LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
1
4.0
Bank
Filtration
4.1
Introduction
Bank
filtration
is
a
surface
water
pretreatment
process
that
uses
the
bed
and
bank
of
a
river
(
or
lake)
and
the
adjacent
aquifer
as
a
natural
filter.
In
optimal
locations
and
under
optimal
conditions,
bank
filtration
is
suitable
for
accomplishing
sufficient
Cryptosporidium
removal
to
partially
meet
the
requirements
of
the
Long
Term
2
Enhanced
Surface
Water
Treatment
Rule.
To
accomplish
this,
a
pumping
well
located
in
the
adjacent
aquifer
induces
surface
water
infiltration
through
the
bed
and
bank.
Bank
filtration
differs
significantly
from
artificial
recharge
and
from
aquifer
storage
and
recovery,
both
of
which
rely
on
engineering
works
to
move
water
into
specially
constructed
and
maintained
recharge
basins
or
wells
for
infiltration
into
or
replenishment
of
the
aquifer.
Although
microorganism
removal
can
occur
in
such
engineered
systems,
they
are
not
bank
filtration.
This
is
because
bank
filtration
relies
solely
on
the
natural
properties
of
the
surface
water
bed
and
aquifer,
unmodified
by
engineered
works
or
activity,
except
for
the
recovery
of
ground
water
via
a
pumping
well.

A
significant
proportion
of
microorganisms
and
other
contaminants
are
removed
by
contact
with
the
aquifer
material
as
the
water
travels
to
the
well
through
the
subsurface.
Flow
to
the
well
may
be
horizontal
or
vertical,
but
more
typically
will
take
a
variable
path
with
both
horizontal
and
vertical
components.
The
water
which
has
been
induced
to
infiltrate
through
the
river's
bed
and
bank
is
known
as
"
bank
filtrate."
It
will
be
mixed
with
ambient
ground
water
that
has
taken
a
different
and
typically
longer
path
to
the
well.
The
ambient
ground
water
may
have
originated
as
bed
or
bank
infiltration
from
an
upstream
portion
of
the
river
or
from
a
lake.
It
may
have
originated
from
infiltrating
precipitation.
Regardless,
ambient
ground
water
is
likely
to
contain
different
contaminants
and
contaminant
concentrations
than
bank
filtrate
because
its
origin
and
flow
pathways
differ
significantly.
Ambient
ground
water
should
not
be
assumed
to
be
uncontaminated.

Aquifers
suitable
for
bank
filtration
are
composed
of
unconsolidated,
granular
material
(
i.
e.,
grains)
and
have
open,
interconnected
pores
that
allow
ground
water
to
flow.
Pathogen
removal
is
enhanced
when
fine­
grained
sediment
is
present
along
the
flow
path.
Geologic
units
consisting
primarily
of
fine­
grained
(
e.
g.,
clay­
sized)
materials
will
have
higher
removal
but
will
be
incapable
of
yielding
economically
significant
water
flow
rates.
In
aquifers
containing
both
sandsized
and
finer
grains,
the
presence
of
fine
grains
increases
the
possibility
that
pathogens
will
encounter
a
grain
surface.
This
is
because
flow
is
slower
and
flow
paths
are
longer
than
they
would
be
in
aquifers
without
such
fine
grains.
Microorganisms
will
be
removed
from
flow
as
they
contact
and
attach
to
grain
surfaces.
Although
microorganism
(
e.
g.,
Cryptosporidium)
detachment
can
occur,
it
usually
does
so
at
slow
rates
(
Harter
et
al.,
2000).
When
little
or
no
detachment
occurs
or
when
detachment
is
slow,
microorganisms
can
become
non­
viable
while
attached
to
grain
surfaces.
Thus,
bank
filtration
provides
physical
removal,
and
in
some
cases,
inactivation,
to
remove
pathogens
from
water
supplies.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
2
The
purposes
of
this
chapter
are:
1)
to
clarify
the
requirements
of
the
LT2ESWTR
related
to
receiving
Cryptosporidium
removal
credit
for
the
use
of
bank
filtration
systems
2)
to
present
the
current
state­
of­
the­
science,
advantages
and
disadvantages
of
Cryptosporidum
removal
by
bank
filtration;
3)
to
explain
how
local
geologic
and
hydrologic
conditions
affect
the
functioning
and
effectiveness
of
bank
filtration
systems;
and
4)
to
provide
suggestions
for
optimal
operation
of
bank
filtration
systems.

This
chapter
is
organized
as
follows:

4.2
LT2ESWTR
Compliance
Requirements
­
describes
requirements
for
receiving
Cryptosporidium
removal
credits
related
to
the
proposed
installation
of
bank
filtration
wells.

4.3
Toolbox
Selection
Considerations
­
describes
the
advantages
and
disadvantages
of
using
bank
filtration
as
a
pretreatment
technology.

4.4
Site
Selection
and
Aquifer
Requirements
­
characterizes
surface
water
and
aquifer
types
that
are
suitable
for
bank
filtration.

4.5
Design
and
Construction
­
describes
the
types
of
wells
eligible
for
bank
filtration
credits
and
the
locations
at
which
such
wells
are
best
placed.

4.6
Operational
Considerations
­
describes
issues
relevant
to
the
optimal
operation
of
bank
filtration
systems
in
order
to
protect
public
health.

4.2
LT2ESWTR
Compliance
Requirements
Systems
that
propose
to
install
bank
filtration
wells
to
meet
any
additional
treatment
requirements
imposed
by
the
LT2ESWTR
may
be
eligible
for
0.5
or
1.0
log
Cryptosporidium
removal
credit
(
40
CFR
141.726(
c)).
Systems
meeting
all
regulatory
requirements
(
e.
g.
systems
with
conventional
or
direct
filtration
that
meet
the
well
siting
requirements)
receive
Cryptosporidium
log
removal
credit
prior
to
construction
of
the
production
wells.
For
those
systems
which
already
use
bank
filtration
as
a
component
of
their
treatment
process
and
which
also
have
existing
conventional
or
direct
filtration
treatment,
the
LT2ESWTR
requires
source
water
monitoring
of
produced
water
from
the
bank
filtration
well.
This
will
determine
the
initial
bin
classification
for
these
systems.
Because
their
source
water
monitoring
accounts
for
any
bank
filtration
treatment,
these
systems
are
not
eligible
for
subsequent
additional
bank
filtration
credits
(
40
CFR
141.704).

Systems
using
ground
water
under
the
direct
influence
(
GWUDI)
of
surface
water
or
bank
filtered
water
without
additional
filtration
must
take
source
water
samples
in
the
surface
water
to
determine
bin
classification
(
40
CFR
141.704).
This
applies
to
systems
using
an
alternative
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
3
filtration
demonstration
to
meet
the
Cryptosporidium
removal
requirements
of
the
IESWTR
or
LT1ESWTR
(
40
CFR
141.173(
b)
and
141.552(
a)).
As
a
result,
the
requirements
and
guidance
provided
in
this
chapter
do
not
apply
to
existing
primacy
agency
actions
providing
alternative
filtration
Cryptosporidium
removal
credit
for
IESWTR
or
LT1ESWTR
compliance.

4.2.1
Credits
The
LT2ESWTR
specifies
the
following
design
requirements
for
systems
to
receive
log
removal
credit
for
bank
filtration
(
40
CFR
141,
Subpart
W,
Appendix
A):

°
Wells
must
draw
from
granular
aquifers
that
are
comprised
of
clay,
silt,
sand,
or
pebbles
or
larger
particles.
Minor
cement
may
be
present.

°
The
aquifer
material
must
be
unconsolidated,
with
subsurface
samples
friable
upon
touch.

°
Granular
aquifers
formed
by
alluvial
or
glacial
processes
are
eligible
for
bank
filtration
credit.

°
Granular
aquifers,
either
unconsolidated
or
partially
consolidated,
and
mapped
as
earlier
than
Quaternary
alluvium,
must
be
considered
on
a
case­
by­
case
basis
by
the
state
to
determine
if
they
are
too
cemented,
and
therefore
too
fractured,
to
provide
sufficient
natural
filtration.

°
Wells
located
in
consolidated
clastic
aquifers
(
e.
g.,
conglomerates),
fractured
bedrock
aquifers,
and
karst
limestone
aquifers
are
not
eligible
for
bank
filtration
credit.

°
Only
horizontal
and
vertical
wells
are
eligible
for
bank
filtration
log
removal
credit.

°
Other
ground
water
collection
devices
such
as
infiltration
galleries
and
spring
boxes
are
ineligible.

°
Systems
using
horizontal
or
vertical
wells
located
at
least
25
feet
from
the
surface
water
source
are
eligible
for
a
0.5
log
removal
credit
and
those
located
at
least
50
feet
from
the
surface
water
source
are
eligible
for
a
1.0
log
removal
credit.

°
Systems
with
vertical
wells
must
identify
the
distance
to
surface
water
using
the
floodway
boundary
or
100
year
flood
elevation
boundary
as
delineated
on
Federal
Emergency
Management
Agency
(
FEMA)
Flood
Insurance
Rate
maps.
If
the
floodway
boundary
or
100
year
flood
elevation
boundary
is
not
already
delineated,
systems
must
determine
the
floodway
or
100
year
flood
elevation
boundary
using
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
4
methods
substantially
similar
to
those
used
in
preparing
FEMA
Flood
Insurance
Rate
maps.

°
Systems
with
horizontal
wells
must
measure
the
distance
from
the
normal
flow
stream
bed
to
the
closest
horizontal
well
lateral.

°
Systems
must
characterize
the
aquifer
at
the
proposed
production
well
site
to
determine
aquifer
properties.

°
At
a
minimum,
the
aquifer
characterization
must
include
the
collection
of
relatively
undisturbed
continuous
core
samples
from
the
surface
to
a
depth
at
least
equal
to
the
projected
bottom
of
the
well
screen
for
the
proposed
production
well.

°
The
recovered
core
length
must
be
at
least
90
percent
of
the
total
depth
to
the
projected
bottom
of
the
well
screen
and
each
sampled
interval
must
be
a
composite
of
no
more
than
2
feet
in
length.

°
Each
composite
sample
must
be
examined
to
determine
if
at
least
10
percent
of
the
grains
in
that
interval
are
less
than
1.0
mm
in
diameter.
Each
composite
sample
with
at
least
10
percent
of
the
grains
less
than
1.0
mm
in
diameter
is
considered
an
interval
with
sufficient
fine­
grained
material
to
provide
adequate
removal.

°
An
aquifer
is
eligible
for
removal
credit
if
at
least
90%
of
the
composited
intervals
contain
sufficient
fine­
grained
material
as
defined
previously.

4.2.2
Monitoring
Requirements
The
LT2ESWTR
requires
systems
to
monitor
turbidity
in
bank
filtration
wells
to
provide
assurance
that
the
assigned
log
removal
credit
is
appropriate.
The
LT2ESWTR
specifically
requires
the
following
monitoring
(
40
CFR
141.726(
c)(
1)):


Turbidity
measurements
must
be
performed
on
representative
water
samples
from
each
wellhead
every
four
hours
that
the
bank
filtration
system
is
in
operation
or
more
frequently
if
required
by
the
state.


Continuous
turbidity
monitoring
at
each
wellhead
may
be
used
if
the
system
validates
the
continuous
measurement
for
accuracy
on
a
regular
basis
using
a
protocol
approved
by
the
state.


If
the
monthly
average
of
daily
maximum
turbidity
values
at
any
well
exceeds
1
NTU,
the
system
must
report
this
finding
to
the
state
within
30
days.
In
addition,
within
30
days
of
the
exceedance
the
system
must
conduct
an
assessment
to
determine
the
cause
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
5
of
the
high
turbidity
levels
and
submit
that
assessment
to
the
state
for
a
determination
of
whether
any
previously
allowed
credit
is
still
appropriate.

4.3
Toolbox
Selection
Considerations
Bank
filtration
is
best
suited
to
systems
that
are
located
adjacent
to
rivers
with
reasonably
good
surface
water
quality
and
that
plan
to
use
bank
filtration
as
one
component
of
their
treatment
process.
For
systems
that
can
meet
the
aquifer
requirements
(
section
4.4)
and
the
design
criteria
(
section
4.5),
bank
filtration
can
be
an
efficient,
cost­
effective
pretreatment
option
to
improve
water
quality
(
Berger,
2002).
Medema
et
al.
(
2000)
and
Wang
et
al
(
2000,
2002)
documented
high
removal
of
Cryptosporidium
indicator
organisms
at
production
well
sites
in
The
Netherlands
and
in
Louisville,
Kentucky.
There
was
very
little
occurrence
of
Cryptosporidium
in
river
water
at
the
Kentucky
site
and
no
Cryptosporidium
was
found
in
the
well
water
at
either
site.
The
amount
of
Cryptosporidium
removal
at
either
site
is
unknown.

The
efficient
removal
of
indicator
organisms
at
the
Netherlands
site
was
likely
due
to
the
relatively
impermeable,
fine­
grained
layer
of
river
sediment
present,
as
well
as
the
effect
of
pyrite
oxidizing
to
ironhydroxides.
Ironhydroxides
may
enhance
the
attachment
of
microorganisms
to
riverbed
sediments
(
Medema
et
al,
2000).
In
Louisville,
Kentucky,
an
alluvial
aquifer
was
chosen
for
the
bank
filtration
site.
Wang
et
al
(
2000,
2002)
found
that
removal
of
particles
increased
with
filtration
distance
of
the
riverbank
filtration
process,
although
most
of
the
removal
occurred
at
the
surface
of
the
riverbed,
within
the
first
two
feet
of
filtration.
Wang
et
al
(
2002)
attributed
the
removal
in
their
bank
filtration
system
to
a
combination
of
mechanical
filtering
and
biological
activity
(
e.
g.,
biofiltering)
at
the
surface
of
the
riverbed.

As
discussed
in
section
4.4,
only
certain
sites
are
suitable
for
bank
filtration.
It
is
important
to
understand
the
type
of
bed
and
aquifer
material
present,
the
dynamics
of
groundwater
flow,
and
the
potential
for
scouring
of
riverbed
materials
at
a
potential
bank
filtration
site.
The
degree
to
which
the
bed
and
banks
of
surface
water
bodies
may
effectively
filter
Cryptosporidium
may
vary
not
only
only
from
site
to
site,
but
also
at
a
single
site
over
time.

