and sand). Considering the linear arrays defined in the
Figs. 3and 7and the similar REE patterns defined by the
four groups of elements (i.e. grey clays, brown clays, cal-
cretes and silts), we suspect that the studied sediments are
Table 4 d
13
C and d
18
O data of the mollusk fossil collected from drill samples of Nawalparasi, Nepal
Samples Location Coordinate (latitude/longitude) Depth below
ground level (m)
d
18
O (PDB) %d
13
C (PDB) %Genus
Dril 1/52–530Suspurwa 27 30 54.3N/83 40 25.8E 16.0 -2.7 ?1.0 Gastropod sp.
Dril 1/55–600Suspurwa 27 30 54.3N/83 40 25.8E 17.5 -9.0 -6.6 Bellamya sp.
PNS 2/550Unwach 27 31 00.5N/83 40 37.4E 16.5 -7.8 -3.5 Gastropod sp.
PNS 2/65B0Unwach 27 31 00.5N/83 40 37.4E 19.5 -5.9 ?0.8 Bellamya sp.
PNS 2/670Unwach 27 31 00.5N/83 40 37.4E 20.1 -2.8 ?1.5 Gastropod sp.
Dril 4/720Jeetpur 27 30 58.3N/83 39 48.1E 22 -3.4 ?1.1 Gastropod sp.
Dril 4/740Jeetpur 27 30 58.3N/83 39 48.1E 22.5 -3.5 -0.1 Bivalvia sp.
PNS 5/140Santapur 27 30 55.0N/83 40 3.8E 4.6 -8.0 -4.9 Gastropod sp.
Fig. 7 Plots of sediments in the bivariant diagram between two
immobile elements. Average compositions of the source rocks are
from Galy and France-Lanord (2001) including our data presented in
Table 1. The trend line including the HHC and the origin shows the
effect of hydraulic sorting whereby the above group of elements
concentrate in finer fraction and are depleted in the coarser fraction.
The points plotting far for this line show the weathering effect. The
symbols and abbreviations are the same as Fig. 6
Environ Earth Sci
123

mainly derived from the same dominant source or sources
having similar chemical composition. Moreover, in a bi-
nary diagram of two immobile elements (Fig. 7), it is ex-
pected that the plots of both finer and coarser sediments
will move away in opposite direction and the tie lines
joining them will pass through their common source
(Fralick and Kronberg 1997). Among the potential sources,
only the Higher Himalayan Crystallines and Lesser Hi-
malaya meet this criteria (Fig. 7). However, some grey clay
and a few brown clays are enriched in Fe
2
O
3
and Al
2
O
3
and define a second line passing through the black schists.
This suggests that these samples specifically derived from
the weathered Lesser Himalaya black schists.
Similarly, in a Fe
2
O
3
/SiO
2
–Al
2
O
3
/SiO
2
diagram
(Fig. 8a), all sediments cluster around the average com-
position of the Higher Himalayan Crystallines and Lesser
Himalayan rocks, suggesting, here again, a predominant
contribution from these two sources. The lability of pla-
gioclase relative to K-feldspar and of K-feldspar relative to
quartz results in substantial modification to the quartz:
plagioclase: K-feldspar proportions by weathering pro-
cesses; thus, the plot Na
2
O/K
2
O versus SiO
2
/Al
2
O
3
can be
used to assess the behaviour of feldspars (Nesbitt et al.
1997). In Fig. 8b, we also observe that about 50 % of the
samples plot along a linear array including the average
Higher Himalayan Crystallines and Lesser Himalaya black
schist composition. This trend reflects feldspar sorting in
suspended sediment during fluvial transport (Nesbitt et al.
1997). We observe again that Siwaliks and Lesser Hi-
malaya are well separated from the trend related to feldspar
sorting, indicating that the Higher Himalayan Crystalline
rocks and black schists from the Lesser Himalaya are the
predominant contributors to the Parasi sediments. The
other 50 % of the samples, including most of the silt/sand
samples and most of the brown clays, plot outside the
linear trend suggesting weathering, reworking, and possible
input by the Siwaliks.
REE elements can also be used to assess the source rock
composition (Taylor and McLennan 1985). All the sedi-
ments in this study show fractionated normalised REE
pattern with average La
N
/Yb
N
ratio of 9.95, 9.74 and 9.62
for grey clays, brown clays and sand/silt, respectively.
