Coastal-erosion processes and assessments of setback distances


Komar, P.D.; Marra, J.J.; Allan, J.C.

Solutions to Coastal Disasters '02: 808-822

2002


Methodologies have been developed to establish setback distances for use on the coast of the Pacific Northwest where the principal hazards are wave runup and surge during major storms, and El Ninos that produce unusually high tides and the northward movement of beach sand within littoral cells, creating "hot spot" erosion sites. Of concern is the discovery that the severity of storms and their generated waves have been increasing for at least the past 25 to 50 years, with the underlying cause and possible connection to global-climate change not being fully understood. Important are extreme water levels that result from combinations of high predicted astronomical tides, processes related to El Ninos that elevate tides still further, and the occurrence of a major storm that produces high levels of wave runup on beaches. Having made such an assessment, a geometric model is employed to evaluate the potentially maximum horizontal retreat of foredunes, thereby serving as a guide for the placement of setbacks in dune areas. In areas where the beach is backed by sea cliffs, the analysis yields a determination of the average number of hours per year the cliff is impacted by the high waters of tides plus wave runup, providing an assessment of the susceptibility of the property to wave attack, one component in the evaluation of an appropriate setback distance.

COASTAL-EROSION
PROCESSES
AND
ASSESSMENTS
OF
SETBACK
DISTANCES
Paul
D.
Komar',
John
J.
Marra
2
,
and
Jonathan
C.
Allan
3
Abstract:
Methodologies
have
been
developed
to
establish
setback
distances
for
use
on
the
coast
of
the
Pacific
Northwest
where
the
principal
hazards
are
wave
runup
and
surge
during
major
storms,
and
El
Nitlos
that
produce
unusually
high
tides
and
the
northward
movement
of
beach
sand
within
littoral
cells,
creating
"hot
spot"
erosion
sites.
Of
concern
is
the
discovery
that
the
severity
of
storms
and
their
generated
waves
have
been
increasing
for
at
least
the
past
25
to
50
years,
with
the
underlying
cause
and
possible
connection
to
global-climate
change
not
being
fully
understood.
Important
are
extreme
water
levels
that
result
from
combinations
of
high
predicted
astronomical
tides,
processes
related
to
El
Nitios
that
elevate
tides
still
further,
and
the
occurrence
of
a
major
storm
that
produces
high
levels
of
wave
runup
on
beaches.
Having
made
such
an
assessment,
a
geometric
model
is
employed
to
evaluate
the
potentially
maximum
horizontal
retreat
of
foredunes,
thereby
serving
as
a
guide
for
the
placement
of
setbacks
in
dune
areas.
In
areas
where
the
beach
is
backed
by
sea
cliffs,
the
analysis
yields
a
determination
of
the
average
number
of
hours
per
year
the
cliff
is
impacted
by
the
high
waters
of
tides
plus
wave
runup,
providing
an
assessment
of
the
susceptibility
of
the
property
to
wave
attack,
one
component
in
the
evaluation
of
an
appropriate
setback
distance.
'College
of
Oceanic
&
Atmospheric
Sciences,
Oregon
State
Univ.,
Corvallis,
OR
97331
USA,
pkomar@oce.orst.edu
2
Shoreland
Solutions,
P.O.
Box
1046,
Newport,
OR
97365
USA,
marrajj@harborside.com
3
Oregon
Dept.
of
Geology
and
Mineral
Industries,
Coastal
Field
Office,
313
2nd
St.,
Suite
D,
Newport,
OR
97365
USA,
jonathan.allan@dogami.state.or.us
808
Columbia
I
River
NDBC
46050
Yaquina
Bay
Newport
OREGON
Coos
Bay
NDBC
46002
Oregon
132.
130•
128.
Cape
Blanco
126
1
124°
122°
NDBC
46041
Cape
Elizabeth.
CDIP
03601
Grays
Harbour
NDBC
46029
Columbia
River
Bar
Neal)
Bay
Strait
of
Juan
de
Fuca
WASHINGTON
Toke
Point
Astoria
NDBC
wave
buoy
NOS
Tide
gauge
0
50
100 150
Km
NDBC
46005
Washington
COASTAL
DISASTERS
'02
809
INTRODUCTION
Many
coastal
communities
have
found
it
useful
to
establish
coastal
hazard
zones
or
setbacks
adjacent
to
beaches
as
part
of
their
management
efforts
to
protect
developments
from
recognized
natural
hazards.
Examples
of
setbacks
applied
along
the
shores
of
the
United
States
have
been
summarized
by
Wood
et
al.
(1990).
In
most
cases
this
involves
a
direct
transformation
of
measured
long-term
recession
rates
into
setback
distances.
That
approach
is
not
particularly
suited
to
the
coast
of
the
Pacific
Northwest
(PNW),
the
states
of
Washington
and
Oregon
(Figure
1),
in
that
long-term
net
recession
rates
are
generally
small
to
negligible,
or
accretion
has
prevailed
[e.g.,
in
the
Columbia
River
Littoral
Cell
on
the
Washington
coast
(Kaminsky
et
al.,
1999)].
Far
more
important
have
been
episodic
occurrences
of
erosion,
most
dramatic
in
foredunes
that
may
be
cut
back
by
lOs
of
meters
during
the
few
days
of
a
storm,
but
with
the
dunes
naturally
rebuilding
in
the
following
decades
(Komar,
1997).