4.3.1
Advantages
and
Disadvantages
4.3.1.1
Removal
of
additional
contaminants
The
two
research
sites
with
published
data
(
Medema
et
al.,
2000;
Wang
et
al.,
2000;
Wang
et
al.,
2002;
Berger,
2002)
have
reported
that
bank
filtration
is
effective
at
removing
Cryptosporidium.
Bank
filtration
has
also
been
shown
at
some
sites
to
be
an
effective
technology
for
attenuating
a
variety
of
additional
microorganisms
as
well
as
particulates,
ammonia,
nitrate,
pesticides
(
e.
g.,
atrazine),
heavy
metals,
ethylenediamine
tetra­
acetic
acid
(
EDTA),
alkylated
and
chlorinated
benzenes
and
other
organic
contaminants,
and
disinfection
by­
product
precursors
(
DBPs)
in
the
form
of
natural
organic
matter
(
NOM)
(
Schijven
et
al.,
2003;
Tufenkji
et
al.,
2002;
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
6
Ray
et
al,
2002;
Kuehn
and
Mueller,
2000).
Bank
filtration
achieves
the
removal
of
these
diverse
contaminants
by
facilitating
or
enhancing
physical
and
chemical
filtering,
sorption,
reduction/
oxidation,
precipitation,
ion
exchange,
and
biodegradation
(
Schijven
et
al.,
2003;
Ray
et
al.,
2002;
Tufenkji
et
al.,
2002).
Bank
filtration
further
reduces
contaminant
concentrations
and
especially
shock
contaminant
loads
from
spills
and
intentional
acts
by
providing
for
the
multidimensional
dispersion
and
dilution
of
contaminants
(
Ray
et
al.,
2002).

The
degree
to
which
any
particular
contaminant
will
be
removed
via
bank
filtration
depends
on
site­
specific
conditions.
For
example,
under
aerobic
conditions,
ammonia
is
often
completely
transformed,
whereas
such
removal
may
not
occur
under
more
reducing
conditions.
Oxygen
is
usually
significantly
depleted
within
5­
15
feet
of
the
riverbed,
due
to
microbial
activity
in
this
zone.
As
infiltrating
water
becomes
increasingly
depleted
of
organic
matter
due
to
degradation,
microbial
activity
diminishes,
and
the
aquifer
may
be
reaerated
at
a
certain
distance
from
the
riverbed
(
Tufenkji
et
al.,
2002).
The
anaerobic
part
of
the
aquifer
was
observed
to
remove
up
to
99%
of
polar
organic
contaminants
at
a
site
in
central
Germany
(
Juttner,
1995).
Miettinen
et
al
(
1994)
found
that
almost
90%
of
the
high
molecular
weight
fraction
of
NOM
had
been
removed
at
a
bank
filtration
site
in
Finland.

The
reduction
in
some
treatment
costs
made
possible
by
bank
filtration
results
from
a
reduced
need
for
other
treatment
technologies.
When
bank
filtration
decreases
the
concentration
of
dissolved
organic
carbon
reaching
a
treatment
plant,
costs
are
lowered
because
a
decreased
proportion
of
dissolved
contaminants
needs
to
be
adsorbed
onto
activated
carbon
filters.
Thus,
each
filter
is
capable
of
operating
for
a
longer
period
of
time,
and
fewer
replacement
filters
are
needed.
Particle
and
microorganism
removal
during
bank
filtration
allows
for
more
efficient
filtration,
use
of
membranes,
and
disinfection
during
subsequent
treatment
steps.
The
removal
of
ammonia
means
that
the
additional
treatment
step
of
oxidizing
ammonia
with
chlorine
may
be
unnecessary.
The
removal
of
nitrate
when
water
is
induced
to
flow
through
anaerobic
areas
may
eliminate
the
need
for
expensive
ion
exchange
or
reverse
osmosis
treatment
processes
(
Kuehn
and
Mueller,
2000).
Finally,
because
it
is
effective
at
biodegrading
many
contaminants,
bank
filtration
reduces
the
need
for
adding
large
quantities
of
flocculants
to
drinking
water,
thereby
reducing
both
costs
and
the
unhealthful
effects
of
water
treatment
residuals
(
Kuehn
and
Mueller,
2000).

Another
advantage
of
bank
filtration
as
a
pretreatment
technology
is
that
it
acts
to
equalize
fluctuations
in
contaminant
concentrations
observed
in
surface
waters.
This
is
due
to
the
effects
of
dilution
and
dispersion
which
serve
to
spread
peaks
in
contaminant
concentrations
over
space
and
time
by
the
time
they
reach
wells.
Contaminant
concentration
peaks
may
be
due
to
variations
in
river
water
levels,
seasonal
effects,
and
runoff,
in
addition
to
spills,
terrorist
acts
and
emissions
by
municipal
and
industrial
institutions
(
Kuehn
and
Mueller,
2000).
Bank
filtration
also
smooths
out
fluctuations
in
water
temperature.
Bank
filtration
is
continuously
active,
and
the
decreased
amplitude
of
the
contaminant
peak
by
the
time
it
reaches
a
well
(
an
inherent
result
of
subsurface
transport
through
porous
material)
allows
for
easier
and
less
expensive
treatment
by
utilities
with
limited
capabilities.
In
addition,
the
time
lag
between
contamination
of
surface
water
and
arrival
of
contaminant
at
a
well
would
give
utilities
more
of
an
opportunity
to
respond
to
a
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
7
threat
or
an
accidental
spill.
Kuehn
and
Mueller
(
2000)
estimate
that
in
many
modern
bank
filtration
systems
bank
filtrate
spends
anywhere
from
5
to
15
days
in
the
subsurface
before
reaching
supply
wells.
At
one
site
in
the
Netherlands,
bank
filtrate
was
estimated
to
spend
45­
65
days
in
the
subsurface
before
reaching
the
supply
well
(
Medema,
et
al.,
2000).
Residence
time
depends
on
site­
specific
hydrogeology
as
well
as
bank
filtration
system
design.

The
removal
of
NOM
during
bank
filtration
is
useful
because
NOM
occurrence
can
result
in
the
production
of
harmful
disinfection
byproducts,
as
discussed
above.
In
addition,
moderate
to
high
concentrations
of
NOM
in
drinking
water
can
result
in
unpleasant
taste
and
odor.
Finally,
NOM
removal
via
bank
filtration
can
also
aid
in
the
removal
of
a
large
variety
of
additional
organic
and
inorganic
contaminants.
These
contaminants
are
sometimes
made
more
mobile
in
surface
and
ground
waters
due
to
a
partitioning
process
whereby
they
are
attached
to
NOM,
which
is
relatively
mobile,
and
thereby
carried
along
a
flow
path.
The
removal
of
NOM
and
associated
contaminants
prior
to
above­
ground
treatment
is
likely
to
lessen
the
overall
cost
of
water
treatment
at
a
given
facility.

4.3.1.2
Clogging
of
pores
Clogging
of
the
surface
water
­
ground
water
interface
has
the
potential
to
be
a
problem
with
any
riverbank
filtration
system,
and
results
from
physical,
chemical,
and
biological
processes.
Partial
clogging
during
riverbank
filtration
system
operation
is
likely
to
be
unavoidable
(
Wang
et
al.,
2001),
however
its
effects
are
not
always
deleterious.
The
disadvantage
of
clogging
is
that
it
can
reduce
hydraulic
conductivity
of
the
local
riverbed
and
the
aquifer,
thereby
temporarily
or
permanently
reducing
well
yields.
On
the
other
hand,
a
limited
accumulation
of
fine­
grained
sediments
and
the
accompanying
development
of
a
biologically
active
zone
can
enhance
pathogen
removal.
Indeed,
this
enhanced
removal
is
a
basic
principle
behind
riverbank
filtration
as
a
water
treatment
technology.
An
optimal
amount
of
clogging
is
beneficial
because
it
can
reduce
the
size
of
large
pores
or
reduce
entrances
to
pores
in
a
stream
bed
or
aquifer.
Pore
size
reduction
and
decreased
hydraulic
conductivity
also
result
in
longer
travel
times
which
can
result
in
additional
pathogen
inactivation.
Transport
of
fewer
pathogens
is
also
likely
because
there
are
more
opportunities
for
pathogen
contact
with
aquifer
grain
surfaces.

Physical
clogging
of
the
surface
water
­
ground
water
interface
results
from
the
deposition
of
fine­
grained,
suspended
sediment
at
the
interface
and
in
the
near
surface
pores.
The
deposition
and
growth
of
microorganisms
also
contribute
to
physical
clogging.
This
clogging
may
be
exacerbated
during
periods
of
low
surface
water
discharge,
and
is
most
apparent
near
the
river's
edge
where
flow
velocities
are
generally
lower
than
at
the
center
of
the
river.
Chemical
clogging
can
result
from
precipitation
of
dissolved
surface
water
constituents
and
may
occur
near
the
interface
or
anywhere
along
the
flowpath.
This
is
due
to
the
change
in
geochemical
conditions
as
infiltrating
water
enters
the
riverbed
and
aquifer.
Factors
to
be
considered
when
evaluating
the
potential
for
chemical
clogging
include
electrolyte
concentration,
pH,
redox
potential,
presence
of
dissolved
or
colloidal
organic
matter,
and
the
mineralogy
and
surface
characteristics
of
stream
bed
and
aquifer
solids.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
8
Finally,
biological
or
microbial
clogging
can
result
from
the
accumulation
of
bacterial
cells
in
pore
spaces,
the
production
of
extra­
cellular
polymers,
the
release
of
gaseous
byproducts
from
denitrifying
bacteria
and
methanogens,
and
the
microbially
mediated
accumulation
of
insoluble
precipitates
(
Vandevivere
et
al.,
1995;
Baveye
et
al.,
1998).
Biogenic
gas
bubbles
have
the
effect
of
blocking
or
partially
blocking
water
flow
through
pores
in
much
the
same
way
that
solid
particles
do
(
Orlob
and
Radhakrishna
1958;
Oberdorfer
and
Peterson
1985;
Sanchez
de
Lozada
et
al.,
1994).
Insoluble
sulfide
salts
can
cause
clogging
due
to
the
activity
of
sulfate
reducing
bacteria,
whereas
iron
hydroxide
and
manganese
oxide
deposition
can
be
brought
on
by
bacterial
iron
metabolism
(
Vandevivere
et
al.,
1995;
Baveye
et
al.,
1998).
Biological
clogging
is
most
likely
to
occur
near
the
surface
water
­
ground
water
interface
where
nutrients
are
most
available.

Some
or
all
of
these
processes
may
act
at
a
particular
site
to
lower
hydraulic
conductivity
and
thus
decrease
flow
velocities.
For
example,
several
months
of
pumping
from
a
new
riverbank
filtration
well
in
Louisville,
Kentucky
resulted
in
a
significant
decline
in
well
production,
presumably
due
to
a
70%
reduction
in
leakance
from
the
river
to
the
adjacent
aquifer.
The
reduced
well
yields
were
attributed
to
the
physical
clogging
of
riverbed
sediments
(
Schafer,
2000).
The
disadvantage
of
reduced
well
yields
accompanies
the
advantages
of
increased
microbial
inactivation
rates
due
to
lower
flow
velocities
(
and
thus
longer
residence
times
in
the
aquifer)
as
well
as
increased
removal
of
pathogens
due
to
smaller
pores.

4.3.1.3
Scour
Both
the
positive
and
negative
effects
of
clogging
on
riverbank
filtration
system
performance
may
be
diminished
following
periodic
flooding.
Scour
refers
to
the
erosion
of
the
river's
bed
and
banks,
and
depends
on
both
flood
conditions
and
the
resistance
of
the
bed
and
bank
material
that
has
been
deposited
at
a
particular
site.
During
flooding
the
river
channel
may
be
scoured,
and
fine
sediments
at
the
surface
water
­
ground
water
interface
mobilized.

Much
of
the
removal
of
the
contaminants
and
microbes
discussed
above
occurs
during
the
first
few
centimeters
of
the
flow
path,
due
to
the
significant
filtering
and
sorptive
capabilities
of
sediments
in
the
riverbed.
These
sediments
are
often
organic­
rich,
highly
biologically
active,
and
fine­
grained.
The
effectiveness
of
bank
filtration,
however,
may
be
temporarily
threatened
during
high
flows
if
this
active
layer
is
washed
away
or
scoured.
EPA
suggests
the
potential
for
stream
channel
scour
be
evaluated
during
riverbank
filtration
site
selection
(
section
4.4).
Section
4.5
provides
further
discussion
of
scour
and
its
implications
for
riverbank
filtration
system
operation.

4.3.1.4
Additional
Treatment
Steps
In
addition
to
clogging
and
scour,
there
are
several
disadvantages
to
bank
filtration
which
utilities
may
wish
to
consider
and
balance
against
the
advantages
and
cost
savings
described
in
section
4.3.1.
One
disadvantage
is
that
an
additional
aeration
step
may
be
required
during
water
treatment
due
to
the
possible
depletion
of
oxygen
as
biological
activity
consumes
oxygen
during
riverbank
filtration
pretreatment
(
Kuehn,
et
al.,
2000).
This
oxygen
depletion
may
lead
to
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
9
extremely
anaerobic
conditions
over
a
portion
of
the
flow
path,
which
may
sometimes
result
in
the
release
of
iron
and
manganese
from
the
bank
sediment
into
the
flowing
water.
This
process
occurs
due
to
a
redox
reaction
which
reduces
iron
and
manganese
to
their
water­
soluble
forms.
This
condition
may
necessitate
the
removal
of
these
metals
during
subsequent
treatment
steps
(
Kuehn,
et
al.,
2000;
Tufenkji
et
al.,
2002).