Those values are similar to modern sediments from the
Narayani (La
N
/Yb
N
ratio of 9.1) and the Ganga (9.9) ac-
cording to Garc¸on et al. (2013). Deeper Eu anomalies
compared to UCC suggest more silicic source than average
continental crust (Fig. 5). Lesser Himalayan and Higher
Himalayan Crystalline rocks share similar pattern with the
Parasi clay sediments confirming that they are the main
sources (Fig. 5). According to the compilation of available
REE data done by Garc¸on et al. (2013), the La
N
/Yb
N
ratio
of the Parasi sediments are closer to the La
N
/Yb
N
ratio of
the Lesser Himalaya derived-sediments (11.3) than to that
of the Higher Himalayan Crystallines derived-sediments
(4.2), but very close to Siwaliks La
N
/Yb
N
ratio (9.09).
Considering that the REE concentrations in the sources and
in the sediments are similar (150 ±50 ppm), the La
N
/Yb
N
ratio in the sediments and the dominant sources (Higher
Himalayan Crystalline and Lesser Himalaya) can be used
as a proxy to estimate their respective contribution in the
Parasi sediments. Considering that the Upper Himalaya and
the Siwaliks contribute to a maximum of 40 % of the
sediments in the flood plain (Garzanti et al. 2007),we es-
timate that the uppermost part of the Lesser Himalaya
contributed at 45 % and the lower part of the Higher Hi-
malayan Crystallines only at 15 %.
This estimate is in contradiction with previous prove-
nance studies on the Ganga plain sediments, (Galy and
France-Lanord 2001; Singh 2009) suggesting that the
Higher Himalayan Crystallines is the major source and that
the Lesser Himalaya contribute to only 20 % of the sedi-
ment load. Based on petrological and mineralogical studies
on modern sediments in the Marsyandi river catchment in
central Nepal (Fig. 1), Garzanti et al. (2007) proposed that
most of the Ganga sediments derived from the lower part of
the Higher Himalayan Crystallines, within the Main
Fig. 8 Plot of sediments in Fe
2
O
3
/SiO
2
vs. Al
2
O
3
/SiO
2
and Log
Na
2
O/K
2
O vs. log SiO
2
/Al
2
O
3
diagrams showing the pathway of
chemical evolution of feldspar from the source to the plain. Average
compositions of the source rocks are from Galy and France-Lanord
(2001) including our data presented in Table 1. The symbols and
abbreviations are the same as Fig. 6
Environ Earth Sci
123

Central Thrust zone where annual rainfall is maximum at
3000 m and reaches 5 m/year. Thus, erosion rate is max-
imum and quite high in this part of the Higher Himalayan
Crystallines, estimated at 5.1 ±1.2 mm/year. However,
Amidon et al. (2005) show *3 times higher modern ero-
sion rates south of the Main Central Thrust in the north-
ernmost Lesser Himalaya where the black schists outcrop.
Present-day erosion rates also decrease in the upper part of
the Higher Himalayan Crystallines to the north despite the
extreme topography with 7000–8000 m peaks, because
precipitation is too scarce to feed significant ice flux and
glacial activity (Harper and Humphery 2003). Southwards,
in the Lesser Himalaya and Siwaliks, erosion rates are
markedly lower (3 times lower) because rainfall decreases
and relief is lower (Garzanti et al. 2007).We conclude then,
that at the front of the Himalayan belt, Pleistocene/Holo-
cene sediments in the Terai plain in Central Nepal re-
semble modern sediments of the Gangetic alluvial plain
and the main Himalayan river catchments from the up-
stream Narayani including the Kali Gandaki and Mar-
syandi reaches. Sediments shed from these source areas are
geochemically dominated by the uppermost part of the
Lesser Himalaya and the lowermost part of the Higher
Himalayan Crystalline because of focused erosion related
to rapid uplift and high rainfall in the Main Central Thrust
zone (Lave
´and Avouac 2001).
Weathering effects
As already discussed, the chemistry of sediments is con-
trolled by provenance but also by weathering and fluvial
processes (McLennan et al. 1993; Singh 2009).
The A-CN-K triangular plot of Nesbitt et al. (1997)
gives the weathering trend for the sediments (Fig. 9). All
the sediments plot along the same weathering line, con-
firming their similar provenance. The silty fraction plots
near the K-feldspar line and on the lower side of the field of
the Higher Himalayan Crystallines source rocks. This
suggests that these sediments are chemically unweatheared,
and the shift towards the CN end-member reflects sorting
effect with reduction of aluminosilicates phases remaining
in the suspended load. They define a linear trend similar to
the upper reaches sediments of the modern Ganga system.
On the other hand, grey clays plot at the higher side of the
source field indicating strong weathering may be in the
suspended load or after entering into the plains by break-
down of the feldspar fraction, and hence loss of CaO and
Na
2
O. Compared to the modern Ganga interfluve sedi-
ments that are the more mature sediments, the grey clays
plot above, again suggesting more drastic weathering. In
between, the brown clays plot in intermediate position.