Generally
the
re-formed
dunes
eventually
extend
out
to
their
former
extent,
and
unfortunately
in
some
locations
homes
have
been
constructed
in
areas
that
had
experienced
earlier
cycles
of
erosion
and
accretion.
Fig.
1.
The
U.S.
Pacific
Northwest,
with
locations
of
NOAA
wave
buoys
and
tide
gauges.
This
paper
summarizes
the
methodologies
that
have
been
developed
for
the
establishment
of
setbacks
along
the
PNW
coast.
This
has
required
careful
consideration
due
to
the
high-energy,
erosive
character
of
this
coast.
The
most
susceptible
areas
consist
of
dune-covered
sand
spits
that
separate
bays
or
estuaries
from
the
ocean,
or
other
low-lying
areas
where
dune
sands
have
accumulated.
Much
of
the
coast
consists
of
sea
cliffs
backing
beaches
that
offer
varying
degrees
of
buffer
protection
from
wave
attack.
In
order
to
establish
rationally-based
setback
lines,
the
assessments
have
had
to
focus
on
the
underlying
process
of
erosion.
This
has
included
quantitative
evaluations
.48
4
44.
42°
810
COASTAL
DISASTERS
'02
of
runup
elevations
due
to
extreme
storm
waves,
the
addition
of
the
runup
to
elevated
measured
tides,
followed
by
a
comparison
between
the
resulting
total
water
level
and
the
elevation
of
the
edge
of
the
foredune
or
base
of
a
sea
cliff.
The
development
of
useful
models
based
on
an
understanding
of
nearshore
processes
has
required
many
years
of
research,
followed
by
a
period
during
which
we
have
gained
experience
in
the
application
of
the
techniques
to
specific
areas
of
the
PNW
coast.
While
the
focus
of
this
paper
is
on
the
PNW,
the
methodologies
can
be
adapted
to
other
coasts,
wherever
it
is
required
that
setbacks
be
rationally
based
on
a
scientific
understanding
of
the
erosion
processes.
PNW
PROCESSES
AND
COASTAL
EROSION
The
processes
important
to
erosion
along
the
PNW
coast
are
reasonably
well
understood
due
to
investigations
of
a
number
of
occurrences
over
the
years
(Komar,
1997).
Furthermore,
the
basic
processes
have
been
well
documented,
including
the
wave
climate
and
occurrences
of
extreme
storm-wave
conditions,
and
water-level
variations
due
to
tides
and
phenomena
related
to
El
Nitlos
that
can
raise
water
levels
by
lOs
of
centimeters
above
predicted
tidal
elevations.
Of
significance,
this
research
has
documented
that
there
has
been
an
increase
in
wave
heights
spanning
at
least
the
past
25
to
50
years,
with
parallel
increases
in
the
wave-related
nearshore
processes
(Allan
and
Komar,
2000,
2001).
Accounting
for
such
long-term
trends
complicates
the
determination
of
setback
distances.
The
PNW
is
noted
for
the
severity
of
its
wave
climate,
with
major
winter
storms
generating
waves
having
deep-water
significant
wave
heights
of
10 to
15
m.
The
available
wave
data
from
offshore
buoys,
Figure
1,
were
analyzed
by
Tillotson
and
Komar
(1997)
using
conventional
procedures
to
establish
the
wave
climate,
but
since
that
compilation
there
has
been
a
series
of
exceptionally
severe
storms,
well
above
previous
experience.
Based
on
wave
measurements
collected
through
1996,
Tillotson
and
Komar
(1997)
and
Ruggiero
et
al.
(1996)
projected
that
the
100-year
storm
would
generate
a
deep-water
significant
wave
height
of
approximately
10
m.
During
the
winter
of
1997-98,
one
storm
produced
waves
that
reached
or
exceeded
this
projected
value,
while
the
1998-99
winter
saw
four
additional
storm-wave
events
exceeding
10
m,
including
the
most
severe
storm
on
2-4
March
1999
that
generated
deep-water
significant
wave
heights
of
14
m.
A
sixth
100-year
storm
occurred
during
the
following
winter
(Allan
and
Komar,
in
press).
Including
these
recent
storms,
the
projected
100-year
deep-water
significant
wave
height
is
now
estimated
to
be
15.0
m.
This
seemingly
abrupt
increase
in
the
recent
storm
severity
and
generated
waves
induced
us
to
re-examine
the
wave
climate
of
the
PNW,
a
study
that
was
then
expanded
to
cover
the
entire
eastern
North
Pacific
(Allan
and
Komar,
2000,
in
review).
This
was
accomplished
through
analyses
of
wave
data
collected
during
the
past
25
years
by
six
deep-water
buoys
extending
from
the
Gulf
of
Alaska
in
the
north
to
Pt.
Arguello
in
south-central
California.
It
was
found
that
there
have
been
progressive
increases
in
wave
heights
and
periods
at
mid-latitudes,
reaching
a
maximum
rate
of
increase
for
wave
heights
off the
Washington
coast,
with
only
a
slightly
lower
rate
of
increase
off
the
Oregon
coast.
The
increase
was
still
smaller
offshore
from
northern
California,
while
buoys
located
off
the
coast
of
south-central
California
and
in
the
Gulf
of
Alaska
showed
no
evidence
for
an
increase
in
wave
heights.