On
the
other
hand,
if
the
flow
path
between
the
riverbank
and
the
well
is
long
enough,
iron
and
manganese
may
precipitate
onto
the
sediments
in
the
subsurface
before
ever
reaching
the
well
(
Tufenkji
et
al.,
2002).
The
aquifer
becomes
reaerated
with
increasing
distance
from
the
riverbed.
This
is
one
reason
for
locating
riverbank
filtration
wells
greater
than
25
or
50
feet
from
the
river,
as
discussed
in
section
4.5.2.2.
Even
though
most
contaminant
removal
occurs
during
the
first
few
centimeters
of
subsurface
transport,
the
reaeration
and
associated
precipitation
reactions
in
the
aquifer
may
significantly
improve
water
quality
before
it
reaches
the
well
(
Tufenkji
et
al.,
2002).
The
location
of
the
aerated
and
anaerobic
portions
of
the
aquifer
vary
seasonally
due
to
variable
microbial
activity
and
changing
pumping
rates.

Finally,
riverbank
filtration
is
ineffective
at
removing
a
few
persistent
compounds,
primarily
non­
polar
organic
compounds
and
highly
soluble
chemical
contaminants
such
as
methyltertiarybutylether
(
MTBE)
and
trichloroethylene
(
TCE),
which
would
need
to
be
addressed
during
subsequent
treatment
steps.
In
addition,
when
bank
filtration
is
used
to
induce
infiltration
of
highly
contaminated
surface
water,
it
may
be
important
to
include
additional
adsorption
steps
during
later
treatment
(
Kuehn,
et
al.,
2000).

4.4
Site
Selection
and
Aquifer
Requirements
Unconsolidated,
granular
aquifers
with
sufficient
amounts
of
fine­
grained
material
(
see
section
4.4.2)
are
eligible
for
Cryptosporidium
removal
credits
under
the
LT2ESWTR.
Partially
consolidated,
granular
aquifers
may
also
be
eligible
for
removal
credits.
Each
granular
aquifer
proposed
as
a
bank
filtration
site
is
to
be
evaluated
on
a
case­
by­
case
basis
with
regard
to
its
grain
size
distribution
and
degree
of
cementation.
For
example,
a
partially
consolidated,
granular
aquifer
may
be
too
cemented,
and
thus
perhaps
too
fractured,
to
provide
adequate
pathogen
removal.
Geophysical
methods,
discussed
in
section
4.5.2.2,
may
be
helpful
in
determining
the
degree
of
fracturing
of
such
aquifers.

This
section
characterizes
river
and
aquifer
types
that
may
be
suitable
for
bank
filtration
surface
water
treatment.
A
list
of
selected
sites
in
the
United
States
and
Europe
which
have
used
bank
filtration
is
provided
for
reference.
No
information
is
available
for
these
sites,
however,
regarding
whether
they
would
meet
the
siting
criteria
in
the
LT2ESWTR.
Some
common
aquifer
types
that
are
clearly
not
appropriate
for
this
technology
are
described
as
well.
Finally,
sitespecific
aquifer
criteria
which
shall
be
met
in
order
for
systems
to
receive
Cryptosporidium
removal
credits
are
outlined
in
section
4.4.3.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
10
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
11
4.4.1
Selected
Bank
Filtration
Sites
Table
4.1
Selected
Bank
Filtration
Systems
in
Europe
and
the
United
States
Site
Location
Well
Type*
Number
of
Wells
Maximum
Capacity
mgd
(
m3/
s)
River
System
Europe
Torgau,
Germany
V
42
39.7
(
1.737)
Elbe
Mockritz,
Germany
V
74
28.8
(
1.260)
Elbe
United
States
Cincinnati,
OH
V
10
40.0
(
1.750)
Great
Miami
Columbus,
OH
H
4
40.0
(
1.750)
Scioto/
Big
Walnut
Louisville,
KY
H
1+
20.0
(
0.875)
Ohio
Terra
Haute,
IN
H
1
12.0
(
0.525)
Wabash
Jacksonville,
IL
H
1
8.0
(
0.350)
Illinois
Galesburg,
IL
H
1
10.0
(
0.438)
Mississippi
Henry,
IL
V
1
0.7
(
0.030)
Illinois
Mt.
Carmel,
IL
V
1
1.0
(
0.044)
Wabash
Quincy,
IL
H
1+
10.0
(
0.438)
Mississippi
Sacramento,
CA
H
1
10.0
(
0.438)
Sacramento
Sonoma
County,
CA
H,
V
5
(
H)
+
7
(
V)
85.0
(
3.727)
Russian
Independence,
MO
H
*
1
15.0
(
0.656)
Missouri
Lincoln,
NB
H,
V
2
(
H)
+
44
(
V)
35.0
(
H)
(
1.530)
Platte
Kennewick,
WA
H
1
3.0
(
0.130)
Columbia
Kalama,
WA
H
1
2.6
(
0.110)
Kalama
St.
Helens,
OR
H
3
5.0
(
0.219)
Columbia
Kansas
City,
KS
H
1
40.0
(
1.750)
Missouri
Sioux
Falls,
OK
H
1+
40.0
(
1.750)
Missouri
*
H
 
horizontal,
V
 
vertical
*
Gravel­
packed
Laterals
Reprinted
from
Journal
AWWA,
Vol.
94,
No.
4
(
April
2002),
by
permission.
Copyright
©
2002,
American
Water
Works
Association.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
12
4.4.2
Aquifer
Type
4.4.2.1
Unconsolidated,
Granular
Aquifers
Unconsolidated,
granular
aquifers
can
be
composed
of
a
wide
range
of
sediment
sizes
including
clay,
silt,
sand,
and
larger
particles.
They
may
also
exhibit
minor
cementation,
but
subsurface
samples
are
typically
friable
(
readily
crumbled
by
hand).
To
be
eligible
for
bank
filtration
credits
under
the
LT2ESWTR,
unconsolidated
granular
aquifers
are
expected
to
contain
a
sufficient
amount
of
fine­
grained
sediments
to
achieve
adequate
pathogen
removal
and/
or
inactivation
(
section
4.4.3
prescribes
the
amount
deemed
sufficient).
In
aquifers
with
these
characteristics,
the
flow
path
is
tortuous
at
the
micro­
scale
(
Figure
4.3),
providing
many
opportunities
for
removal
of
microorganisms
by
straining
or
by
their
attachment
to
grain
surfaces.

Many
alluvial
aquifers
contain
significant
amounts
of
well­
sorted,
fine­
grained
sediments.
Alluvial
aquifers
are
produced
by
fluvial
depositional
processes
and
are
adjacent
to
modern
streams.
Aquifers
formed
in
glacial
deposits
may
also
contain
sufficient
amounts
of
fine­
grained
material.
These
may
be
"
till"
deposits,
which
have
a
wide
range
of
poorly
sorted
sediment
sizes,
or
glacial
outwash
deposits
that
are
formed
by
meltwater
and
often
contain
well­
sorted,
sandsized
sediments.
Any
of
these
alluvial
or
till
aquifers
would
be
likely
to
be
suitable
for
a
bank
filtration
system.
On
the
other
hand,
coarse
gravel
aquifers
produced
by
the
rapid
drainage
of
glacial
lakes,
or
in
outwash
environments
that
deposit
little
fine­
grained
material,
may
not
be
eligible
for
bank
filtration
credits
unless
sieve
analysis
shows
sufficient
fine­
grained
material
as
discussed
in
section
4.4.3.2.

Alluvial
aquifers
may
be
identified
on
detailed
hydrogeologic
maps
simply
as
"
Quaternary
alluvium",
indicating
both
their
genesis
and
relative
age.
Glacial
deposits
are
documented
on
surficial
geology
maps
and,
where
aquifer­
forming,
may
be
identified
on
large­
scale
hydrogeologic
maps.

4.4.2.2
Karst,
Consolidated
Clastic,
and
Fractured
Bedrock
Aquifers
In
karst,
consolidated
clastic,
and
fractured
bedrock
aquifers,
ground
water
velocities
are
fast,
and
flow
paths
may
be
direct,
allowing
microbial
contaminants
to
travel
rapidly
to
a
well
with
little
removal
or
inactivation.
Therefore,
these
aquifer
types
are
not
eligible
for
bank
filtration
treatment
credits.

Karst
may
be
broadly
defined
as
a
region
where
the
dissolution
of
calcitic
or
other
soluble
bedrock,
primarily
limestone
(
calcium
carbonate),
produces
a
unique
subsurface
drainage
network
and
associated
surface
landforms.
Ground
water
movement
in
karst
aquifers
differs
from
that
in
porous,
granular
aquifers
in
that
flow
in
the
former
occurs
predominantly
in
conduits
and
dissolution­
enlarged
fractures.
Consequently,
there
is
little
physical
removal
of
microbes
and
other
particles
by
filtration
and
few
opportunities
for
microbes
to
come
in
contact
with
the
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
13
surfaces
of
aquifer
materials.
Furthermore,
rapid
flow
creates
conditions
where
inactivation
is
less
likely
to
occur
before
ground
water
reaches
a
well.

Although
fractures
have
a
role
in
ground
water
movement
through
any
aquifer,
fractures
provide
the
dominant
flow­
path
in
fractured
consolidated
clastic
and
fractured
bedrock
aquifers.
Most
consolidated
aquifers
can
be
presumed
to
be
fractured.
Similar
to
solution
conduits
in
karst
aquifers,
fractures
in
consolidated
aquifers
provide
preferential
flow
paths
that
may
transmit
ground
water
at
high
velocities,
and
in
a
relatively
direct
flow
path
to
a
well,
with
little
time
or
opportunity
for
inactivation
or
removal
of
microbial
pathogens.
Wells
located
in
these
aquifers
would
not
be
eligible
for
bank
filtration
credit.

4.4.2.3
Partially
Consolidated,
Granular
Aquifers
Granular
aquifers
formed
by
marine
processes
earlier
than
Quaternary
alluvial
or
glacial
deposition
may
be
partially
consolidated
by
natural
cement
that
fills
pores,
connects
grains,
and
makes
the
aquifer
material
less
friable.
Partially
consolidated,
granular
aquifers
are
present
within
the
Atlantic
Coastal
Plain,
Gulf
Coast
Lowland,
Texas
Coastal
Upland,
and
Mississippi
Embayment
aquifer
systems
(
USGS
1998).
When
significant
proportions
of
cement
are
present,
fractures
are
more
likely
to
exist.
As
in
consolidated
aquifers,
fractures
in
partially
consolidated,
granular
aquifers
create
direct
paths
for
microbial
contamination
that
minimize
the
natural
filtration
capabilities
of
the
aquifer
system.
EPA
suggests
that
partially
consolidated
aquifers
be
evaluated
at
the
proposed
well
location
to
determine
if
they
may
be
too
cemented,
and
thus
perhaps
too
fractured,
to
provide
sufficient
natural
filtration.

The
degree
of
cementation
can
be
evaluated
by
a
variety
of
methods.
Geologic
material
collected
from
below
the
aquifer's
weathered
zone
that
is
friable
upon
touch
is
likely
to
be
adequate
for
bank
filtration
purposes.
Another
test
for
the
degree
of
cementation
includes
the
slaking
test,
which
involves
alternate
wetting
and
drying
of
the
sample
in
water,
or
in
salt
or
alcohol
solutions.
Finally,
a
triaxial
compression
test
can
be
used
to
measure
strain
in
three
mutually
perpendicular
directions.
Less
cemented
samples
will
be
more
deformable
during
such
tests.

4.4.3
Aquifer
Characterization
Systems
seeking
Cryptosporidium
removal
credit
are
required
to
characterize
the
aquifer
properties
between
their
surface
water
source
and
their
well.
The
aquifer
characterization
will
include,
at
a
minimum,
core
sampling
to
determine
grain
size
distribution.
This
data
will
establish
whether
enough
fine­
grained
sediment
is
present
to
provide
adequate
filtration.
The
following
procedure
outlines
the
steps
necessary
to
perform
such
a
characterization,
which
will
ultimately
determine
eligibility
for
bank
filtration
treatment
credits
under
the
LT2ESWTR.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
14
16)
Collect
relatively
undisturbed
continuous
core
samples
from
the
surface
to
a
depth
at
least
equal
to
the
projected
bottom
of
the
well
screen
for
the
proposed
production
well.

17)
Determine
if
recovered
core
consists
of
at
least
90%
of
the
interval
from
the
surface
to
the
planned
location
of
the
well
screen
bottom.
If
core
recovery
is
insufficient,
another
well
core
must
be
obtained.

18)
Examine
each
2
foot
long
composite
sample
of
recovered
core
in
a
laboratory
using
sieve
analysis
to
determine
grain
size
distribution.
Core
intervals
are
typically
2
feet
long
for
a
conventional
split­
spoon
sampler
and
3
or
4
feet
long
for
soil
probes
(
e.
g.,
a
Giddings­
type
soil
probe).

19)
If
more
than
10
percent
of
the
sediments
in
each
2
foot
long
composite
sample
are
less
than
1.0
mm
in
diameter
(
very
coarse
sand),
then
the
core
interval
from
which
it
was
taken
is
noted
as
containing
a
sufficient
quantity
of
fine­
grained
material
to
provide
adequate
pathogen
removal.

20)
To
receive
Cryptosporidium
removal
credit,
at
least
90
percent
of
the
analyzed
composited
core
intervals
from
the
sampled
aquifer
will
meet
criterion
number
(
4)
above.

4.4.3.1
Coring
The
collection
of
relatively
undisturbed
cores
in
unconsolidated
aquifers
can
be
quite
difficult,
especially
when
gravel­
sized
clasts
are
present.
The
two
most
important
criteria
for
successful
test
drilling
to
obtain
a
core
are
sample
accuracy
and
drilling
speed.
Borehole
stability
is
a
major
problem
in
drilling
in
an
unconsolidated
gravelly
formation.
Rotary
core
drilling
is
particularly
suited
to
drilling
in
unconsolidated
formations
because
the
drilling
fluid,
which
cools
the
drill
bit
and
carries
up
the
core,
also
acts
to
stabilize
the
borehole
(
Driscoll,
1986).