To better separate the effects of sorting versus weath-
ering, some of the other diagrams are useful. Using Fig. 7a,
b, and c, 30 % of the samples do no plot on the line
showing this more intense weathering effect. Fe
2
O
3
and
Al
2
O
3
are enriched mostly in the grey clay fractions, while
decreases in SiO
2
content, Na
2
O and CaO (Fig. 3)is
symptomatic of alteration of gneissic rocks under tropical
conditions (Tardy and Nahon 1985). In contrast, some
coarse sediments are only enriched in Al
2
O
3
; as they in-
tersect the tie line in its lower part (Fig. 8b), Al
2
O
3
en-
richments in the silts suggest that they weathered under
different conditions. This hypothesis is confirmed by
Fig. 7d. In this diagram, sand and silt are enriched in MnO
relative to Fe
2
O
3
while clays are depleted in MnO relative
to Fe
2
O
3
. This reflects contrasting redox conditions during
weathering, with more reduced conditions present during
grey clay weathering. As Fe concentrations are higher than
Mn, the majority of the soil cations exchange complex in
reduced conditions is occupied by Fe
2?
, thus leaving Mn
2?
ions in solution. In contrast, in the silt/sand fraction, oxic
conditions precipitated Mn according to the reaction
(Reddy and DeLaune 2008):
2Mn2þþO2þ2H2O¼2 MnO2þ4H
þ
As already discussed, the Na
2
O/K
2
O versus SiO
2
/Al
2
O
3
plot (Fig. 8b) reflects feldspar behaviour during sorting but
also weathering processes. 50 % of the samples, including
most of the silt/sand samples and a large portion of the
brown clays, plot outside of the linear trend suggesting
reworking processes. The upper linear trend reflects sorting
processes with higher concentration of Na
2
O in the coarser
sediments, whereby aluminosilicates (mostly micas, chorite
Fig. 9 A-CN-K plots (Nesbitt et al. 1997) for sediment samples.
Average compositions of the source rocks are from Galy and France-
Lanord (2001) including our data presented in Table 1. Ganga
sediments are from Singh (2009). Notice that the Nawalparasi clays
are much more weathered than the modern Ganga interfluve
sediments suggesting more drastic weathering conditions than today
Environ Earth Sci
123

and illite) are preferably taken in the suspended load, re-
sulting in increased K
2
O and Al
2
O
3
along the trend line
(Galy and France-Lanord 2001).
In the diagram Al
2
O
3
/SiO
2
vs. K
2
O/SiO
2
(Fig. 10), we
report the evolution of the studied samples compared to
modern river sediments covering the whole Ganga basin
from the Himalayan front in Central Nepal to Bangladesh
(data from Lupker et al. 2012). Most of our samples, except
the grey clays, plot along the Himalayan front linear array
reflecting sorting effect without any evidence of chemical
weathering. The grey clays and some brown clays are
shifted downward towards the Ganga plain linear array,
suggesting a loss in K
2
O and a slight enrichment in Al
2
O
3
.
Lupker et al. (2012) interpreted this K
2
O loss in modern
Ganga sediments as reflecting chemical weathering in the
flood plain because of a longer residence time under wet
tropical conditions. Moreover, grey clays are enriched in
organic carbon (Fig. 6). This can reflect successive
weathering from source rocks to plain deposits in the
presence of organic matter (Meharg et al. 2006) or, alter-
natively, the initial content of organic carbon in the source
zone because the black schists contain up to 30 % of car-
bon (Paudel 2012). In continental mudstones, maroon to
purple colours appear to be mostly early diagenetic in
origin and form by alternative wetting and drying on food
plains in semi-arid to arid climates, where little organic
matter is buried. In contrast, wet local conditions and
abundant vegetation favour greyer facies (Potter et al.
2005). Finally, the lower MnO concentrations in the grey
clays fraction suggest weathering under reduced
conditions. Thus, we conclude that the grey clays, enriched
in organic carbon, were possibly developed by wet
weathering processes under reduced conditions.
Relationship between sedimentation, arsenic content
and climate evolution
The climatic condition of the sedimentary environment
during the deposition of the sediments in the inter-fan area
of Nepal’s Narayani and Tinau Rivers has oscillated be-
tween humid and arid regimes as depicted by the stable
isotopic composition of the mollusk fossils (Fig. 11). The
stable isotopic compositions of freshwater mollusk shells
have been used to characterize the climatic conditions as-
sociated to sedimentary environments in the Indo-Gangetic
plain (Gurung et al. 2005; Gajurel et al. 2006; Sharma et al.