The
findings
of
Allan
and
Komar
(2000)
of
increasing
wave
heights
at
mid
latitudes
in
the
North
Pacific
have
been
supported
by
the
study
of
Graham
and
Diaz
(2001)
of
the
storm
systems,
demonstrating
that
the
frequencies
and
intensities
of
extreme
storms
have
increased
progressively
since
1948.
1980
1984
1988
1992
1996
2000
Year
Washing
ton
(#46005)
188
32x
-
60
Y
=
0
0
R
Z
=
0.48
4.0
3.5
3.0
2.5
2.0
1976
COASTAL
DISASTERS
'02
811
Figure
2
is
a
graph
of
the
annual
averages
of
the
"winter"
(October
through
March)
significant
wave
heights
measured
by
buoy
#46005
off
the
coast
of
Washington.
The
regression
line
has
a
slope
0.032
m/yr,
the
rate
of
increase
in
the
average
winter
waves
during
the
past
25
years,
representing
a
0.80-m
increase
during
the
record
of
measurements
[these
values
are
slightly
less
than
found
by
Allan
and
Komar
(2000),
produced
by
four
years
of
additional
measurements
when
wave
conditions
were
lower].
The
same
type
of
analysis
but
for
the
average
maximum
measured
waves
of
the
six
"winter"
months
yields
a
regression
slope
of
0.084
m/yr
with
R
2
=
0.36,
a
2.1-m
increase
in
25
years.
This
decadal
increase
of
the
more
extreme
waves
generated
by
winter
storms
has
the
greatest
relevance
to
coastal
erosion
occurrences.
Fig.
2.
Decadal
increase
in
the
"winter"
deep-water
significant
wave
heights
measured
by
buoy
#46005
off
the
Washington
coast.
In
terms
of
the
erosion
of
properties
backing
beaches,
one
of
the
most
important
processes
is
swash
runup
on
the
beach
once
the
deep-water
waves
reach
the
nearshore
(Komar,
1998a).
In
recognition
of
this
importance,
we
undertook
measurements
of
runup
elevations
on
Oregon
beaches
under
a
range
of
conditions,
relating
the
elevations
to
the
deep-water
wave
heights
and
periods,
and
to
the
beach
slope
(Ruggiero
et
al.,
1996,
2001).
When
our
Oregon
data
were
combined
with
those
of
Holman
(1986)
from
the
Field
Research
Facility,
Duck,
North
Carolina,
the
results
established
the
correlation
R
2%
=
O.
27(SH
L
o
.)
1/2
=
0.11e
2
S
1
/
2
1/f
2
T
(1)
where
S
is
the
beach
slope,
H..
and
L..
are
the
deep-water
wave
height
and
wave
length,
and
T
is
the
wave
period.
The
calculated
runup,
12
29
,
is
the
2%
exceedence
elevation,
that
is,
only
2%
of
the
measured
runup
values
during
35
minutes
of
data
collection
exceeded
this
level.
This
is
the
vertical
component
of
the
runup,
and
includes
the
wave
setup
as
well
as
the
swash
runup
elevation.
Equation
(1)
has
been
shown
to
yield
good
results
in
applications
to
specific
sites
on
the
Oregon
coast
under
an
appreciable
range
of
beach
slopes
(Komar
et
al.,
2001).
In
that
both
the
deep-water
wave
heights
and
periods
have
increased
during
the
past
25
to
50
years,
it
follows
that
R.
2%
must
have
increased
as
well.
This
is
shown
in
S
ig
n
i
fican
t
wav
e
he
ig
ht,
H,
(n
-i
)
Washington
(046005)
1
80
1
40
Y
x
0.010x
-
17.744
R
2
-0.49
1
20
1976
1980
1984
1988
1992
1
896
z000
Year
812
COASTAL
DISASTERS
'02
Figure
3
in
terms
of
the
annual
winter
(October
through
March)
averages
of
runup
levels,
calculated
with
equation
(1)
for
a
beach
slope
S
=
0.04,
representative
of
sandy
beaches
on
the
PNW
coast.
The
significance
of
Figure
3
is
that
in
response
to
the
long-term
increases
in
deep-water
wave
heights
and
periods,
there
also
has
been
a
progressive
increase
in
runup
levels
of
winter
waves.
The
regression
line
is
statistically
significant
with
R
2
=
0.49,
and
the
slope
of
increasing
is
0.010
m/yr,
amounting
to
a
0.25-m
vertical
increase
during
a
25-year
time
span.
The
corresponding
migration
of
the
shoreline
on
PNW
beaches
with
a
slope
of
0.04
would
have
been
over
6
m,
if
"shoreline"
is
defined
in
terms
of
the
level
reached
by
the
runup.
This
degree
of
landward
shift
in
the
extent
of
the
average
wave
runup
during
the
winter
clearly
could
be
an
important
factor
in
the
progressive
erosion
of
the
coast.
Much
of
the
associated
property
erosion
would
of
course
be
produced
by
the
maximum
values
of
R,
achieved
each
year
during
the
most
extreme
events
those
values
are
examined
later.
Fig.
3.
Decadal
increase
in
the
runup
calculated
using
eq.
(1)
with
deep-
water
wave
data
and
for
a
beach
slope
S
=
0.04.
It
is
important
to
understand
the
global-climate
controls
that
have
produced
the
increases
in
storm
intensities
and
wave
conditions
in
the
North
Pacific.