Other
drilling
methods
require
the
installation
of
a
casing
to
stabilize
the
borehole,
a
process
which
slows
down
the
speed
of
drilling.
Rotary
core
drilling
is
the
fastest
method
for
drilling
in
an
unconsolidated
formation.
One
disadvantage
to
rotary
core
drilling
is
the
separation
of
different
sized
core
particles
as
they
rise
(
smaller
particles
rise
faster)
and
cross­
contamination
by
overlying
borehole
material.
An
experienced
driller
can
avoid
cross
contamination
by
using
the
dual­
wall
method
of
rotary
core
drilling.
In
the
dual­
wall
method,
the
core
is
pushed
up
the
inner
pipe
of
the
drill
rather
than
traveling
in
the
space
between
the
drill
and
the
borehole
wall
(
Driscoll,
1986).
Shallow
wells
will
have
fewer
particle
size
separation
problems
than
deeper
wells.

Auger
drilling
is
another
method
for
drilling
test
wells.
In
this
method
an
earth
auger
is
screwed
into
the
earth
by
rotation.
Auger
drilling
in
an
unconsolidated
formation
is
slower
than
rotary
core
drilling,
due
to
the
necessary
installation
of
casing
to
support
the
borehole.
Sampling
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
15
with
augers
can
provide
reliable
samples
from
any
depth.
A
split
spoon
sampler
can
be
used
wherein
a
split
spoon
is
driven
to
the
bottom
of
the
hole.
The
depth
to
which
an
auger
can
drill
is
dependent
on
the
size
of
the
rig.
The
maximum
drilling
depth
possible
for
a
small
drill
rig
is
approximately
250
ft.
(
Driscoll,
1986).
Information
about
drilling
and
finding
a
driller
can
be
found
through
the
National
Groundwater
Association
(
NGWA)
website:
http://
www.
ngwa.
org/.
In
addition,
the
EnviroDirectory
 
provides
listings
for
laboratories
and
drillers
in
New
England,
the
Mid­
Atlantic,
and
the
Great
Lakes
regions
(
www.
envirodirectory.
com).

4.4.3.2
Sieve
Analysis
The
American
Society
for
Testing
and
Materials
(
ASTM)
has
a
published
standard
for
conducting
sieve
analysis,
the
Standard
Test
Method
for
Sieve
Analysis
of
Fine
and
Coarse
Aggregates:
Standard
C
136­
1
(
ASTM,
2003).

Sieve
analysis
is
used
to
determine
the
particle
size
distribution
of
a
sample
of
dry
aggregate
of
known
mass
by
passing
the
sample
through
a
series
of
sieves
with
progressively
smaller
openings.
Sieve
analysis
requires
the
following
equipment:

1.
A
balance,
accurate
to
0.1g
or
0.1%
of
test
load
for
fine
aggregate,
or
accurate
to
0.5g
or
0.1%
of
test
load
for
a
mixture
of
fine
and
coarse
aggregate
2.
Stackable
sieves
3.
A
mechanical
sieve
shaker
(
for
sample
sizes
greater
than
20kg)
4.
An
oven
capable
of
maintaining
110
±
5oC
(
230
±
9oF)

In
the
first
step
of
sieve
analysis
the
sample
is
dried
using
the
oven.
Once
dry,
its
weight
is
measured
and
recorded.
While
the
sample
dries,
sieves
are
selected
with
suitable
openings
to
furnish
the
information
required.
For
bank
filtration
related
sieve
analyses,
it
is
only
necessary
to
determine
what
percentage
of
the
sample
is
less
than
1.0mm;
however,
it
is
recommended
that
sieves
covering
a
range
of
sizes
be
used
so
as
to
prevent
the
overloading
of
any
one
sieve.
Once
the
sample
is
dry
and
the
sieves
are
stacked
in
order
of
decreasing
mesh
size,
the
sample
is
placed
in
the
top
sieve
and
sieving
either
by
machine
or
hand
begins.
Sieving
should
be
continued
until
no
more
than
1%
by
mass
of
the
material
retained
on
an
individual
sieve
will
pass
through
that
sieve
during
1
minute
of
continuous
hand
sieving.
Finally
the
mass
on
each
sieve
is
weighed.
The
total
mass
of
the
material
after
sieving
should
correspond
closely
with
the
original
mass
of
the
sample.
Using
the
mass
for
each
size
increment
and
the
total
mass
of
the
sample,
the
size
distribution
of
the
sample
can
be
determined
(
ASTM,
2003).

Further
information
about
sieve
analysis
can
be
found
at
the
ASTM
web
site
(
www.
astm.
org).
A
multi­
media
sieve
analysis
demonstration
can
be
found
at
Geotechnical,
Rock
and
Water
Resources
Library
(
GROW)
(
http://
www.
grow.
arizona.
edu/
geotechnical/
virtual_
labs/
sieveanalysis/
sieveanalysisexp.
shtml).
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
16
ASTM
also
provides
a
search
engine
which
allows
the
user
to
search
for
laboratories
that
perform
sieve
analyses
(
http://
astm.
365media.
com/
astm/
labs/).
The
EnviroDirectory
 
provides
listings
for
laboratories
and
drillers
in
New
England,
the
Mid­
Atlantic,
and
the
Great
Lakes
regions
(
www.
envirodirectory.
com).

4.4.4
Site
Selection
as
it
Relates
to
Scour
Stream
channel
scour
may
often
be
an
important
consideration
in
choosing
sites
that
are
suitable
for
riverbank
filtration.
This
section
discusses
stream
channel
erosional
processes
in
general,
as
well
as
reasons
sites
with
certain
characteristics
may
be
unsuitable
for
riverbank
filtration.
Section
4.6
discusses
the
implications
of
periodic
scour
for
riverbank
filtration
system
operations.
Detailed
information
on
fluvial
erosional
processes
can
be
obtained
from
any
of
a
number
of
texts
on
fluvial
geomorphology,
hydrology,
and
river
hydraulics
(
e.
g.,
Leopold
et
al.,
1964;
Ritter
et
al.,
1995;
Chow
1964).

4.4.4.1
Stream
Channel
Erosional
Processes
This
discussion
focuses
on
the
dominant
erosional
processes
of
alluvial
rivers
because,
given
the
LT2ESWTR's
aquifer
requirements,
such
rivers
may
be
among
the
most
suitable
for
bank
filtration
credits.
Although
many
lake
banks
are
also
suitable
sites
for
bank
filtration,
lakes
will
not
be
discussed
in
detail
in
this
section.
Lake
bank
filtration
settings
typically
do
not
change
rapidly
with
time
and
climate.
Their
hydrologic
properties
are
not
highly
variable
and
thus
do
not
require
the
detailed
evaluation
discussed
here
for
riverbank
filtration
settings.

The
width,
depth,
and
gradient
of
an
undisturbed
alluvial
river
has
typically
adjusted
to
prevailing
discharge
conditions
and
sediment
loads
such
that
no
net
erosion
or
deposition
occurs
over
long
time
periods
(
Mackin
1948;
Leopold
and
Maddock
1953).
The
quasi­
equilibrium
condition
of
such
rivers,
which
are
referred
to
as
"
graded
streams"
(
Mackin1948),
may
be
disrupted
over
short
time
periods
(
e.
g.,
due
to
floods),
when
erosion
or
deposition
may
be
considerable.

The
dominant
scouring
process
in
alluvial
rivers
is
lateral
migration.
This
process
is
responsible
for
the
stream
meanders
visible
on
many
floodplains,
and
is
accomplished
by
the
progressive
erosion
of
the
outside
bank
of
a
river
bend
with
concurrent
deposition
on
the
inside
bank.
Because
erosion
is
generally
matched
by
deposition
in
this
process,
channel
dimensions
do
not
change
significantly
over
time,
and
the
net
result
is
migration
of
the
channel
across
the
floodplain
(
Figure
4.1).
Stream
channel
meanders
are
characteristic
of
many
alluvial
rivers
and
are
indicative
of
a
graded
stream.

Downcutting,
another
type
of
scour
that
can
occur
in
fluvial
environments,
is
the
vertical
erosion
of
the
streambed.
Downcutting
is
fairly
uncommon
in
alluvial
rivers
except
during
floods
or
if
the
stream
is
not
graded.
The
long­
term
dynamic
equilibrium
of
a
graded
stream
can
be
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
17
disrupted
by
a
variety
of
changing
hydrologic
and
geologic
conditions
and
especially
by
anthropogenic
activity.
Human
activities
in
a
watershed
or
river
channel
may
alter
the
conditions
to
which
an
alluvial
river
has
become
adjusted,
initiating
a
period
of
readjustment
marked
by
either
progressive
downcutting
or
aggradation
(
deposition).

Urbanization
generally
increases
the
proportion
of
impervious
surface
in
a
watershed,
increasing
flood
volumes
during
precipitation
events
because
less
water
is
able
to
infiltrate
the
land
surface
and
recharge
ground
water
(
Jacobsen
et
al.,
2001).
Increased
flood
volumes
may
cause
higher
water
levels
in
a
river
channel,
increasing
the
shear
stress
on
the
channel
bed
and
causing
scour
(
Booth,
1990).
Downcutting
may
continue
until
the
channel
gradient,
and/
or
channel
dimensions,
become
adjusted
to
the
new
flooding
regime.

Impoundment
is
another
activity
that
may
disrupt
the
quasi­
equilibrium
state
of
a
graded
river
and
initiate
readjustment
of
the
channel.
The
sharp
decrease
in
sediment
supply,
which
commonly
occurs
subsequent
to
dam
and
reservoir
construction,
may
initiate
downcutting
in
the
reach
immediately
downstream
until
the
channel
adjusts
to
the
lightened
sediment
load.
This
has
been
observed
downstream
of
many
dams
throughout
the
world.
One
of
the
most
dramatic
examples
is
the
7.5
meters
of
channel­
bed
degradation
that
occurred
twelve
kilometers
downstream
of
the
Hoover
Dam
after
its
completion
in
1935
(
Williams
and
Wolman,
1984).

The
construction
of
artificial
levees
(
raised
banks
along
a
stream
channel)
also
may
result
in
flooding
downstream.
Levees
allow
greater
quantities
of
water
to
be
carried
by
the
stream,
thus
decreasing
the
probability
of
flooding
in
the
vicinity
of
the
levee,
but
increasing
flood
hazards
downstream
(
Montgomery,
2000).
Even
if
flooding
downstream
does
not
result,
the
high
flows
downstream
may
cause
downcutting
of
the
river,
removal
of
fine­
grained
bed
material,
and
thus
a
threat
to
the
protectiveness
of
a
riverbank
filtration
system.
Another
possible
effect
of
levees
is
an
increase
in
sedimentation
in
the
channel.
Sediment
that
would
otherwise
be
deposited
on
the
floodplain
may
be
trapped
within
the
channel.
This
can
raise
the
channel
bottom
and
thus
raise
stream
stage
or
the
elevation
of
the
water
surface
in
the
channel
(
Montgomery,
2000).
The
consequences
of
this
for
a
riverbank
filtration
system
are
variable.
Increased
sedimentation
may
lead
to
clogging
and/
or
decreased
well
yields.
On
the
other
hand,
higher
stream
stages
may
result
in
flooding
and
scour
along
certain
portions
of
the
river
as
the
channel
adjusts
to
a
new
equilibrium
condition.
Understanding
the
impact
of
current
or
planned
upstream
activities
can
be
an
important
part
of
site
selection
for
a
riverbank
filtration
system.

4.4.4.2
Unsuitable
Sites
As
discussed
in
section
4.4.2.2,
some
sites
may
be
ineligible
for
bank
filtration
credit
due
to
the
type
of
aquifer
adjacent
to
the
river.
The
following
section,
however,
focuses
on
importance
of
understanding
the
nature
of
the
surface
water­
ground
water
interface
at
a
potential
bank
filtration
site.
In
some
localities,
frequent
scour,
the
absence
of
a
sufficiently
fine­
grained
interface,
and/
or
the
coarse­
grained
or
fractured
aquifer
materials
may
suggest
that
a
certain
river
or
reach
is
not
the
best
possible
location
for
a
bank
filtration
system.
A
system
may
choose
to
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
18
evaluate
such
situations
on
a
site­
by­
site
basis,
however,
except
as
specified
in
the
LT2ESWTR,
EPA
does
not
require
such
evaluations
or
any
particular
decisions
made
on
the
basis
of
such
evaluations.
EPA
recommends,
however,
that
this
information
be
considered
in
order
to
ensure
that
bank
filtration
systems
are
protective
of
public
health.

Data
from
studies
of
aerobic
and
anaerobic
spore
(
indicator
organism)
removal
in
bank
filtration
systems
indicate
that
much
of
the
removal
of
spores
between
a
surface
water
body
and
a
bank
filtration
pumping
well
takes
place
at
the
surface
water
 
ground
water
interface
and
in
the
aquifer
material
proximal
to
that
interface.
The
interface,
which
lines
the
bottom
of
the
riverbed,
is
typically
comprised
of
very
fine­
grained,
biologically­
active
material
a
few
inches
to
a
foot
in
thickness.
Some
rivers,
however,
(
especially
many
in
the
western
United
States)
lack
this
finegrained
bed
and
the
important
organic­
rich
materials
or
silts
usually
associated
with
the
surface
water
­
ground
water
interface
(
Ray
et
al.,
2002).
For
this
reason,
such
rivers
should
be
evaluated
carefully
to
determine
if
bank
filtration
is
a
suitable
treatment
technology.

Other
rivers
may
sometimes
possess
this
important
layer
of
fine­
grained
sediment
for
the
first
few
inches
or
feet
from
the
surface
water
source,
but
may
at
other
times
be
subject
to
periodic
scour.
The
nature
and
performance
of
the
surface
water
 
ground
water
interface
on
such
rivers
may
be
altered
temporarily
by
flood
scour,
specifically
during
the
high
river
stages
that
occur
periodically
throughout
the
year.
This
situation
is
most
likely
to
occur
on
uncontrolled
rivers.
Higher
flow
velocities
in
the
river
and
increased
bedload
transport
at
such
times
mobilizes
fine
sediments
(
which
were
deposited
when
discharges
were
lower).
(
Note
that
bedload
transport
is
the
carrying
of
heavy,
coarse
sediments
by
saltation
along
the
stream
bed
rather
than
by
suspension.
Saltation
is
the
process
in
which
particles
jump
from
one
point
to
the
next
along
the
stream
bed.)
Lower
log
removals
are
thus
expected
to
occur
during
floods.
If
such
situations
are
expected
to
occur
very
frequently,
and
if
a
system
cannot
envision
a
way
of
managing
the
system
so
as
to
adequately
protect
its
water
supply
during
such
events,
sites
on
such
rivers
may
be
inappropriate
for
riverbank
filtration.
On
the
other
hand,
through
careful
management
it
may
be
feasible
to
protect
drinking
water
wells
from
the
potentially
negative
consequences
of
occasional
scour,
as
discussed
in
section
4.6.2.