2004). The average d
13
C and d
18
O values of modern
freshwater mollusk shells in the Indo-Gangetic plain are
reported as -9.5 and -7.4 %, respectively. The d
13
C
values in the Parasi drilling biogenic sediments are outside
of d
13
C from the surface shell, and d
18
O values are
otherwise highly deviated from the reported averages
(Fig. 11). The enriched C and O isotopic compositions of
the biogenic carbonate possibly indicate a dry climatic
regime, while the depleted values potentially suggest more
humid climatic condition. This is relative to d
13
C[-7%,
and d
18
O[-4%, which are modern values (Fig. 11).
Three arid periods are clearly recognized at 23.5, 20 and
11 m. These periods coincide with relatively higher arsenic
content, up to 30 ppmat 20 m depth (Fig. 11). Using our
14
C age data, with the data of Gurung et al. (2005), we
approximately date these periods at \22 kyr BP for the
two oldest periods (23.5 and 20 m depth) and at 12 kyr BP,
at the Pleistocene–Holocene transition for the youngest
period at 11 m. The transition from arid to more humid
period at 17–18 m agrees with a *22 kyr isotopic arid-
humid transition recorded in marine cores (Hayashi et al.
2008; Gibling et al. 2005; Reichart et al. 1998; Tzedakis
et al. 1997) and by a decrease in the extent of glaciations in
the Nepal Himalaya due to reduced precipitation (Finkel
et al. 2003; Tsukamoto et al. 2002). Between 18 and 16 m
depths, the d
13
C and d
18
O in the fossil shells strongly de-
crease, suggesting a more humid period preceding the
Pleistocene’s Late Glacial Maximum (Zhao et al. 1995).
Above the Pleistocene series, the development of the
2 m thick black clay layers is observed from the Terai to
the Bengal fan and well dated between 13 and 10 kyr BP
(Sharma et al. 2004; Gurung et al. 2005). It marks the
beginning of the Holocene period and its warmer and hu-
mid climate. Nevertheless, the upper 13 m thick sedimen-
tary deposit was also accumulated during several humid
and arid climatic oscillations (Zhao et al. 1995; Sharma
et al. 2004). Moreover, the boundaries between humid and
Fig. 10 Evolution of the K
2
O/SiO
2
ratio as a function of Al
2
O
3
/K
2
O,
showing a depletion in K
2
O and a slight enrichment in Al
2
O
3
from the
Himalayan front rivers downstream to the Ganga plain in Bangladesh
(data from Lupker et al. 2012). We reported our samples in this
diagram showing the weathering effect in the foot plain. Average
compositions of the source rocks are from Galy and France-Lanord
(2001) including our data presented in Table 1
Environ Earth Sci
123

arid climatic oscillations in the area are generally marked
by calcrete bearing paleosols (Figs. 2,11). At about
6 kyr BP, the monsoon intensification was five times
higher that of today (about 200 mm/year) in India (Bryson
and Swain 1981) corresponding to the worldwide optimum
climatic (Kaufman et al. 2004). During this period, in the
Nawalparasi area, an important amount of point bar and
overbank sediments with lacustrine facies, as those oc-
curring between 8 and 13 m depth, were deposited (Fig. 2)
and the sediment-arsenic content remains low to very low
(Fig. 11). Towards the upper part of the section, a sig-
nificant calcrete bearing paleosol with dominant overbank
deposits was developed at around 4 m depth that can co-
incide with the period of reduced monsoon precipitation in
Fig. 11 Synthetic log of the
Nawalparasi area showing the
relationship between the
lithology along the five studied
boreholes and the boreholes
studied of Gurung et al. (2005),
the arsenic concentration the C
and O isotopes and the inferred
climate variation (see text for
‘‘Discussion’’ )
Environ Earth Sci
123

Midlands of Nepal between 2.3 and 1.5 kyr BP (Denniston
et al. 2000) and corresponds to a slight increase of the
arsenic content at 6 m depth (Gurung et al. 2005).
The relationship between past sedimentary environ-
ments and the climatic regimes for the concentration of
arsenic in the Parasi sediments is an important aspect that
needs to be addressed to understand how arsenic accumu-
lates in the Indo-Gangetic plain. In most of the sedimentary
rocks (e.g. Smedley and Kinniburgh 2002), arsenic is
concentrated in finer clay fractions, possibly reflecting
accumulation in secondary iron or aluminium oxide phases.
Indeed, we observe a good relationship between Al
2
O
3
,
Fe
2
O
3
and arsenic content in the studied sediments (Fig. 6).