Without
such
an
understanding,
it
is
not
possible
to
project
them
into
the
future
with
confidence,
necessary
if
they
are
to
be
included
in
assessments
of
setbacks.
Allan
and
Komar
(2000,
in
review)
attempted
to
relate
the
decadal-long
trends
of
increasing
wave
heights
to
various
climate
indices,
including
the
East
Pacific
Teleconnection
Index
(EPI),
the
difference
in
atmospheric
pressures
between
the
Aleutian
Low
and
Hawaiian
High,
expected
to
be
important
to
the
strengths
of
storms
and
sizes
of
generated
waves.
Unfortunately,
no
convincing
correlation
could
be
found
that
accounts
for
the
long-term
trends.
Graham
and
Diaz
(2001)
suggested
that
increasing
sea-surface
temperatures
in
the
western
tropical
Pacific
are
a
likely
cause
of
the
increasing
storm
intensities
and
wave
heights,
thereby
possibly
connecting
them
to
global
warming.
However,
this
remains
to
be
demonstrated
by
additional
research.
Allan
and
Komar
(2000)
found
that
while
the
EPI
and
other
climate
indices
did
not
explain
the
long-term
progressive
increase
in
wave
conditions,
the
indices
could
account
in
part
for
the
numbers
of
storms
and
the
generated
wave
heights
either
above
or
below
the
long-term
trend,
the
variability
from
year
to
year.
It
was
shown
that
this
variability
in
the
PNW
depends
primarily
on
the
EPI,
while
at
the
lower
latitudes
of
California
it
depends
on
the
range
of
climate
conditions
between
El
Nifios
and
La
Nifias
as
measured
by
a
modified
Southern
Oscillation
Index
(MEI).
Seymour
(1996,
COASTAL
DISASTERS
'02
813
1998)
demonstrated
that
wave
heights
increase
along
the
coast
of
southern
California
during
an
El
Nifio,
attributing
this
to
the
more
southerly
paths
of
the
storms
so
they
cross
the
California
coast,
contrasting
with
normal
or
La
Nifia
years
when
the
storms
pass
more
directly
over
the
PNW.
He
further
suggested
that
wave
heights
along
the
PNW
coast
would
decrease
during
an
El
Nitio
due
to
the
more
southerly
tracks
of
the
storms.
However,
while
it
is
true
that
the
storms
do
tend
to
cross
the
California
coast
during
El
Nifios,
Allan
and
Komar
(2000,
in
review)
found
that
there
is
little
actual
decrease
in
wave
heights
along
the
PNW
coast
associated
with
this
southward
displacement.
A
more
important
consequence
of
this
southward
displacement
of
storms
during
an
El
Nifio
is
that
their
generated
waves
reach
the
PNW
coast
from
a
more
southwesterly
quadrant.
The
Oregon
coast
consists
of
a
series
of
littoral
cells,
in
effect
"pocket
beach"
stretches
of
shore
isolated
by
large
headlands
that
extend
well
offshore.
During
El
Nifio
winters
there
is
an
abnormally
large
northward
transport
of
sand
on
beaches
within
the
littoral
cells.
This
was
first
documented
during
the
1982-
83
El
Niiio
when
it
was
observed
that
the
movement
resulted
in
large
quantities
of
sand
accumulation
to
the
south
sides
of
headlands,
with
extreme
erosion
on
their
north
sides,
and
that
inlets
without
jetties
were
pushed
toward
the
north
(Komar,
1986).
The
same
patterns
were
observed
during
the
1997-98
El
Nifio
(Komar,
1998b),
leading
to
the
identification
of
"hot
spot"
erosion
sites
directly
related
to
El
Nitio
occurrences,
principally
erosion
at
the
south
ends
of
littoral
cells
and
to
the
north
of
inlets
that
are
free
to
migrate.
Where
jetties
have
been
constructed
to
control
inlets,
to
a
degree
they
have
acted
as
small
headlands,
similarly
producing
erosion
to
their
immediate
north;
the
primary
example
of
this
occurred
at
Ocean
Shores,
Washington,
north
of
the
Grays
Harbor
jetties
(Kaminsky
et
al.,
1999).
Direct
measurements
of
the
northward
displacement
of
sand
within
the
Netarts
Littoral
Cell
on
the
Oregon
coast
during
the
1997-98
El
Nino,
and
the
resulting
erosion
impacts,
have
been
made
possible
from
LIDAR
surveys
analyzed
by
Revell
et
al.
(in
press).
Another
El
Nitio-related
effect
important
to
West
Coast
erosion
is
the
increased
water
levels
that
enhance
the
elevations
of
measured
tides.
Predicted
high
Spring
tides
on
the
PNW
coast
range
up
to
about
3.5
m
MLLW
(2.3
m
NGVD29),
but
extreme
measured
tides
can
be
10s
of
centimeters
higher
than
predicted
(Komar
et
al.,
2000;
Ruggiero
et
al.,
2001;
Allan
and
Komar,
in
press).
This
is
especially
true
during
the
winter
when
monthly-averaged
sea
levels
reach
their
highest
elevations.
Upwelling
during
the
summer
produces
colder
offshore
water
than
in
the
winter,
depressing
mean
sea
levels
during
the
summer,
raising
them
in
the
winter
due
to
the
thermal
expansion
of
the
warmer
water.