EPA
recommends
that
the
potential
for
scouring
be
considered
during
site
selection.
If
a
site
that
undergoes
occasional
scour
is
selected
for
riverbank
filtration,
the
system
may
wish
to
locate
its
wells
at
greater
than
the
required
separation
distance
from
the
surface
water
body,
as
discussed
in
section
4.5.2.2.
Such
a
solution
helps
to
ensure
the
protection
of
public
health.
The
drawback
of
this
solution
to
the
problem
of
scour,
however,
is
that
wells
located
at
very
great
distances
from
surface
water
sources
are
drawing
in
more
ambient
ground
water
and
less
riverbank
filtrate
than
wells
located
closer
to
the
river
or
lake.
One
result
of
this
is
that
the
yield
to
the
wells
is
likely
to
be
smaller
when
the
wells
are
located
far
from
the
surface
water
source.

The
potential
for
scour
can
be
evaluated
initially
by
examining
the
past
frequency
of
high
flow
and
flood
events.
Data
on
flood
history
and
discharge
is
typically
available
from
the
US
Geological
Survey,
the
Army
Corps
of
Engineers,
the
US
Bureau
of
Reclamation
and
the
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
19
Department
of
Homeland
Security
(
formerly
FEMA).
State
and
county
highway
and
transportation
departments
typically
evaluate
river
scour
to
determine
the
safety
of
bridge
supports.
A
more
comprehensive
evaluation
of
the
potential
for
scour
can
be
conducted
when
the
effect
of
past
and
current
human
activities
(
as
discussed
in
section
4.4.4.1)
is
considered
in
comparison
to
the
history
of
flood
events.

Sources
of
high
flow
and
flood
data
USGS
Main
Page:
http://
water.
usgs.
gov
The
National
Flood
Frequency
(
NFF)
Program:
http://
water.
usgs.
gov/
pubs/
wri/
wri024168/
pdf/
entirereport.
pdf
°
A
computer
program
developed
by
the
USGS
for
estimating
the
magnitude
and
frequency
of
floods
for
ungaged
sites.
Since
1993,
updated
equations
have
been
developed
by
the
USGS
for
various
areas
of
the
nation.
These
new
equations
have
been
incorporated
into
an
updated
version
of
the
NFF
Program.
USGS
Fact
Sheets
(
listed
by
state):
http://
water.
usgs.
gov/
wid/
index­
state.
html
°
Includes
NFF
program
methods
for
estimating
flood
magnitude
and
frequency
(
in
rural
and
urban
areas)
for:
AL,
AZ,
AR,
CA,
CN,
HI,
LA,
MD,
MO,
NV,
NM,
NC,
OK,
SC,
SD,
TX
UT,
VT,
VA,
and
WA.
These
fact
sheets
describe
the
application
of
the
updated
NFF
Program
to
various
waterways
within
the
specific
State.
Includes
maps
of
each
of
the
above
state's
hydrologic
regions,
as
well
as
regression
equations
and
statistics.
WaterWatch:
http://
water.
usgs.
gov/
cgi­
bin/
dailyMainW?
state=
us&
map_
type=
flood&
web_
type=
map
°
Map
of
current
flood
and
high
flow
conditions
in
the
United
States.
The
map
shows
the
location
of
streamgages
where
the
water
level
is
currently
at
or
above
flood
stage
(

)
or
at
high
flow
(

)
.
The
high
flow
conditions
are
expressed
as
percentiles
that
compare
the
current
(
i.
e.,
within
the
past
several
hours)
flow
value
to
historical
daily
mean
flow
values
for
all
days
of
the
year.
The
real­
time
data
used
to
produce
the
maps
have
not
been
evaluated
or
edited.

Army
Corps
of
Engineers
Main
Page:
http://
www.
usace.
army.
mil/
°
Flood
control
and
management
pages.
For
example,
river
and
reservoir
reports
including
flood
level
data
are
available
for
the
St.
Louis
district
of
Missouri
(
see
example
below)
(
http://
mvs­
wc.
mvs.
usace.
army.
mil/
dresriv.
html).

Mississippi
River:

River
Mile
Gage
Station
6
am
Levels
24­
hr
Change
National
Weather
Service
River
Forecasr
Flood
Level
Gage
Zero
Record
Level
Record
Date
Next
3
Days
Crest
Date
309.0
301.2
Hannibal
Dam
22
tw
16.9
15.9
­
0.2
­
0.3
16.6
16.1
15.6
15.8
15.3
14.7
16.0
16.0
449.3
446.1
31.80
29.58
07/
10/
93
07/
16/
93
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
20
US
Bureau
of
Reclamation
Main
Page:
http://
www.
usbr.
gov/
main/
Dams
and
Reservoirs
Page:
http://
www.
usbr.
gov/
dataweb/
html/
dam_
selection.
html
°
The
project
DataWeb
provides
the
most
current
information
on
the
bureau's
projects,
facilities,
and
programs
including
dam
and
reservoir
information
for
western
states.
This
data
can
be
obtained
by
selecting
a
dam
or
from
the
State
and
Region
maps.

The
Department
of
Homeland
Security
(
formerly
FEMA)
Main
Page:
http://
www.
fema.
gov/
Flood
Hazard
Mapping:
http://
www.
fema.
gov/
fhm/
°
The
flood
maps
describe
where
the
flood
risks
are,
based
on
local
hydrology,
topology,
precipitation,
flood
protection
measures
such
as
levees,
and
other
scientific
data.
Fee
to
obtain
maps.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
21
Figure
4.1
Generalized
Depiction
of
Stream
Channel
Lateral
Migration
(
a)
Map
of
a
Stream
Meander;
(
b)
Cross­
section
of
the
Channel
from
A­
A

with
Channel
Positions
at
3
Successive
Times
(
t
0,
t
1,
and
t
2);
(
c)
Map
of
Stream
Meander
Showing
Location
After
Migration
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
22
4.5
Design
and
Construction
This
section
describes
the
type
of
wells
eligible
for
bank
filtration
credits
under
the
LT2ESWTR.
Because
specific
well
construction
requirements
(
e.
g.,
casing
depths)
vary
by
state
and
with
geologic
conditions,
this
guidance
will
address
these
issues
only
briefly
where
appropriate.
Readers
are
referred
to
the
agency
within
their
state
that
makes
regulations
or
recommendations
regarding
well
construction
for
details
on
issues
such
as
casing
depths,
annular
seals,
drilling
methods,
filter
packs,
etc.
Other
good
general
references
on
well
construction
include
Driscoll
(
1986)
and
USEPA
(
1975).

Figure
4.2
Taking
a
Water
Level
Reading
The
pump
house
for
the
horizontal
collector
well
caisson
is
in
the
background.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
23
4.5.1
Well
Type
Only
vertical
and
horizontal
wells
are
eligible
for
bank
filtration
credits.
Other
types
of
ground
water
collection
devices
may
not
provide
adequate
filtration
of
pathogens.
For
example,
a
spring
box
is
a
ground
water
collection
device
located
at
the
ground
surface
and
is
designed
to
contain
spring
outflow
and
protect
it
from
surface
contamination
until
the
water
is
used.
Spring
boxes
are
found
where
local
hydrogeologic
conditions
have
focused
ground
water
discharge
into
a
smaller
area
(
i.
e.,
a
spring)
and
at
a
faster
volumetric
flow
rate
than
elsewhere.
Often,
localized
fracturing
or
dissolution­
enhanced
channels
are
the
cause
of
the
focused
discharge
to
the
spring.
As
noted
in
section
4.4.2.2,
fractures
and
dissolution
channels
have
significant
potential
to
transport
microbial
contaminants.
Thus,
spring
boxes
are
not
eligible
for
bank
filtration
credit.

Infiltration
galleries
(
or
filter
cribs)
are
also
not
eligible
for
bank
filtration
credits.
Infiltration
galleries
are
designed
to
collect
water
infiltrating
from
the
surface,
or
to
intercept
ground
water
flowing
naturally
toward
surface
water,
using
a
slotted
pipe
installed
horizontally
in
a
trench
and
backfilled
with
granular
material
(
Symons
et
al.,
2000).
An
infiltration
gallery
is
not
bank
filtration
because
the
material
overlying
an
infiltration
gallery
may
be
engineered
to
optimize
oocyst
removal.
Bank
filtration
systems
are
defined
as
relying
solely
on
the
natural
properties
of
the
system
to
remove
microbial
contaminants.
An
infiltration
gallery
may,
however,
be
eligible
for
Cryptosporidium
removal
credit
as
an
alternative
treatment
technology
[
40
CFR
141.73(
d)].

Horizontal
and
vertical
wells
are
both
eligible
for
bank
filtration
credits.
They
are
distinguished
from
each
other
by
the
orientations
of
their
well
screens,
and
the
important
implications
this
has
for
their
well
hydraulics
(
Figure
4.3
and
4.4).
Collector
horizontal
wells
are
constructed
by
the
excavation
of
a
central
vertical
caisson
or
pipe.
One
or
more
laterals
(
i.
e.,
collector
lateral
well
screens)
extend
horizontally
from
the
caisson
bottom
and
may
be
very
long.
Laterals
may
extend
radially
in
all
directions
­
resulting
in
a
radial
collector
well­
or
primarily
in
the
direction
of
the
river
(
Driscoll,
1986;
Ray,
2001a).
The
lateral
well
screens
are
often
installed
near
the
bottom
of
the
formation,
allowing
a
greater
proportion
of
the
saturated
thickness
of
the
aquifer
to
be
used.
A
greater
proportion
of
pathogens
and
other
contaminants
are
removed
when
the
distance
between
the
surface
water
body
and
the
laterals
is
increased
(
Ray,
2001a).
Section
4.5.2.2
contains
a
discussion
of
when
it
may
be
appropriate
to
locate
wells
at
separation
distances
greater
than
those
required
by
the
LT2ESWTR.
Laterals
may
extend
underneath
a
surface
water
body
in
the
United
States.
This
is
generally
not
how
horizontal
wells
are
placed
in
Europe
(
Ray
2001a)
because
in
Europe
such
wells
are
required
to
meet
a
55­
60
day
average
travel
time
requirement.
An
example
of
a
pump
house
for
a
horizontal
collector
well
in
Louisville,
KY
is
shown
in
Figure
4.2.
It
is
elevated
to
prevent
flood
waters
from
entering
it.

The
choice
between
using
a
vertical
or
horizontal
well
for
a
bank
filtration
system
depends
on
the
site
hydrogeology
and
the
pumping
requirements.
For
systems
with
large
production
requirements
(
e.
g.,
many
Public
Water
Systems)
or
for
pumping
in
shallow
alluvial
aquifers,
horizontal
wells
may
be
preferred
because
they
are
designed
to
capture
large
volumes
of
surface
water
recharge
with
little
drawdown
(
Driscoll,
1986).
Vertical
wells
with
large
production
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
24
requirements
are
not
well
suited
to
shallow
alluvial
aquifers
because
the
necessary
low
drawdown
cannot
be
sustained
(
Ray,
2001a).

Finally,
a
comparison
of
construction
expense
with
the
costs
of
well
maintenance
may
play
a
role
in
the
choice
of
well
type.
Horizontal
collector
wells
are
substantially
more
costly
than
vertical
wells
(
Driscoll,
1986).
However,
moderately
large
utilities
may
need
many
smaller
capacity
vertical
wells
to
match
the
capacity
of
a
horizontal
well.
The
maintenance
of
these
vertical
wells
may
require
significant
effort
and
expense
(
Ray,
2001a).
In
such
cases,
horizontal
collector
wells
may
be
preferred.

Figure
4.3
Schematic
Showing
Generalized
Flow
and
Required
Separation
Distance
to
a
Vertical
Well
(
Inset
shows
tortuous
ground
water
flow
at
the
micro­
scale.)
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
25
Figure
4.4
Schematic
Showing
Generalized
Flow
and
Required
Separation
Distance
to
a
Horizontal
Well
With
Three
Laterals
4.5.2
Filtrate
Flow
Path
and
Well
Location
For
systems
to
receive
Cryptosporidium
log
removal
credits,
the
ground
water
flow
path
length
between
the
edge
of
the
surface
water
body
and
the
well
is
expected
to
be
sufficient
for
effective
oocyst
removal.
This
section
discusses
EPA's
requirements
for
appropriate
flow
path
lengths,
and
associated
well
locations,
for
the
log
removal
credits
available
under
the
LT2ESWTR.
The
ground
water
flow
path
length
necessary
to
receive
credits
is
specified
for
both
vertical
and
horizontal
wells.
A
discussion
of
how
to
obtain
information
necessary
to
define
the
edge
of
the
surface
water
body
is
also
included.

4.5.2.1
Required
separation
distance
between
a
well
and
the
surface
water
source
Cryptosporidium
oocyst
removal
may
vary
significantly
throughout
the
year
in
many
bank
filtration
systems.
At
most
typical
bank
filtration
locations,
high
log
removal
rates
(
e.
g.
3.5
log
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
26
removal
over
13
m)
may
be
expected
with
the
surface
water
discharges
that
predominate
during
most
of
the
year.
During
short
flood
periods,
however,
there
may
be
substantially
lower
removal
(
e.
g.
0.5
to
1.0
log
removal
over
13
m)
due
to
scouring
of
the
surface
water
 
ground
water
interface,
as
discussed
below
in
section
4.6.2.
In
summary,
a
number
of
different
factors
may
contribute
to
increased
risk
of
Cryptosporidium
reaching
wells.
These
factors
include
the
presence
of
coarse­
grained
aquifer
or
stream
bed
sediments,
high
river
velocities,
and
frequent
scouring
of
riverbeds.
Given
the
need
to
protect
water
supplies
during
periods
of
high
surface
water
discharge
with
their
potentially
lower
log
removal
capabilities,
the
LT2ESWTR
rule
language
(
40
CFR
141.726(
c))
provides
0.5
log
removal
credit
for
systems
with
bank
filtration
wells
located
greater
than
25
feet
from
a
surface
water
source
and
1.0
log
removal
credit
for
wells
located
greater
than
50
feet
from
a
surface
water
source.