Moreover, we also observe a rather good correlation be-
tween K
2
O and arsenic content, suggesting that biotite or
almost-weathered biotite contributed to the fixation of this
element (Nath et al. 2009). Similarly, a rough relationship
exists between the organic content and the arsenic content
(Fig. 6). In contrast, no relationship is observed between
MnO, CaO, and P
2
O
3
, suggesting that ferromanganese
nodules, carbonates, or apatite do not contribute to the
arsenic enrichments in clay minerals, as observed in the
Bengal fan sediments (Smedley and Kinniburgh 2002;
Plant et al. 2003; van Geen et al. 2008).
As previously discussed, there is likely a correlation
between the late Quaternary climate conditions and the
concentration of arsenic in the sediments; arsenic is ob-
served to preferentially concentrate in sediments deposited
during more arid periods. The lower arsenic concentrations
in sediments plain during humid periods can be explained
by the process of arsenic eluviations in sandy and silty
sediments due to intense summer monsoon rain during
interglacial periods (Gourlan et al. 2010). Three factors
seem to dominantly explain the enrichment of arsenic
during the arid periods in the grey clays. Arsenic concen-
trates in clay sediments, associated with specific elements
(FeO, Al
2
O
3
,K
2
O and C) and linked to the degree of al-
teration (Fig. 6). In arid periods, when the rainfall was
reduced, there was minor terrigenous clastic input from the
Higher Himalayan Crystallines relative to the black schists
from the Lesser Himalaya but also from watershed soil
erosion, which favoured the development of swamp lands
in the Ganga plain (Sharma et al. 2004). This environment
favoured the development of aquatic plants and bacteria
growing within the moist land areas, enhancing the strong
weathering of initially suspended load particles (micas,
clays), which were preferentially deposited inquiet hy-
draulic environments. In this environment, clays mostly
deriving from C-rich black schists are chemically weath-
ered losing Na
2
O and K
2
O relative to the source and
relatively enriched in immobile elements such as Al
2
O
3
and Fe
2
O
3
. During this chemical weathering process,
arsenic remains relatively immobile until reduced
conditions persist. Nevertheless, except in one sample, we
do not observe an over-enrichment in arsenic compared to
the source.
Conclusion
The average arsenic concentration in the Terai sediments is
within the range of normal sediments (8 ppm) but the dis-
tribution is not homogeneous throughout the local aquifer.
Abundances are greater in finer sediments, particularly in the
black to grey clays (maximum 27 ppm) than in coarser silts
and fine sands (3 ppm). A positive correlation exists between
arsenic concentrations and clay content. In contrast, there is
no correlation between arsenic and calcium content. The
sediments represent homogeneous mixtures of parent rocks
of felsic origin. Correlation diagrams of major elements and
REE patterns suggest that the dominant source of the Late
Pleistocene to Early Holocene sediments was the uppermost
part of the Lesser Himalaya and the Higher Himalayan
Crystalline with a possible input from the Siwaliks. In this
context, the dominant source of arsenic in the present-day
aquifer is the black schists from the Lesser Himalaya. The
results of O and C isotopic analyses indicate a possible
linkage between late Quaternary climate conditions and
concentration of arsenic in the sediments. During the
last *25 kyr BP, arsenic seems to be preferentially con-
centrated in sediments deposited during arid periods. Lower
arsenic concentrations during more humid periods can be
explained by a process of arsenic eluviations in sandy and
silty sediments. In contrast, during dry periods, the seasonal
precipitation was smaller leading to local wetter soils. This
environment favoured the development of aquatic plants and
bacteria growing within the moist land areas, enhancing the
strong weathering of initially suspended load particles (mi-
cas, clays), which were preferentially deposited in quite
hydraulic and reduced environments.
Acknowledgments The research project was supported by NSF, the
CNRS INSU EC2CO and Labex OSUG20@20 programs. We ac-
knowledge James W. LaMoreaux and the two anonymous reviewers
for fruitful comments.
References
Acharyya SK, Lahiri S, Raymahashay BC, Bhowmik A (2000)
Arsenic toxicity of groundwater in parts of the Bengal basin in
India and Bangladesh: the role of Quaternary stratigraphy and
Holocene sea-level fluctuation. Environ Geol 39:1127–1137
Amidon WH, Burbank DW et al (2005) U-Pb Zircon ages as a
sediment mixing tracer in the Nepal Himalaya. Earth Planet Sci
Lett 235:244–260
Ayres M, Harris N (1997) REE and Nd-Isotope fractionation during
crustal anatexis: constraints from Himalayan leucogranites.