In
addition,
the
geostrophic
effects
of
the
southward-
flowing
currents
in
the
summer
further
lower
sea
levels
along
the
shore,
while
the
northward
flows
of
the
winter
raise
water
levels.
As
a
result,
winter
monthly-averaged
sea
levels
are
on
average
about
30
cm
higher
than
in
the
summer.
These
processes
are
magnified
during
El
Nitios,
with
still
warmer
offshore
water
during
the
winter
and
stronger
northward-flowing
currents,
combining
to
elevate
water
levels
along
the
coast
well
above
normal
winters.
The
historically
most
extreme
examples
of
elevated
water
levels
from
these
various
processes
occurred
during
the
strong
1982-83
and
1997-98
El
Nifios
when
monthly
mean-water
levels
during
those
winters
were
50
to
60
cm
higher
than
in
the
preceding
summer,
20
to
30
cm
higher
than
the
long-term
average,
increasing
the
measured
tides
by
some
70
cm
above
predicted
levels
(Komar
et
al.,
2000).
This
raises
the
elevations
of
the
tides
at
all
stages,
not
just
the
high
tides,
tending
to
flood
out
PNW
beaches
during
El
Nifios.
814
COASTAL
DISASTERS
'02
Storm
surges
can
also
elevate
measured
tides
above
predicted
levels,
although
surges
on
the
PNW
coast
during
even
the
most
extreme
storms
are
substantially
lower
than
occur
during
hurricanes
as
experienced
on
the
East
and
Gulf
Coasts
of
the
USA.
During
typical
PNW
storms
the
surge
elevates
the
tide
by
some
20
to
30
cm,
while
a
severe
storm
like
that
on
2-4
March
1999
produced
a
storm
surge
of
1.6
m
above
the
predicted
tide
(Allan
and
Komar,
in
press).
COMBINED
PROCESSES,
EXTREME
EVENTS,
AND
SETBACK
DISTANCES
Models
have
been
developed
to
link
the
processes
discussed
above
that
are
responsible
for
foredune
and
sea-cliff
erosion,
and
then
to
assess
the
extent
of
the
potential
erosion
in
order
to
provide
guidance
in
the
establishment
of
setbacks.
While
the
models
were
developed
for
use
on
the
PNW
coast
and
the
example
application
presented
here
is
for
that
area,
they
are
general
and
could
be
adapted
for
application
on
other
coasts.
The
first
model
used
in
an
analysis,
depicted
in
Figure
4,
is
that
developed
by
Ruggiero
et
al.
(1996, 2001).
In
essence
it
involves
a
summation
of
the
processes
that
determine
the
total
water
level
at
the
shore,
the
sum
of
the
predicted
tide,
the
several
processes
discussed
above
that
elevate
measured
tides
above
predicted
levels,
and
finally
the
addition
of
the
wave
runup.
The
water
levels
of
these
individual
processes,
and
thus
the
total
water
elevation
resulting
from
their
addition,
vary
from
hour
to
hour
during
a
storm,
change
with
the
seasons,
and
are
affected
by
climate
events
such
as
El
Nifios.
Ultimately,
as
depicted
in
Figure
4,
of
importance
are
occurrences
when
the
addition
of
the
elevation
of
the
measured
tide
(E
T
)
and
wave
runup
(R
2%
)
exceeds
the
elevation
of
the
beach-dune
junction
(E,)
or
the
toe
of
the
sea
cliff,
erosion
only
being
possible
when
E
T
±
R2%
>
E.
On
the
coast
of
the
PNW,
as
shown
by
analyses
of
several
major
storm
events
(Allan
and
Komar,
in
press),
this
condition
for
erosion
is
typically
achieved
during
only
a
few
hours
in
each
storm,
controlled
in
large
part
by
the
cycle
of
the
tides.
Foredune
Erosion
Model
foredune
dune
erosion
occurs
when
E
r
+
R
>
E,,
beach-dune
junction
wave
swash
R=w
a
runup
measured
tide
level
predicted
tide
E
r
=
measured
tide
NGVD
"sea
level'
Fig.
4.
Model
of
Ruggiero
et
al.
(1996,
2001)
for
an
evaluation
of
the
total
water
level
due
to
the
summation
of
the
measured
tides
and
wave
runup.
Having
completed
the
analysis
of
Figure
4,
the
question
is
how
much
foredune
or
sea-cliff
erosion
might
be
expected
during
a
major
storm
in
the
hours
when
COASTAL
DISASTERS
'02
815
E
T
+
R
IB
,
>
El?
In
the
case
of
foredune
erosion,
we
have
developed
a
simple
geometric
model,
shown
in
Figure
5,
to
assess
the
potentially
maximum
dune
retreat
(DE
max
)
produced
by
a
total
water
level
WL
=
E
T
+
R
N
,
(Komar
et
al.,
1999).
On
the
high-energy
PNW
coast,
the
beach
face
dominated
by
wave
swash
is
wide
and
has
a
nearly
uniform
slope
angle,
13
(Fig.
5).
The
model
assumes
that
this
slope
is
maintained
as
the
dunes
are
eroded
back,
so
the
analysis
focuses
on
the
right
triangle
depicted
in
Figure
5
where
erosion
due
to
the
high
water
level
alone
cuts
back
the
foredune
to
point
B
where
the
projected
sloping
beach
intersects
the
elevated
water
level.