4.5.2.2
Locating
wells
at
greater
than
required
distances
from
the
surface
water
source
Given
the
dynamic
nature
of
riverbanks
and
aquifer
systems,
including
scouring
processes,
as
discussed
in
section
4.3.1.3,
it
may
sometimes
be
advisable
to
place
bank
filtration
wells
at
distances
greater
than
25
or
50
feet
from
a
surface
water
source.
This
extra
precaution
may
also
be
advisable
when
a
system
is
uncertain
as
to
whether
the
riverbed
and
bank
contain
sufficient
fine­
grained
material
to
provide
adequate
removal
of
Cryptosporidium
oocysts.
That
is,
EPA
is
requiring
the
separation
distances
of
25
feet
and
50
feet
for
the
log
removal
credits
discussed
above,
but
greater
separation
distances
may
result
in
additional
public
health
protection
at
some
sites.
The
disadvantage
of
using
greater
separation
distances
between
the
surface
water
source
and
the
bank
filtration
well
is
that
water
yields
to
the
well
will
be
decreased.
When
a
system
makes
a
decision
to
place
wells
at
a
greater
distance
from
a
surface
water
source
than
EPA
requires,
it
will
need
to
balance
the
sacrifice
in
well
yield
with
the
added
public
health
protection.

The
remainder
of
section
4.5.2.2
discusses
geophysical
methods
which
may
be
of
use
in
constructing
a
conceptual
model
of
subsurface
flow
conditions
in
riverbank
filtration
systems.
By
obtaining
hydrogeologic
information
through
geophysical
or
other
means
(
e.
g.,
pre­
existing
hydrogeologic
or
geologic
maps),
systems
can
determine
the
degree
to
which
local
conditions
may
affect
Cryptosporidium
removal
at
the
bank
filtration
site.
For
example,
if
mapping
the
bedrock­
alluvial
interface
and
the
water
table
at
a
particular
site
indicates
that
the
aquifer
is
fairly
thin,
it
is
unlikely
that
infiltrating
river
water
will
be
diluted
by
much
ambient
ground
water.
In
such
a
case
it
may
be
advisable
to
locate
wells
at
greater
than
the
required
distance
from
the
surface
water
source.
On
the
other
hand,
if
detailed
hydrogeologic
investigations
indicate
that
the
aquifer
contains
a
large
proportion
of
fine­
grained
sediments,
it
would
not
be
advisable
to
locate
the
well
at
greater
than
the
required
distance
from
the
surface
water
source,
because
the
aquifer
is
already
likely
to
be
an
efficient
pathogen
filter,
and
it
would
be
inadvisable
to
further
sacrifice
well
yields.

When
the
aquifer
contains
fine­
grained
material,
it
is
possible
that
well
over­
pumping
may
break
the
hydraulic
connection
between
ground
water
and
surface
water,
yielding
a
variably
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
27
saturated
zone
underneath
a
perched
stream,
as
shown
below
in
Figure
4.5.
Formation
of
such
a
variably
saturated
zone
during
periods
of
high
pumping
can
greatly
alter
the
existing
ground
water
flow
paths.
New
ground
water
flow
paths
could
result
in
marked
changes
in
water
quality.
For
example,
surface
water
infiltration
could
occur
further
upstream,
resulting
in
a
longer
ground
water
flow
path
for
infiltrating
surface
water
flowing
towards
the
well.
The
increase
in
flow
pathlength
could
improve
water
quality.
Alternatively,
the
result
of
over­
pumping
could
be
decreased
water
quality.
This
may
occur
because
the
decreased
thickness
of
the
saturated
aquifer
­
due
to
the
formation
of
a
large
variably
saturated
zone
­
may
cause
faster
ground
water
flow
(
assuming
pumping
rates
remain
constant).
Faster
ground
water
flow
provides
less
time
for
contaminant
attenuation
within
the
aquifer.
Finally,
the
variably
saturated
zone
itself,
to
the
extent
that
it
transmits
water,
can
improve
water
quality
because
contaminant
attenuation
is
usually
increased
under
variably
saturated
conditions.
If
possible,
the
potential
for
formation
of
a
variably
saturated
zone
can
be
investigated
in
order
to
provide
additional
information
regarding
the
desirability
of
locating
wells
at
greater
than
required
distances
from
the
surface
water
source.

Figure
4.5
The
Streambed
of
a
Perched
Stream
Is
Well
above
the
Water
Table
Geophysical
methods
generally
do
not
disturb
subsurface
materials.
They
are
often
less
expensive
than
labor­
intensive
digging
of
trial
pits
or
drilling
of
boreholes.
Furthermore,
the
useful
information
gleaned
by
using
geophysical
methods
can
aid
in
choosing
the
best
locations
for
wells
(
Reynolds,
1997).
Geophysical
methods
include
gravity
and
magnetic
methods,
seismic
methods,
electrical
methods,
and
ground
penetrating
radar.

Hydrogeophysical
methods
can
be
used
in
pre­
existing
boreholes,
thereby
providing
high
resolution
data
for
a
very
localized
area
around
the
borehole.
Alternatively,
surface
geophysical
methods
can
be
used
to
obtain
more
generalized
information
over
a
large
area,
including
information
on
the
depth
to
the
water
table,
the
depth
to
bedrock,
and
stratigraphy
(
Hubbard,
2003).
The
discussion
below
provides
only
a
generalized
overview
of
currently
available
geophysical
methods.
More
detailed
information
can
be
obtained
from
texts
by
Hearst
(
2000),
Reynolds
(
1997),
Rubin
and
Hubbard
(
2003),
Keys
(
1990)
and
Burger
(
1992).
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
28
Gravity
surveying
measures
variations
in
the
acceleration
due
to
the
Earth's
gravitational
field
which
are
caused
by
density
variations
in
subsurface
rocks.
Subsurface
cavities
can
be
detected
with
this
technology,
however
sites
with
such
cavities
would
not
be
suitable
for
bank
filtration.
Reynolds
(
1997)
states
that
gravity
methods
are
fairly
uncommon
in
hydrogeological
work
compared
to
electrical
methods.
On
the
other
hand,
in
the
Arizona
district
of
the
United
States
Geological
Survey,
gravity
methods
have
been
in
use
for
over
15
years
to
evaluate
changes
in
water
storage
in
aquifers.
These
methods
can
detect
water
table
elevation
changes
of
as
little
as
a
few
inches
(
Callegary,
2003).
Thus,
gravity
methods
may
be
useful
at
riverbank
filtration
sites
for
assessing
the
depth
to
water
table,
aquifer
thickness,
and
seasonal
effects
on
the
dilution
of
infiltrating
river
water
with
ambient
groundwater.
Magnetic
surveying
or
magnetic
anomolies
can
also
be
used
in
hydrogeologic
investigations.
For
example,
clay
infilling
bedrock
cavities
can
be
detected
due
to
slight
changes
in
the
magnetic
susceptibility
of
clay
and
most
bedrock
(
Reynolds,
1997).

Seismic
methods
are
widely
used
in
hydrogeologic
investigations.
Applied
seismology
involves
generating
a
signal
through
an
explosion
or
other
method
at
a
specific
time.
The
generated
seismic
waves
travel
through
the
subsurface,
are
reflected
and
refracted
back
to
the
surface,
and
the
return
signals
are
detected
on
monitoring
instruments.
The
amount
of
time
that
elapses
is
the
basis
for
determining
the
nature
of
subsurface
layers/
materials
(
Reynolds,
1997).
Reynolds
(
1997)
provides
a
detailed
example
of
the
use
of
seismic
refraction
surveying
for
locating
the
bedrock/
alluvial
interface
at
one
particular
site.

Seismic
methods
can
be
used
to:

°
estimate
depth
to
bedrock
(
ideal
for
riverbank
filtration
applications)
°
determine
the
nature
of
bedrock
(
e.
g.,
cavernous)
or
location
of
cavities.
Note
that
karst
buried
by
alluvium
may
contain
unexpected
ground
water
flowpaths.
°
determine
the
location
of
faults
that
may
juxtapose
bedrock
against
alluvial
material
°
determine
stratigraphy
(
useful
where
sands
and
clays
may
be
interlayered)
°
determine
porosity
°
determine
ground
water
particle
velocities
(
an
important
parameter
for
riverbank
filtration
systems)

Electrical
resistivity
methods
are
used
extensively
in
downhole
logging
to
identify
hydrogeologic
units
that
will
produce
high
flow
rates.
Electrokinetic
surveying
makes
use
of
electrodes
implanted
at
the
ground
surface
to
identify
the
location
of
the
water
table.
This
may
be
useful
at
riverbank
filtration
sites,
where
water
table
layer
and
depth
to
bedrock
can
be
used
to
determine
aquifer
thickness
­
an
important
parameter
in
determining
how
much
dilution
of
bank
filtrate
with
ambient
groundwater
is
occurring.
A
more
recent
development
is
the
use
of
electrokinetic
methods
to
measure
flowrates
in
boreholes
(
Reynolds,
1997).

The
spontaneous
polarisation
or
self­
potential
(
SP)
method
is
conducted
by
measuring
differences
in
ground
electrical
potential
at
different
locations,
but
is
still
fairly
uncommon.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
29
Another
electrical
method,
the
induced
polarisation
(
IP)
method
can
be
used
to
detect
ground
water
and
water
tables,
however
electromagnetic
induction
methods
are
generally
considered
more
practical
for
these
purposes
in
the
field.
Contaminated
ground
water
within
subsurface
clays
can
also
sometimes
be
detected
with
the
IP
method
(
Reynolds,
1997).

Electromagnetic
(
EM)
methods
have
been
used
in
groundwater
investigations
to
delineate
contaminant
plumes,
and
thus
can
be
useful
in
conceptualizing
flow
systems
in
a
riverbank
filtration
context
when
the
quality
of
infiltrating
river
water
is
especially
poor.
Pulse­
transient
EM
(
TEM)
surveys
(
a
type
of
EM
method)
may
be
useful
in
conceptualizing
flow
for
riverbank
filtration
systems
where
infiltrating
water
quality
is
poor.
It
may
also
be
useful
in
monitoring
the
quality
of
infiltrating
water.
When
data
is
available
from
both
borehole
and
surface
instruments,
EM
and
electrical
methods
can
be
used
to
map
subsurface
geology
such
as
the
locations
of
coarse­
grained
and
fine­
grained
units.

Ground
penetrating
radar
(
GPR)
has
been
used
as
a
surface
method
for
contaminant
plume
mapping
and
monitoring
pollutants
in
groundwater.
To
operate
such
a
system,
a
signal
generator,
transmitting
and
receiving
antennae,
and
a
receiver
must
be
used.
Radiowaves
are
generated,
which
travel
in
a
broad
beam
at
high
speeds.
Energy
is
lost
or
attenuated
depending
on
the
subsurface
materials
through
which
the
waves
travel.
GPR
has
proven
valuable
in
mapping
sediment
sequences,
and
can
be
used
to
investigate
sediments
through
freshwater
up
to
27
m
deep
(
Reynolds,
1997).
Thus,
it
may
be
of
use
in
gaining
information
about
the
composition
of
riverbeds,
and
for
monitoring
the
effects
of
scour
on
riverbed
composition.
GPR
can
also
be
used
to
locate
water
tables,
delineate
sedimentary
structures
which
may
contain
pockets
of
coarsegrained
alluvium,
and
determine
the
spatial
extent
and
continuity
of
buried
clay
and
peat
layers
within
subsurface
deposits.
Borehole
radar
can
also
be
used
for
hydrogeologic
investigations.

Before
choosing
a
specific
geophysical
method
it
may
be
important
to
consider
the
following:
desired
level
of
resolution,
area
of
coverage,
site­
specific
conditions
and
their
influence
on
the
applicability
of
the
method,
possible
non­
uniqueness
of
the
geophysical
attribute,
resources
needed
to
interpret
the
geophysical
data,
and
possible
integration
with
direct
measurements.
In
general,
mapping
the
water
table
and
finding
the
depth
to
bedrock
are
considered
standard
hydrogeophysical
procedures.
Other
applications
such
as
estimating
permeabilities
or
porosities
are
at
an
earlier
stage
of
development
and
may
not
yet
be
appropriate
for
routine
use
at
riverbank
filtration
sites
(
Hubbard,
2003).

4.5.2.3
Delineating
the
edge
of
the
surface
water
source
The
flow
paths
due
to
induced
infiltration
to
a
vertical
well
have
both
vertical
and
horizontal
components,
and
are
tortuous
at
the
micro­
scale
(
Figure
4.3).
Such
flow
will
typically
have
a
significant
horizontal
component,
especially
if
the
vertical
well
is
screened
in
a
shallow,
unconsolidated,
alluvial
aquifer
that
is
eligible
for
bank
filtration
credits.
Therefore,
for
the
purpose
of
receiving
log
removal
credits,
the
flow
path
length
to
a
vertical
well
is
to
be
determined
using
the
measured
horizontal
distance
from
the
edge
of
the
surface
water
body
to
the
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
30
well
intake.
The
edge
of
the
surface
water
body
is
defined
as
the
edge
of
either
the
100­
year
floodplain
or
the
floodway,
discussed
below.
The
100­
year
floodplain
is
defined
by
its
boundary
­
the
flood
elevation
that
has
a
one
percent
chance
of
being
equaled
or
exceeded
each
year.

