Research Square (Research Square),
Journal Year:
2021,
Volume and Issue:
unknown
Published: May 7, 2021
Abstract
Groundwater
supports
agriculture
and
provides
domestic
water
for
over
250
million
people
in
the
Bengal
Basin.
Our
analysis
of
stable
isotope
ratios
rain,
surface,
groundwater
shows
that
proportion
recharge
originating
from
stagnant
surface
bodies
has
increased
by
about
50%
last
seventy
years
while
relative
contribution
direct
infiltration
rain
decreased.
This
regional
shift
source
how
simultaneous
expansion
irrigated
rice,
excavated
ponds
pumping
changed
hydrologic
system
cycling
evaporated
standing
through
subsurface.
Analysis
data
also
reveals
most
enters
during
latter
part
dry
season
(February-April),
rainwater
occurs
early
months
monsoon
(June-August)
before
aquifers
fill
to
capacity
reject
additional
rainwater.
Hydrological Processes,
Journal Year:
2020,
Volume and Issue:
34(16), P. 3506 - 3523
Published: May 26, 2020
Abstract
Groundwater
supplies
a
significant
proportion
of
water
use
in
the
United
States
and
is
critical
to
maintenance
healthy
ecosystems
environmental
processes,
thus
characterizing
aquifer
hydrology
important
managing
preserving
these
resources.
While
groundwater
isotopes
provide
insight
into
hydrologic
ecologic
their
application
limited
where
measurements
exist.
To
help
overcome
this
limitation,
we
used
random
forest
algorithm
develop
predictive
model
for
shallow
conterminous
States.
Our
uses
variables
(e.g.
temperature,
elevation,
precipitation
isotopes)
as
predictors.
We
our
first
isoscape
δ
2
H
18
O
describe
patterns
using
both
observations
modelled
isoscape.
find
that
throughout
much
Eastern
States,
are
close
annual
amount
weighted
precipitation,
while
significantly
depleted
relative
across
High
Plains
Western
Furthermore,
compare
compiled
study
which
allows
us
determine
recharge
efficiency
(i.e.
ratio
precipitation)
between
seasons
occurs
given
season.
findings
suggest
winter
generally
more
efficient
than
summer
recharge;
however,
dominant
season
varied
it
product
seasonal
timing
precipitation.
Parts
central
have
recharge,
likely
result
heavy
precipitation/nocturnal
Interestingly,
parts
coastal
California
appear
could
be
due
from
fog‐drip.
results
summarize
spatial
processes
affecting
groundwater,
valuable
information
future
studies.
Environmental Science & Technology Letters,
Journal Year:
2020,
Volume and Issue:
7(12), P. 916 - 922
Published: Sept. 30, 2020
Iron
oxides
control
the
mobility
of
a
host
contaminants
in
aquifer
systems,
and
microbial
reduction
iron
subsurface
is
linked
to
high
levels
arsenic
groundwater
that
affects
greater
than
150
million
people
globally.
Paired
observations
solid-phase
composition
are
critical
understand
spatial
temporal
trends
contamination
effectively
manage
changing
water
resources,
yet
field-representative
mineralogical
data
sparse
across
redox
gradients
relevant
contamination.
We
characterize
mineralogy
using
X-ray
absorption
spectroscopy
natural
gradient
Vietnam.
Hierarchical
cluster
analysis
classifies
sediments
into
meaningful
groups
delineating
weathering
changes,
diagnostic
depositional
history,
this
first
direct
characterization
transformations
field.
Notably,
these
groupings
reveal
signature
minerals
undergoing
active
before
onset
groundwater.
Pleistocene
postdepositional
may
be
more
extensive
previously
recognized
due
previous
misclassification.
By
upscaling
similar
environments
South
Southeast
Asia
via
multinomial
logistic
regression
modeling,
we
show
reduction,
therefore
susceptibility
future
contamination,
widely
distributed
presumably
pristine
aquifers
anticipated.
Ground Water,
Journal Year:
2021,
Volume and Issue:
60(3), P. 362 - 376
Published: Dec. 24, 2021
Manganese
(Mn)
concentrations
and
the
probability
of
arsenic
(As)
exceeding
drinking-water
standard
10
μg/L
were
predicted
in
Mississippi
River
Valley
alluvial
aquifer
(MRVA)
using
boosted
regression
trees
(BRT).
BRT,
a
type
ensemble-tree
machine-learning
model,
created
predictor
variables
that
affect
Mn
As
distribution
groundwater.
These
included
iron
(Fe)
specific
conductance
from
previously
developed
BRT
models,
groundwater
flux
age
estimates
MODFLOW,
hydrologic
characteristics.
The
models
also
results
first
airborne
geophysical
survey
conducted
United
States
to
target
an
entire
system.
Predictions
high
occurred
where
Fe
was
high.
Predicted
correlated
with
fraction
young
(less
than
65
years)
computed
MODFLOW
results.
High
probabilities
exceedance
relatively
old
electromagnetic
resistivity
high,
typically
proximal
streams.
Two-variable
partial-dependence
plots
sensitivity
analysis
used
provide
insight
into
factors
controlling
maps
can
be
identify
areas
these
constituents
may
could
targeted
for
further
study.
This
paper
shows
incorporation
selected
set
process-informed
data,
such
as
geophysics,
model
improves
interpretability.
Incorporation
process-rich
information
will
likely
useful
addressing
wide
range
problems
interest
hydrologists.