Additional
dune
retreat
could
result
from
the
lowering
of
the
beach
face
due
to
general
beach
erosion
during
the
storm,
or
locally
produced
by
the
presence
of
a
rip
current
that
cuts
an
embayment
into
the
beach.
This
vertical
shift
in
the
profile
is
represented
by
the
beach-level
change
ABL,
which
results
in
a
further
retreat
of
the
dunes
to
point
C.
The
total
retreat
is
then
given
by
the
line
segment
iTC
,
which
is
taken
as
equivalent
to
DE„,
to
represent
the
"maximum
dune
erosion".
From
the
right
triangle
formed
by
this
erosion
and
projection
of
the
beachface,
Figure
5,
we
have,
(WL—
E
.
,)+
ABL
DE.
a
.
(2)
where
S
=
tan/3
is
the
slope
of
the
beach
within
the
swash
zone
fronting
the
dunes.
GEOMETRIC
MODEL
OF
FOREDUNE
EROSION
A
minims
wafer
level
WL
'At
-E,
OAL
E,)
ABL
=
1.0
E
a
:lune-tem
eLPArabc."
anaee
Fig.
5.
Geometric
dune-erosion
model
of
Komar
et
al.
(1999).
This
geometric
model
used
to
assess
potential
foredune
erosion
during
storms
is
similar
in
concept
to
the
well
known
model
of
Bruun
(1962)
where
the
erosive
process
is
the
long-term
rise
in
sea
level.
Closer
in
application
are
models
that
have
been
developed
in
the
Netherlands
to
evaluate
dune
erosion
when
the
water
is
temporarily
raised
by
a
storm
surge
(Edelman,
1972;
Vellinga,
1982).
These
models
are
"geometric"
in
that
they
all
involve
the
upward
and
landward
shift
of
a
triangle,
one
leg
of
which
corresponds
to
the
elevated
water
level.
The
geometric
model
we
have
developed
is
much
the
same,
differing
in
that
the
total
water
level
is
determined
by
extreme
high
tides
plus
the
runup
of
waves
during
a
storm,
the
factors
depicted
in
Figure
4.
Our
dune-erosion
model
and
the
evaluation
of
DE„,
ax
with
equation
(2)
tends
to
represent
the
maximum
potential
dune
retreat
for
the
water
level
WL,
in
that
the
model
816
COASTAL
DISASTERS
'02
does
not
account
for
the
fact
that
the
water
may
only
reach
that
level
for
a
couple
of
hours,
and
the
erosional
response
will
lag
behind
the
causative
processes.
We
have
attempted
to
assess
this
lag
through
application
of
beach
profile
and
dune
erosion
process-based
models,
specifically
SBEACH
(Larson
and
Kraus,
1989),
EBEACH
(Kriebel
and
Dean,
1985)
and
COSMOS
(Nairn
and
Southgate,
1993).
We
found
that
these
process-based
models
are
inadequate
in
applications
on
the
PNW
coast
due
to
their
having
been
calibrated
on
much
lower
energy
beaches
(or
in
laboratory
wave
tanks),
and
in
particular
because
processes
important
to
erosion
on
the
PNW
coast
are
not
included
(e.g.,
infragravity
surf
motions
and
rip
currents).
It
was
found
that
the
process-based
models
not
only
predict
less
dune
retreat
during
a
storm
than
the
geometric
model,
they
also
under
predicted
the
actual
extent
of
dune
erosion
during
major
storms
due
to
having
left
out
important
processes.
We
have
suggested
that
in
applications
both
the
geometric
model
and
the
process-based
models
be
used,
the
former
over-predicting
the
extent
of
dune
retreat
while
the
latter
under-predict
the
erosion,
thereby
bracketing
the
actual
extent
of
erosion
during
a
major
storm
(Komar
et
al.,
1999).
However,
in
management
applications
to
establish
setbacks
this
generally
has
not
been
done,
in
part
due
to
the
difficulty
of
routinely
using
the
process-based
models.
Instead,
equation
(2)
from
the
geometric
model
has
been
used
alone
to
guide
the
placement
of
setbacks,
recognizing
that
the
evaluated
DE„,
represents
the
maximum
likely
erosion,
and
accordingly
additional
uncertainty
factors
are
not
generally
needed
in
the
assessment.
The
direct
use
of
the
geometric
model
to
assess
the
potential
extent
of
dune
erosion
and
to
establish
setbacks
has
been
supported
by
tests
under
the
extreme
storm
conditions
experienced
on
the
PNW
coast
in
recent
years.
As
discussed
in
the
preceding
sections,
the
winter
of
1997-98
was
a
strong
El
Nino,
at
which
time
monthly
mean
water
levels
were
elevated
so
that
measured
tides
were
substantially
higher
than
predicted.
Of
some
surprise
was
the
occurrence
of
one
major
storm
when
the
deep-
water
significant
wave
height
exceeded
10
m,
the
value
we
had
projected
to
be
the
100-
yr
occurrence;
this
was
a
surprise
in
that
following
the
suggestion
by
Seymour
(1996),
we
had
thought
that
an
El
Nitio
would
produce
lower
wave
heights
in
the
PNW
due
to
the
more
southerly
tracks
of
the
storms.
The
following
winter,
1998-99,
a
medium
La
Nitia,
saw
the
occurrence
of
four
additional
storms
when
deep-water
wave
heights
exceeded
the
100-year
event.