As
a
first
step,
utilities
may
use
the
online
maps
available
at
the
following
website
to
get
a
general
idea
of
the
mapped
extent
of
the
100­
year
floodplain
in
their
area:
http://
www.
esri.
com/
hazards/
makemap.
html.
In
order
to
satisfy
the
requirements
of
the
LT2ESWTR
for
the
location
of
the
wells
of
a
bank
filtration
system,
however,
an
official
Federal
Emergency
Management
Agency
(
FEMA)
(
now
part
of
the
Department
of
Homeland
Security)
flood
hazard
map
must
be
used.
Such
maps
can
be
ordered
in
either
paper
or
digital
formats
from
FEMA.
The
following
website
can
be
used
to
order
these
maps:
http://
msc.
fema.
gov/
MSC/.

Although
in
some
areas
of
the
United
States
the
mapped
extent
of
the
100­
year
floodplain
may
be
more
easily
accessible
than
the
mapped
extent
of
the
floodway,
some
utilities
may
choose
to
use
the
edge
of
the
floodway
as
a
starting
point
for
measuring
separations
distances
to
wells
because
it
typically
allows
wells
to
be
placed
slightly
closer
to
the
river
and
is
thus
a
somewhat
less
restrictive
requirement.
The
floodway
is
a
regulatory
concept,
and
is
defined
as
that
portion
of
the
overbanks
that
must
be
kept
free
from
encroachment
to
discharge
the
one
percent
annual
chance
flood
(
i.
e.
the
100­
year
flood)
without
increasing
flood
levels
by
more
than
1.0
foot.
It
is
determined
by
specified
methods
according
to
FEMA
guidelines,
as
described
below.

For
many
areas,
the
mapped
extent
of
the
floodway
will
also
be
drawn
on
the
flood
hazard
map
obtained
by
FEMA.
The
utility
may
choose
to
use
the
edge
of
the
floodway
rather
than
the
edge
of
the
100­
year
floodplain
for
the
purpose
of
determining
the
required
separation
distance
between
a
river
and
a
riverbank
filtration
well.
If
the
mapped
extent
of
the
floodway
is
unavailable,
the
utility
may
opt
to
perform
the
mapping
using
one
of
a
number
of
hydraulic
models
approved
by
FEMA.
A
list
of
these
approved
models
is
available
at
http://
www.
fema.
gov/
mit/
tsd/
en_
hydra.
htm.
EPA
recommends
using
the
US
Army
Corps
of
Engineers'
HEC­
RAS
model
for
mapping
floodway
limits.
The
HEC­
RAS
software,
User's
Manual,
Applications
Guide,
and
Hydraulic
Reference
Manual
are
available
for
free
downloading
from
http://
www.
hec.
usace.
army.
mil/
software/
software_
distrib/
index.
html.

When
a
utility
elects
to
determine
the
edge
of
the
floodway,
and
to
model
the
floodway
boundaries
if
they
are
not
available
from
FEMA,
the
preferred
encroachment
method
within
HECRAS
is
Method
4.
Method
4
can
be
summarized
as
follows,
according
to
FEMA's
Map
Assistance
Center
(
2003):

The
Method
4
encroachment
operates
by
analyzing
the
hydraulic
conveyance
for
the
unencroached
one
percent
annual
chance
floodplain
at
each
cross
section,
then
equally
reducing
the
conveyance
from
both
overbank
areas
by
moving
toward
the
stream
channel
from
the
edge
of
the
floodplain
until
the
resulting
water­
surface
elevation
is
one
foot
higher
than
the
unencroached
elevation,
and
the
resulting
encroached
conveyance
is
approximately
equal
to
the
unencroached
conveyance.
The
new
left
and
right
cross­
section
limits
are
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
31
assumed
to
be
vertical
walls.
Finally,
a
backwater
energy
balance
is
calculated
using
the
new
cross
sections,
which
results
in
the
encroached
or
floodway
water­
surface
profile.
The
floodway
modeling
process
requires
adjustments
and
rerunning
of
the
model
because
the
final
calculation
is
the
backwater
energy
balance
between
new
cross
sections.
Many
times
the
1.0­
foot
target
cannot
be
achieved
exactly
at
each
cross
section
because
of
energy
balance
considerations.
Floodplain
geometry,
constrictions
at
culvert
and
bridge
crossings,
and
constrictions
from
other
man­
made
obstructions
in
the
floodplain
may
require
adjustments
to
the
encroachment
widths
to
stay
at
or
below
the
1.0­
foot
maximum
water­
surface
increase.
Chapter
10
of
the
HEC­
RAS
User's
Manual
includes
a
discussion
of
performing
a
floodway
encroachment
analysis.

In
most
areas,
however,
EPA
expects
that
utilities
will
find
it
preferable
and
simpler
to
use
the
previously
mapped
limits
of
the
100­
year
floodplain
to
determine
the
edge
of
the
river
for
riverbank
filtration
separation
distances.

4.5.2.4
Measuring
separation
distances
for
horizontal
wells
and
wells
that
are
neither
horizontal
nor
vertical
As
noted
in
section
4.5.1,
horizontal
wells
may
have
laterals
that
extend
underneath
a
surface
water
body.
The
flow
direction
for
induced
infiltration
to
a
horizontal
well
that
extends
under
a
surface
water
body
is
predominately
downward.
Therefore,
the
flow
path
length
to
a
horizontal
well
is
the
measured
vertical
distance
from
the
bed
of
the
river
under
normal
flow
conditions
to
the
closest
horizontal
well
lateral's
intake
(
Figure
4.4).

Some
wells
may
be
constructed
so
that
the
well
is
neither
truly
horizontal
nor
truly
vertical.
In
these
cases,
there
is
greater
uncertainty
about
the
definition
of
separation
distance
from
surface
water.
For
simplicity,
if
the
well
if
closer
to
being
a
vertical
well
than
to
being
a
horizontal
well
(
i.
e.
the
well
is
oriented
at
greater
than
a
45
degree
angle
to
a
horizontal
line),
the
separation
distance
is
defined
for
the
purposes
of
this
toolbox
option
to
be
the
horizontal
distance
from
the
edge
of
the
river
to
the
closest
(
in
terms
of
horizontal
distance)
intake
on
the
well.
Similarly,
if
the
well
is
closer
to
being
a
horizontal
well
as
opposed
to
a
vertical
well,
separation
distance
is
defined
as
the
shortest
possible
vertical
distance
from
the
riverbed
to
an
intake
on
the
well.
To
ensure
that
the
assigned
log
removal
credit
is
realized,
systems
are
expected
to
perform
continuous
turbidity
monitoring
for
all
wells
that
receive
a
credit.
Continuous
turbidity
monitoring
is
discussed
in
section
4.2.2.

4.6
Operational
Considerations
4.6.1
High
River
Stage
When
the
river
stage
(
i.
e.
the
elevation
of
the
water
surface)
is
high,
the
increased
head
gradient
between
the
river
and
the
adjacent
aquifer
results
in
increased
infiltration
and
increased
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
32
ground
water
flow
rates.
This
condition
can
be
expected
to
occur
periodically
throughout
the
year
at
many
sites,
and
will
generally
be
associated
with
reduced
log
removals
(
Gollnitz,
1999;
Ray,
2001b).
High
river
stage
is
often
associated
with
scouring
of
riverbed
sediments.
Nevertheless,
even
when
scour
does
not
occur,
the
high
ground
water
velocities
associated
with
high
river
stage
can
be
a
significant
threat
to
a
riverbank
filtration
system.

One
solution
to
this
problem
is
that
pumping
rates
can
be
temporarily
decreased
during
periods
of
high
river
flow
(
Medema
et
al.,
2000).
Decreased
pumping
rates
will
in
turn
decrease
the
head
gradient
between
the
river
and
the
well,
thereby
decreasing
subsurface
velocities,
increasing
residence
times,
and
facilitating
pathogen
inactivation.

4.6.2
Implications
of
Scour
for
Bank
Filtration
System
Operations
Periodic,
short­
term
flood
scour
can
have
both
negative
and
positive
impacts
on
the
performance
of
a
bank
filtration
system.
As
noted
in
section
4.5.2
above,
lower
log
removals
of
oocysts
are
expected
during
floods
because
higher
river
shear
velocities
and
associated
increases
in
bedload
transport
mobilize
fine
sediments
deposited
when
discharges
were
lower.

Removal
of
fine
sediments
opens
large
pore
spaces,
increasing
the
hydraulic
conductivity
across
the
surface
water
 
ground
water
interface
(
Gollnitz,
1999;
Ray,
2001a;
Ray,
2001b).
Unfortunately,
this
potentially
increases
the
number
of
pathogens
transported.
Furthermore,
the
microbial
activity
and
unique
geochemical
environment
of
the
riverbed,
which
serves
to
facilitate
the
removal
of
pathogens
via
sorption
and
other
processes,
may
not
be
present
for
short
periods
following
flood
scour.
Recent
work
in
Germany
(
Baveye
et
al.,
2003)
suggests
that
the
biologically
active
zone
is
re­
established
very
quickly
after
scour,
perhaps
within
3
days,
at
least
when
measured
in
terms
of
the
ability
to
degrade
certain
organic
compounds.
Limited
scour
can
reduce
clogging
at
the
surface
water
 
ground
water
interface
and
improve
well
yields
(
Wang
et
al.,
2001).
The
continuous
turbidity
monitoring
required
by
the
LT2ESWTR
for
bank
filtration
credits
can
be
used
to
help
systems
manage
the
threat
posed
by
periodic,
short­
term
flood
scour.

When
high
river
stages
or
high
turbidity
levels
indicate
that
flood
scour
may
be
occurring
and
compromising
the
effectiveness
of
a
bank
filtration
system,
pumping
rates
can
be
decreased.
This
will
lead
to
lower
velocities
and
longer
subsurface
residence
times,
thereby
increasing
the
protectiveness
of
the
system
(
Medema
et
al.,
2000;
Juhasz­
Holterman,
2000).

4.6.3
Anticipating
high
flow
events
/
flooding
Many
factors
are
involved
in
increasing
the
probability
that
a
flood
will
occur.
Intense
rainfall
is
the
most
apparent
factor,
however
the
geomorphology
of
a
watershed
is
important
in
determining
how
quickly
water
will
enter
a
stream
system
after
a
rainfall
event,
as
well
as
how
quickly
water
will
enter
a
major
river
from
smaller
tributaries.
Systems
can
anticipate
that
a
high
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
33
flow
event
will
occur
if
a
rapid
spring
thaw
follows
a
winter
of
unusually
heavy
snowfall.
It
is
also
important
to
be
aware
of
recent
changes
in
vegetation
due
to
wildfires
or
urbanization.
When
vegetation
is
removed
or
decreased
there
are
fewer
barriers
to
rapid
surface
runoff,
plant
roots
no
longer
keep
soil
loose
and
permeable
(
thus
more
compact
soils
will
be
less
able
to
decrease
surface
runoff),
and
plants
themselves
will
be
unavailable
to
take
in
a
certain
proportion
of
precipitation
(
Montgomery,
2000).
Therefore,
systems
may
wish
to
monitor
for
pathogens
more
frequently
or
change
pumping
regimes
in
riverbank
filtration
systems
when
high
flows
are
anticipated.

4.6.4
Possible
responses
to
spill
events
and
poor
surface
water
quality
One
response
to
a
serious
water
quality
threat
is
to
stop
pumping
from
all
bank
filtration
production
wells.
Other
pumping
regime
changes
can
also
be
implemented
to
reduce
risks,
including
decreasing
the
number
of
hours
the
system
is
in
operation
each
day.
For
systems
that
have
a
number
of
wells
in
operation,
it
may
be
advisable
to
increase
pumping
rates
for
wells
further
from
the
surface
water
source
and
decrease
pumping
rates
for
wells
that
are
closer
(
Juhasz­
Holterman,
2000).
Juhasz­
Holterman
(
2000)
recommended
that
this
kind
of
change
be
implemented
seasonally
at
a
site
in
the
Netherlands.
Her
study
of
the
site's
hydrology
indicated
that
during
the
winter
months
pumping
wells
were
more
vulnerable
to
contamination
due
to
"
short­
circuited"
flow
paths
from
the
polluted
river
through
the
subsurface.
Her
solution
involved
both
restricting
extraction
rates
to
a
few
hours
a
day
(
which
was
acceptable
due
to
decreased
demand
during
the
winter
months)
as
well
as
an
altered
pumping
regime
which
relied
more
on
wells
located
further
from
the
river.

4.6.5
Maintaining
required
separation
distances
Alluvial
rivers
that
are
experiencing
active,
progressive
erosion
as
an
adjustment
to
new
flooding
regimes
or
sediment
loads,
or
in
relation
to
natural
lateral
migration,
may
pose
serious,
longer­
term
challenges
to
bank
filtration
systems.
For
example,
significant
log
removal
reductions
may
be
more
frequent
in
an
urbanizing
basin
as
a
consequence
of
more
frequent
flooding
and
associated
scouring.
In
extreme
cases,
long
term
degradation
of
the
bed
or
banks
may
reduce
the
threshold
separation
distances
between
the
surface
water
source
and
bank
filtration
well.
Recall
that
these
separation
distances
­
25
feet
for
0.5
log
removal
credit
and
50
feet
for
1.0
log
removal
credit
­
are
required
to
receive
log
removal
credits
under
the
LT2ESWTR.

Systems
may
wish
to
assess
their
sites
for
active,
progressive
erosion.
Lateral
migration
rates
can
be
calculated
using
sequential
aerial
photography
and/
or
topographic
maps,
if
available.
Systems
without
such
data
may
need
to
obtain
the
needed
information
by
conducting
sequential
field
surveys
of
the
floodplain
area
proposed
for
the
site.
This
will
require
a
far
more
lengthy
investigation
period.
Progressive
downcutting
could
also
be
measured
with
sequential
field
surveys
of
the
channel
bed
elevation
over
a
period
of
years.
Regardless
of
the
method
used,
the
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
34
threshold
separation
distances
between
the
surface
water
source
and
the
bank
filtration
well
must
be
maintained.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
35
References
ASTM
(
American
Society
for
Testing
and
Materials),
2003.
Standard
Test
Method
for
Sieve
Analysis
of
Fine
and
Coarse
Aggregates
­
Standard
C
136­
1.