Chem Geol 139:249–269
Environ Earth Sci
123

Bryson RA, Swain AM (1981) Holocene variations of monsoon
rainfall in Rajasthan. Quat Res 16:125–145
Charlet L, Polya D (2006) Arsenic in shallow reducing groundwaters
in southern Asia: an environmental health disaster. Elements
2:91–96
Chauvel C, Bureau S, Poggi C (2011) Comprehensive chemical and
isotopic analyses of basalt and sediment standards. Geost
Geoanal Res 35:125–143
Colchen M, Le Fort P, Pe
ˆcher A (1986) Recherches ge
´ologiques dans
l’Himalaya du Ne
´pal. Annapurna, Manaslu, Ganesh. Paris: Ed.
du Centre national de la recherche scientifique, p 136
Cotten J, Le Deza A, Baub M, Caroff RC, Maury RC, Dulski P,
Fourcade S, Bohn M, Brousse R (1995) Origin of anomalous
rare-earth element and yttrium enrichments in subaerially
exposed basalts: evidence from French Polynesia. Chem Geol
119:115–138
Denniston RF, Gonza
´lez LA, Asmerom Y, Sharma RH, Reagan MK
(2000) Speleothem evidence for changes in Indian summer
monsoon precipitation over the last *2300 years. Quat Res
53:196–202
Dettman DL, Reische AK, Lohamann CK (1999) Controls on the
stable isotope composition of seasonal growth bands in
aragonitic fresh-water bivalve (Unionidae). Geochim Cos-
mochim Acta 63:1049–1057
Finkel RC, Owen LA, Barnard PL, Cafee MW (2003) Beryllium-10
dating of Mount Everest moraines indicates a strong monsoon
influence and glacial synchroneity throughout the Himalaya.
Geology 31:561–564
Fralick PW, Kronberg BI (1997) Geochemical distribution of clastic
sedimentary rock source. Sediment Geol 113:111–124
France-Lanord C, Derry L, Michard A (1993) Evolution of the
Himalaya since Miocene time: isotopic and sedimentologic
evidence from the Bengal fan. In: Treloar PJ, Searle M (eds)
Himalayan tectonics. Geol Soc Spe Pub, London 74:445–465
Gajurel AP, France-Lanord C, Huyghe P, Guilmette C, Gurung D
(2006) C and O Isotope compositions of modern fresh-water
mollusc shells and river waters from Himalaya and Ganga plain.
Chem Geol 233:156–183
Galy A, France-Lanord C (2001) Higher Erosion rates in the
Himalaya: geochemical constraints on riverine fluxes. Geology
29:23–26
Galy A, France-Lanord C, Derry LA (1999) The strontium isotopic
budget of Himalayan rivers in Nepal and Bangladesh. Geochim
Cosmochim Acta 63:1905–1925
Garzanti E, Vezzoli G, Ando S, France-Lanord C, Singh SK, Foster G
(2004) Sand Petrology and focused erosion in collision orogens:
the Brahmaputra case. Earth Planet Sci Lett 220:157–174
Garzanti E, Vezzoli G, Ando S, Lave
´J, Attal M, France-Lanord C,
DeCelles PG (2007) Quantifying sand provenance and erosion
(Marsyandi River, Nepal Himalaya. Earth Planet Sci Lett
258:500–515
Garc¸on M, Chauvel C, France-Lanord C (2013) Sedimentary
processes decouple Nd and Hf isotopes in river sediments on
continents. Geochim Cosmochim Acta 121:177–195
Gibling MR et al (2005) Discontinuity-bounded alluvial sequences of
the southern Gangetic Plains, India: aggradation and degradation
in response to monsoonal strength. J Sedim Res 75:369–385
Gourlan A, Meynadier L, Alle
`gre CJ, Tapponnier P, Birck JL, Joron
JL (2010) Northern Hemisphere climate control of the Bengali
rivers discharge during the past 4 Ma. Quat Res 29:2484–2498
Guillot S, Charlet L (2007) Bengal arsenic, an archive of Himalaya
orogeny and paleohydrology. J Environ Sci Health Part
A42:1785–1794
Guillot S, Le Fort P (1995) Geochemical constraints on the bimodal
origin of high Himalayan leucogranites. Lithos 35:221–234
Guillot S, LeFort P, Pe
ˆcher A, Barman MR, Aprahamian J (1995)
Contact metamorphism and depth of emplacement of the
manaslu granite (Central Nepal). Implications for Himalayan
orogenesis. Tectonophysics 241:99–119
Guillot S (1999) An overview of the metamorphic evolution of central
Nepal. In: Upreti BN, Le Fort PJ (eds) ‘‘Geology of Nepal’’.