In
that
the
beaches
had
not
recovered
from
the
El
Nirio
erosion
of
the
preceding
winter,
the
two
years
and
respective
climate
events
represented
something
of
a
"one-two
punch"
of
erosional
impacts.
These
two
winters
of
unusual
storms
and
erosion
provided
us
with
the
opportunity
to
test
the
methodologies
and
models
developed
to
assess
potential
occurrences
of
foredune
erosion.
The
results
are
presented
in
a
report
(Komar
and
Allan,
2000)
and
in
a
pair
of
published
papers
(Komar
et
al.,
2000;
Allan
and
Komar,
in
press).
Allan
and
Komar
(in
press)
focused
on
developing
detailed
analyses
of
the
hourly
changes
in
measured
tides,
storm-generated
wave
heights
and
periods,
and
of
calculated
/?
2
,
runup
levels
for
each
of
the
major
storms,
and
then
applied
model
of
Ruggiero
et
al.
(1996,
2001),
Figure
4,
to
determine
the
hourly
variations
in
total
water
levels.
Before
and
after
beach
and
dune
profiles
were
obtained
at
a
number
of
sites
along
the
PNW
coast,
documenting
the
resulting
extent
of
the
cumulative
erosional
response
during
the
winters
of
1997-98
and
1998-99.
Confirmation
of
the
calculated
total
water
levels,
WL,
resulting
from
the
combined
processes
was
provided
by
general
agreement
with
the
surveyed
elevations
of
the
eroded
dune
scarps.
This
also
provided
partial
confirmation
of
the
geometric
dune-erosion
model
in
that
the
basic
assumption
in
its
derivation
is
that
the
total
water
level,
WL,
controls
the
elevation
at
which
the
dunes
are
cut
back
(i.e.,
the
positions
of
points
B
or
C
shown
in
Figure
5).
However,
it
was
COASTAL
DISASTERS
'02
817
found
that
the
surveyed
horizontal
retreat
of
the
dunes
was
less
than
the
calculated
DE,„.
(as
we
had
expected
would
be
the
case).
On
the
other
hand,
under
the
"one-
two
punch"
of
the
successive
winters,
with
the
last
storm
in
the
series,
that
on
2-4
March
1999,
having
been
the
largest
and
yielding
the
highest
total
water
levels
at
most
coastal
sites,
the
resulting
surveyed
cumulative
dune
retreat
increased
to
the
extent
that
it
approached
the
calculated
DE
ma
,
for
that
final
storm
(Komar
et
al.,
2000).
Thus,
although
one
storm
may
not
have
sufficient
time
to
produce
dune
erosion
to
the
extent
calculated
with
the
geometric
model,
a
series
of
storms
could,
justifying
the
use
of
the
evaluated
DE
max
in
coastal
management
to
establish
setback
distances.
The
primary
problem
we
have
had
to
deal
with
is
what
exactly
constitutes
an
"extreme"
erosive
event
for
the
PNW,
the
event
for
which
setback
distances
are
need.
The
difficulty
is
that
multiple
processes
play
a
role,
those
reviewed
above,
some
of
which
tend
to
occur
at
the
same
time
(as
in
a
particular
storm,
or
during
an
El
Nitio),
while
occurrences
of
other
processes
are
independent.
With
multiple
processes,
each
potentially
with
its
own
series
of
25-
through
100-year
probabilities
of
occurrence,
the
question
is
how they
can
be
combined
in
a
rational
way
to
form
joint
probabilities
where
the
combined
processes
have
25-
through
100-yr
expected
occurrences.
Our
approach
to
accomplish
this
has
evolved
with
time,
and
continues
to
change
as
we
learn
more
about
the
processes.
As
depicted
in
Figure
4,
there
are
two
components
that
control
occurrences
of
extreme
total
water
levels
measured
tides
(
E
T
)
and
the
runup
levels
expected
during
major
storms
(R
2%
).
One
might
(mistakenly)
simply
add
the
corresponding
extreme
occurrences
of
E
T
and
R.
.
Our
present
best
estimates
of
those
values
are
listed
in
Table
1,
based
on
standard
extreme-value
analyses,
typically
the
fit
to
a
Weibel
distribution
(Allan
and
Komar,
2001).
The
extreme
values
of
E
T
are
based
on
the
long-term
record
of
measured
tides
in
Yaquina
Bay
at
Newport
on
the
Oregon
coast
(Fig.
1);
having
been
derived
from
measured
water
elevations,
they
include
both
the
effects
of
El
Nitios
that
raise
monthly
mean
water
levels
plus
occurrences
of
storm
surges.
The
list
in
Table
1
for
extreme
values
of
R
N
,
is
derived
from
the
measured
deep-water
wave
heights
and
periods
measured
by
the
Washington
buoy
#46005,
with
the
runup
having
been
calculated
using
equation
(1),
followed
by
an
extreme-value
analysis.
The
extreme
values
of
the
deep-water
wave
heights,
H..,
are
also
listed
in
Table
1,
but
they
were
not
used
directly
in
the
calculation
of
the
extreme
values
of
R2%
since
it
is
the
joint
occurrence
of
deep-water
wave
heights
and
periods
that
controls
the
runup,
and
the
100-yr
value
of
might
correspond
to
a
comparatively
short
wave
period
(certainly
not
the
100-yr
wave
period).