Baveye,
P.,
P.
Vandevivere,
B.
L.
Hoyle,
P.
C.
DeLeo,
and
D.
Sanchez
de
Lozada.
1998.
Environmental
impact
and
mechanisms
of
the
biological
clogging
of
saturated
soils
and
aquifer
materials.
Critical
Reviews
in
Environmental
Science
and
Technology.
28(
2):
123­
191.

Baveye,
P.,
Berger,
P.,
Schijven,
J.,
and
Grischek,
T.
2003.
Research
needs
to
improve
knowledge
of
bank
filtration
removal
of
pathogens,
in
Riverbank
Filtration:
Improving
Source
Water
Quality,
edited
by
Ray,
C.,
Melin,
G.
and
Linsky,
R.,
Kluwer,
Dordrecht,

Berger,
P.
2002.
Removal
of
Cryptosporidium
Using
Bank
Filtration
in
Riverbank
Filtration:
Understanding
Contaminant
Biogeochemistry
and
Pathogen
Removal,
C.
Ray
(
ed.).
The
Netherlands:
Kluwer
Academic
Publishers.
p.
85­
121.

Burger,
H.
R.,
D.
C.
Burger,
and
R.
H.
Burger,
1992.
Exploration
Geophysics
of
the
Shallow
Subsurface.
Upper
Saddle
River,
NJ:
Prentice
Hall.

Booth,
D.
B.
1990.
Stream­
channel
incision
following
drainage
basin
urbanization.
Water
Resources
Bulletin
26(
3):
407
 
417.

Callegary,
James,
United
States
Geological
Survey,
personal
communication,
3/
03.

Chow,
V.
T.
1964.
Handbook
of
Applied
Hydrology.
New
York:
McGraw
Hill.

Driscoll,
F.
G.
1986.
Groundwater
and
Wells.
2nd
Edition.
St.
Paul,
Minnesota:
Johnson
Division.

Dunne,
T.,
and
Leopold,
L.
B.
1978.
Water
in
Environmental
Planning.
New
York:
W.
H.
Freeman
and
Company.
818
pp.

FEMA
Map
Assistance
Center.
2003.
Personal
communication.

Juttner,
F.
1995.
Elimination
of
terpenoid
odorous
compounds
by
slow
sand
and
river
bank
filtration
of
the
Ruhr
River,
Germany.
Water
Science
&
Technology.
31:
211­
217.

Goldenberg,
L.
C.,
I.
Hutcheon,
N.
Wardlaw,
and
A.
J.
Melloul.
1993.
Rearrangement
of
fine
particles
in
porous
media
causing
reduction
of
permeability
and
formation
of
preferred
pathways
of
flow:
experimental
findings
and
a
conceptual
model.
Transport
in
Porous
Media
13:
221­
237.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
36
Gollnitz,
W.
D.
1999.
Induced
infiltration
rate
variability
and
water
quality
sampling
issues.
Proceedings
of
the
International
Riverbank
Filtration
Conference.
November
4­
6,
1999.
Louisville,
Kentucky.

Harter,
T.,
S.
Wagner,
and
E.
R.
Atwill,
2000.
Colloid
Transport
and
Filtration
of
Cryptosporidium
parvum
in
Sandy
Soils
and
Aquifer
Sediments,
Environmental
Science
and
Technology,
34(
1),
pp.
62­
70.

Hearst,
J.
R.,
P.
H.
Nelson,
and
F.
L.
Paillet,
2000.
Well
Logging
for
Physical
Properties:
A
Handbook
for
Geophysicists,
Geologists,
and
Engineers,
2nd
Edition.
New
York:
Wiley.

Hubbard,
Susan
M.,
Lawrence
Berkeley
National
Laboratory,
personal
communication,
3/
03.

Hubbs,
S.,
J.
Z.
Wang,
and
R.
Song.
2001.
Use
of
Riverbank
Filtration
to
Meet
the
Requirements
of
SWTR
and
DBP
Rules.
Presentation
at
the
American
Water
Works
Association
Water
Quality
Technology
Conference,
November
11­
15,
Nashville,
TN.

Jacobson,
R.
B.,
S.
R.
Femmer,
and
R.
A.
McKenney.
2001.
Land­
use
Changes
and
the
Physical
Habitat
of
Streams
:
a
review
with
emphasis
on
studies
within
the
U.
S.
Geological
Survey
Federal­
State
cooperative
program.
U.
S.
Geological
Survey
Circular
1175.
Reston,
VA.:
U.
S.
Geological
Survey.
63
pp.

Juhasz­
Hoterman,
M.
H.
A.
2000.
Reliable
drinking
water
by
bank
filtration
along
the
river
Maas
(
Meuse),
by
knowledge
of
the
system
combined
with
simple
resources,
in
Proceedings
of
International
Riverbank
Filtration
Conference,
Nov.
2­
4,
Duesseldorf,
W.
Julich
and
J.
Schubert
(
eds.),
International
Arbeitgemeinschaft
der
Wasserwerke
im
Rheineinzugsgebiet,
Amsterdam.

Jüttner,
F.
1999.
Efficacy
of
bank
filtration
for
the
removal
of
fragrance
compounds
and
aromatic
hydrocarbons.
Water
Science
Technology
40(
6):
123­
128.

Keys,
W.
S.,
1990,
Borehole
geophysics
applied
to
groundwater
investigations:
U.
S.
Geological
Survey,
Techniques
of
Water­
Resources
Investigations
of
the
U.
S.
Geological
Survey,
Book
2,
Collection
of
Environmental
Data,
Chapter
E2,
150
p.

Kuehn,
W.
and
U.
Mueller.
2000.
Riverbank
filtration:
an
overview.
AWWA
Journal.
92(
12):
60­
69.

Leopold,
L.
B.
and
Maddock,
T.
1953.
The
Hydraulic
Geometry
of
Stream
Channels
and
Some
Physiographic
Implications.
U.
S.
Geological
Survey
Professional
Paper
252.
Washington:
U.
S.
Government
Printing
Office.

Leopold,
L.
B.,
M.
G.
Wolman,
and
J.
P.
Miller.
1964.
Fluvial
Processes
in
Geomorphology.
San
Francisco:
W
H
Freeman
and
Co.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
37
Mackin,
J.
H.
1948.
Concept
of
the
graded
river.
Geological
Society
of
America
Bulletin
59:
463­
512.

Medema,
G.
J.,
M.
H.
A.
Juhasz­
Hoterman,
and
J.
A.
Luitjen.
2000.
Removal
of
micro­
organisms
by
bank
filtration
in
a
gravel­
sand
soil,
in
Proceedings
of
International
Riverbank
Filtration
Conference,
Nov.
2­
4,
Duesseldorf,
W.
Julich
and
J.
Schubert
(
eds.),
International
Arbeitgemeinschaft
der
Wasserwerke
im
Rheineinzugsgebiet,
Amsterdam.

Miettinen,
I.
T.,
P.
J.
Martikainen,
T.
Vartiainen.
1994.
Humus
transformation
at
the
bank
filtration
water
plant.
Water
Science
&
Technology,
30
(
10):
179.

Montgomery,
Carla
W.,
2000.
Environmental
Geology,
updated
5th
edition.
Boston:
McGraw
Hill.

Oberdorfer,
J.
A.
and
F.
L.
Peterson.
1985.
Waste­
water
injection:
geochemical
and
biogeochemical
clogging
processes.
Ground
Water
23:
753­
761.

Orlob,
G.
T.
and
G.
N,
Radhakrishna.
1958.
The
effects
of
entrapped
gases
on
the
hydraulic
characteristics
of
porous
media.
Transactions
of
the
American
Geophysical
Union
39:
648­
659.

Purdue
University.
2001.
Well
Water
Location
and
Condition
on
the
Farm.
Available
on
the
Internet
at:
http://
www.
epa.
gov/
seahome/
well/
src/
title.
htm
,
accessed
November
27,
2002.

Ray,
C.,
T.
Grischek,
J.
Schubert,
J.
Wang,
and
T.
Speth.
2002.
A
perspective
of
riverbank
filtration,
J.
AWWA,
94(
4):
149­
160.

Ray,
C.
2001a.
Riverbank
filtration:
an
analysis
of
parameters
for
optimal
performance.
Proceedings
of
the
Annual
Conference
of
the
American
Water
Works
Association.
June
17­
21,
2001.
Washington,
DC.

Ray,
C.
2001b.
Modeling
riverbank
filtration
systems
to
attenuate
shock
loads
in
rivers.
Proceedings
of
the
Annual
Conference
of
the
American
Water
Works
Association.
June
17­
21,
2001.
Washington,
DC.

Reynolds,
J.
M.,
1997.
An
Introduction
to
Applied
and
Environmental
Geophysics.
New
York:
Wiley.

Ritter,
D.
F.,
C.
R.
Kochel,
and
J.
R.
Miller.
1995.
Process
Geomorphology.
3rd
edition.
Dubuque,
Iowa:
Wm.
C.
Brown
Publishers.

Rubin
Y.
and
S.
Hubbard,
2003.
Hydrogeophysics.
Kluwer.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
38
Sanchez
de
Lozada,
D.,
P.
Vandevivere,
P.
Baveye,
and
S.
Zinder.
1994.
Decrease
of
the
hydraulic
conductivity
of
sand
columns
by
Methanosarcina
barkeri.
World
Journal
of
Microbiology
and
Biotechnology
10:
325­
333.

Schafer,
D.
2000.
Groundwater
modeling
in
support
of
riverbank
infiltration
for
Louisville
Water
Company,
in
Proceedings
of
International
Riverbank
Filtration
Conference,
Nov.
2­
4,
Duesseldorf,
W.
Julich
and
J.
Schubert
(
eds.),
International
Arbeitgemeinschaft
der
Wasserwerke
im
Rheineinzugsgebiet,
Amsterdam.

Schijven,
J.,
Berger,
P.,
Miettenen,
I.
2003.
Removal
of
viruses,
bacteria,
protozoa
and
toxins
using
bank
filtration,
in
Riverbank
Filtration:
Improving
Source
Water
Quality,
edited
by
Ray,
C.,
Melin,
G.
and
Linsky,
R.,
Kluwer,
Dordrecht,

State
Coordinating
Committee
on
Ground
Water
(
SCCGW).
2000.
state
of
Ohio
Technical
Guidance
for
Well
Construction
and
Ground
Water
Protection.
Available
on
the
Internet
at:
http://
www.
dnr.
state.
oh.
us/
water/
pubs/
pdfs/
wellconguide.
pdf
,
accessed
November
27,
2002.

Symons,
J.
M.,
L.
C.
Bradley,
Jr.,
T.
C.
Cleveland,
eds.
2000.
The
Drinking
Water
Dictionary.
American
Water
Works
Association,
Denver,
CO.

Tufenkji,
N.,
J.
N.
Ryan,
and
M.
Elimelech.
2002.
The
promise
of
bank
filtration:
a
simple
technology
may
inexpensively
clean
up
poor­
quality
raw
surface
water.
Environmental
Science
and
Technology.
36:
422A
­
428A.

United
States
Environmental
Protection
Agency
(
USEPA).
2001.
Long
Term
2
Enhanced
Surface
Water
Treatment
Rule
(
LT2ESWTR).
Pre­
proposal
draft
for
stakeholder
review.
November
27,
2001.

USEPA.
1975.
Manual
of
Water
Well
Construction
Practices.
Office
of
Water
Supply.
EPA/
570/
9­
75­
001.
Washington,
D.
C.
156
pp.

United
States
Geological
Survey
(
USGS).
1998.
The
National
Atlas
of
the
United
States
of
America.
Principal
Aquifers.
[
Map].
Reston,
VA:
U.
S.
Department
of
the
Interior,
U.
S.
Geological
Survey.

Vandevivere,
P.,
P.
Baveye,
D.
Sanchez
de
Lozada,
and
P.
DeLeo.
1995.
Microbial
clogging
of
saturated
soils
and
aquifer
materials:
Evaluation
of
mathematical
models.
Water
Resources
Research
31(
9):
2173­
2180.

Wang,
J.
Z.,
S.
A.
Hubbs,
and
M.
Unthank.
2001.
Factors
Impacting
the
Yield
of
Riverbank
Filtration
Systems.
Presentation
at
the
American
Water
Works
Association
Water
Quality
Technology
Conference,
November
11­
15,
Nashville,
TN.
Chapter
4
­
Bank
Filtration
LT2ESWTR
Toolbox
Guidance
Manual
Proposal
Draft
June
2003
4­
39
Wang,
J.
Z.,
R.
Song,
and
S.
A.
Hubbs.
2000.
Particle
removal
through
riverbank
filtration
process,
in
Proceedings
of
the
International
Riverbank
Filtration
Conference,
Nov.
2­
4,
Duesseldorf,
W.
Julich
and
J.
Schubert
(
eds.),
International
Arbeitgemeinschaft
der
Wasserwerke
im
Rheineinzugsgebiet,
Amsterdam.

Wang,
J.
Z.,
Hubbs,
S.
A.
and
Song,
R.
2002.
Evaluation
of
Riverbank
Filtration
as
a
Drinking
Water
Treatment
Process,
American
Water
Works
Association
Research
Foundation
Report
90922,
145
p.

Warner,
J.
W.,
T.
K.
Gates,
R.,
Namvargolian,
P.
Miller,
and
G.
Comes.
1994.
Sediment
and
microbial
fouling
of
experimental
groundwater
recharge
trenches.
Journal
of
Contaminant
Hydrology
15:
321­
344.

Weiss,
W.
J.,
E.
J.
Bouwer,
W.
P.
Ball,
C.
R.
O'Melia,
H.
Arora,
T.
F.
Speth.
2003.
Reduction
in
DBP
precursors
and
pathogens
during
riverbank
filtration
at
three
midwestern
drinking
water
utilities,
in
Riverbank
Filtration
for
Water
Supply,
C.
Ray
and
R.
Linsky
(
eds).
Kluwer
Academic
Publishers.

Williams,
G.
P.,
and
Wolman,
M.
G.
1984.
Downstream
Effects
of
Dams
on
Alluvial
Rivers.
U.
S.
Geological
Survey
Professional
Paper
1286.
Washington:
U.
S.
Government
Printing
Office.