J Asian Earth Sci 17:713–725
Gurung JK, Ishiga H, Khadka M (2005) Geological and geochemical
examination of arsenic contamination in groundwater in the
Holocene Terai Basin, Nepal. Environ Geol 49:98–113
Harper JT, Humphery NF (2003) High altitude Himalayan climate
inferred from glacial ice flux. Geophy Res Lett. doi:10.1029/
2003GLO17329
Harris N, Inger S (1992) Trace element modelling of pelite-derived
granites. Contrib Miner Petrol 110:46–56
Hayashi T, Tanimura Kuwahara Y, Ohno M, Mampuku M, Fujii R,
Sakai H, Yamanaka T, Maki T, Uchida M, Yahagi W, Sakai H
(2008) Ecological variations in diatom assemblages in the Paleo-
Kathmandu Lake linked with global and Indian monsoon climate
changes for the last 600,000 years. Quat Res 72:377–387
Huyghe P, Mugnier JL, Gajurel AP, Delcaillau B (2005) Tectonic and
climatic control of the changes in the sedimentary record of the
Karnali River section (Siwaliks of western Nepal). Isl Arcs
14:311–325. doi:10.1111/j.1440-1738.2005.00500.x
Ishiga H, Dozen K, Yamazaki CFA, Islam MB, Rohman MH, Sattar
MA, Yamamoto H, Itoh K (2000) Geological constraints on
arsenic contamination of groundwater in Bangladesh. In:
Proceedings of the 5th forum of Arsenic in Asia, Nov 2000
Asia Arsenic Network (AAN), Yokohama Japan, pp 53–62
Kaufman DS, Ager TA, Anderson NJ, Anderson PM, Andrews JT,
Bartlein PJ, Brubaker LB, Coats LL, Cwynar LC, Duvall ML,
Dyke A, Edwards ME, Eisner WR, Gajewski K, Geirsdottir A,
Hu FS, Jennings AE, Kaplan, Kerwin MW, Lozhkin AV,
MacDonald GM, Miller GH, Mock CJ, Oswald WW, Otto-
Bliesner BL, Porinchu DF, Ruhland K, Mol JP, Steig EJ, Wolfe
BB (2004) Holocene thermal maximum in the western Arctic
0–180 W. Quat Sci Rev 23:529–560. doi:10.1016/j.quascirev.
2003.09.007
Lave
´J, Avouac JP (2001) Fluvial incision and tectonic uplift across
the Himalayas of central Nepal. J Geophy Res
106:26561–26591. doi:10.1029/2001JB000359
Lupker M, France-Lanord C, Galy V, Lave
´J, Gaillardet J, Gajurel
AP, Guilmette C, Rahman M, Singh SK, Sinha R (2012)
Predominant floodplain over mountain weathering of Himalayan
sediments Ganga basin. Geochim Cosmochim Acta 84:410–432
Le
´cuyer C, Reynard B, Martineau F (2004) Stable isotope frac-
tionation between mollusc shells and marine waters from
Martinique Island. Chem Geol 213:293–305
McArthur JM, Ravenscroft P, Safiullah S, Thirlwall MF (2001)
Arsenic in groundwater: testing pollution mechanism for
sedimentary aquifers in Bangladesh. Water Res Res 37:109–117
McCrea JM (1950) On the isotopic chemistry of carbonates and a
paleotemperature scale. J Chem Phys 18:849–857
McDonough W, Sun SS (1995) The composition of the earth. Chem
Geol 120:223–253
McLennan SM, Hemming S, McDennial DK, Hanson GN (1993)
Geochemical approaches to sedimentation provenance and
tectonics. Geol Soc Amer Bull 284:21–40
Meharg AA, Scrimgeour C, Hossain SA, Fuller K, Cruickshank K,
Williams PN, Kinniburgh DG (2006) Codeposition of organic
carbon and arsenic in Bengal delta aquifers. Environ Sci Technol
40:4928–4935
Milliman JD, Sivitsli JPM (1992) Geomorphic/tectonic control of
sediment discharge to the ocean: the importance of small
mountainous rivers. J Geol 100:525–544
Environ Earth Sci
123

Mugnier JL, Leturmy P, Huyghe P, Chalaron E (1999) The Siwaliks
of Western Nepal: mechanism of the Thrust Wedge. In: Le Fort
P, Upreti BN (eds) Geology of the Nepal Himalaya: recent
advances. J Asia Earth Sci 17:643–657
Nesbitt HW, Fedo CM, Young GM (1997) Quartz and feldspar
stability, steady and non-steady-state weathering and petroge-
nesis of siliclastic sands and muds. J Geol 105:173–191
Nath B, Chakraborty S, Burnol A, Stu
¨ben D, Chatterjee D, Charlet L
(2009) Mobility of arsenic in the subsurface environment: an
integrated hydrogeochemical study and sorption model of the
sandy aquifer materials. J Hydrol 364:236–248
Paudel LP (2012) Carbonaceous schists of the Main Central Thrust
zone as a source of graphite: a case study from the Kali Gandaki
valley, west Nepal. Bull Dept Geol Tribhuvan Univ Kathmandu
Nepal 14:9–14
Plant JA, Kinniburgh DG, Smedley PL, Fordyce FM, Klinck BA
(2003) Arsenic and selenium. Treatise on geochemistry. In:
Lollar BS, Heinrich D. Holland, Karl K (eds) Turekian9. doi:10.
1016/B0-08-043751-6/09047-2:17-66
Pokhrel D, Bhandari BS, Viraraghavan T (2009) Arsenic contamina-
tion of groundwater in the Terai region of Nepal: an overview of
healths concerns and treatment options. Environ Geol
35:157–161
Potter PE, Maynard JB, Depteris P (2005) Mud and mudstones,
introduction and overview. Springer-verlag, Berlin, p 218
Rashid SA (2002) Geochemical characteristics of Mesoproterozoic
clastic sedimentary rocks from the Chakrata Formation, Lesser
Himalaya; implications for crustal evolution and weathering
history in the Himalaya. J Asian Earth Sci 21:283–293
Reddy KR, DeLaune RD (2008) Biogeochemistry of wetlands:
science and applications. Taylor and Francis group, London 774
Reichart GJ, Lourens LJ, Zachariasse WJ (1998) Temporal variability
in the northern Arabian Sea Oxygen Minimum Zone OMZ
during the last 225,000 years. Paleoceanography 13:607–621
Shah BA (2008) Role of Quaternary stratigraphy on arsenic-
contaminated groundwater from parts of Middle Ganga Plain,
UP-Bihar, India. Environ Geol 53:1553–1561
Sharma S, Joachimski M, Sharma ML, Tobschal HJ, Singh IB,
Sharma C, Chaulan MS, Morgenroth G (2004) Late glacial and
Holocene environmental changes in Ganga plain, Northern India.
Quat Sci Rev 23:145–159
Shukla U, Bora K (2009) Sedimentation model of gravel-dominated
alluvial piedmont fan, Ganga Plain, India. Int Earth Sci
98:443–459
Singh P (2009) Major, trace and REE geochemistry of the Ganga
River sediments: influence of provenance and sedimentary
process. Chem Geol 266:2516264
Sinha R, Friend KH (1994) River systems and their flux, Indo-
Gangetic plains, Norther Bihar, India. Sedimentology
41:825–845
Smedley PL, Kinniburgh DG (2002) A review of the source,
behaviour and distribution of arsenic in natural waters. Appl
Geochem 17:517–568
Tardy Y, Nahon D (1985) Geochemistry of laterites, stability of Al-
goethite, Al-hematite, and Fe3?-kaolinite in bauxites and
ferricretes: an approach to the mechanism of concretion forma-
tion. Am J Sci 285:865–903
Taylor SR, McLennan SM (1985) The continental crust: its compo-
sition and evolution. Oxford Blackwell, London 312
Tripathi JK, Ghazanfari P, Rajamani V, Tandon SK (2007)
Geochemistry of sediments of the Ganges alluvial plains:
evidence of large-scale sediment recycling. Quat Int
159:119–130
Tsukamoto S, Asahi K, Watanabe T, Rink WJ (2002) Timing of past
glaciations in Kanchenjunga Himal, Nepal by optically stimulat-
ed luminescence dating of tills. Quat Int 97–98:57–67
Tzedakis PC, Andrieu V, De Beaulieu JL, de Crowhurst S, Follieri M,
Hooghiemstra H, Magri D, Reille M, Sadori L, Shackleton NJ,
Wijmstra TA (1997) Comparison of terrestrial and marine
records of changing climate of the last 500,000 years. Earth
Planet Sci Lett 150:171–176
UNDP and Nepal HMGO (1989) Shallow groundwater investigation
in Terai, Nawalparasi district West. Technical report 5 p 21
Upreti BN (1999) An overview of the stratigraphy and tectonics of the
Nepal Himalaya. J Asian Earth Sci 17:577–606
van Geen A, Radloff K, Aziz Z, Cheng Z, Huq MR, Ahmed KM,
Weinman B, Goodbred S, Jung HB, Zheng Y, Berg M, Trang
PTK, Charlet L, Metral J, Tisserand D, Guillot S, Chakraborty P,
Gajurel AP, Upreti BN (2008) Comparison of arsenic concen-
trations in simultaneously-collected groundwater and aquifer
particles from Bangladesh, India, Vietnam, and Nepal. Appl
Geochem 23:3244–3251
Zhao M, Beveridge NAS, Shackleton NJ, Sarnthein M, Eglinton G
(1995) Sediment core ODP 658, interpreted sea surface
temperature, Eastern Tropical Atlantic. Paleoceanography
10:661–675. doi:10.1029/94PA03354
Environ Earth Sci
123