A
similar
problem
is
faced
in
evaluating
the
joint
occurrence
of
extreme
values
of
E
T
and
R
N
,
combined
to
determine
the
total
water
level,
WL
=
E
T
+
R
2
,.
For
example,
if
the
100-year
extreme
tide
(E
T
=
2.50
m
NGVD29)
is
added
to
the
100-year
storm
wave
runup
(
R
N
=
5.58
m),
a
total
water
level
of
8.08
m
NGVD29
is
predicted
for
the
extreme
combination.
However,
such
an
event
has
a
low
probability,
since
while
the
highest
water
levels
tend
to
occur
during
El
Nifios,
the
arrival
of
the
most
extreme
storm
waves
on
the
Oregon
coast
can
occur
at
any
time.
In
this
respect
the
PNW
coast
differs
from
areas
where
hurricanes
or
typhoons
represent
the
extreme
event,
with
the
high
water
of
the
storm
surge
corresponding
in
time
to
the
highest
generated
waves.
On
the
PNW
coast
the
extreme
measured
tides
are
statistically
independent,
so
the
joint
probability
is
the
product
of
their
two
individual
probabilities.
With
a
statistical
lack
of
correlation
between
the
100-year
extreme
water
level
and
100-year
storm
waves,
the
8.08
m
NGVD29
total
818
COASTAL
DISASTERS
'02
water
level
calculated
above
by
adding
the
two
processes
has an
expected
return
period
of
100
X
100
=
10,000
years.
So
simple
additions
at
any
level
in
Table
I
are
not
particularly
representative
of
what
might
actually
occur
on
the
PNW
coast.
Table
1.
Extreme
values
of
measured
tides,
deep-water
significant
wave
heights,
runup
levels
on
beaches,
and
total
water
levels
due
to
tides
plus
wave
runup
(all
values
in
meters).
Extreme
Meas.
Tidal
Deep-Water
Sig.
Wave
Runup
Total
Elev.'
Condition
Elev.',
E
T
Height',
H_
Leve1
3
,
R
2
,
6
E
T
+
R
2
,
2-year
9.4
3.82
5.29
5-year
2.23
11.4
4.40
5.97
10-year
2.33
12.4
4.74
6.35
25-year
2.41
13.6
5.11
6.78
50-year
2.46
14.3
5.35
7.07
75-year
2.48
14.5
5.48
7.22
100-year
2.50
15.0
5.58
7.33
'Referenced
to
the
1929
NGVD
elevation
according
to
the
Newport,
OR,
tide
gauge.
2
Wave
statistics
derived
from
Washington
buoy
#46005.
'Calculated
from
eq.
(1)
using
a
beach
slope
S
=
0.04,
representative
of
PNW
beaches.
One
approach
we
have
taken
is
to
develop
a
series
of
extreme-event
scenarios
that
reflect
the
expected
contrasting
processes
of
El
Nifios,
La
Nifias
and
intermediate
("normal")
winters
(Komar
and
Allan,
2000).
In
brief,
the
three
scenarios
considered
were:
Scenario
#1:
Normal
(non-El
Nifio)
winter
when
monthly
mean
water
levels
are
close
to
the
long-term
average,
with
the
occurrence
of
a
100-year
storm
primarily
being
responsible
for
extreme
erosion.
Scenario
#2:
An
El
Nifio
winter
characterized
by
high
measured
tides,
but
with
a
lower
probability
of
a
major
storm.
Scenario
#3:
A
"worst
case"
situation
when
a
major
storm
does
occur
during
an
El
Nifio
when
there
also
are
elevated
measured
tides.
For
each
scenario
the
analysis
of
the
extreme
total
water
level
involved
a
summation
of
the
individual
processes
according
to
the
model
of
Figure
4.
Accordingly,
a
reasonably
expected
predicted
high
tide
was
enhanced
by
a
monthly
mean
water
level,
with
the
latter
being
greater
for
Scenarios
#2
and
#3
in
order
to
represent
El
Nino
conditions.
The
measured
tide
may
be
further
increased
by
the
occurrence
of
a
storm
surge,
the
value
of
which
depends
on
the
extreme
nature
of
the
storm
itself,
which
also
governs
the
value
of
the
runup,
Because
the
storm
tracks
during
an
El
Niho
tend
to
cross
the
coast
of
California
to
the
south,
resulting
in
reduced
wave
activity
on
the
PNW
coast
(Seymour,
1996),
Scenario
#2
represented
a
lesser
storm
event
than
did
Scenario
#1
for
a
non-El
Nino
winter
when
it
was
expected
that
the
storms
would
pass
directly
over
the
PNW,
bringing
a
higher
probability
for
a
major
storm.
On
average
that
may
be
correct,
but
the
occurrence
of
a
major
storm
(then
assessed
as
a
100-year
event)
on
the
PNW
coast
during
the
El
Nino
winter
of
1997-98
demonstrated
that
it
is
still
possible
to
have
an
extreme
storm
with
accompanying
storm
surge
and
high
R,,
runup
levels
during
El
Nitios.
This
led
to
the
development
of
Scenario
#3,
the
"worst
case"
event
when
a
major
storm
does
occur
during
an
El
Niiio
winter.
This
Scenario,
COASTAL
DISASTERS
'02
819
presented
in
Komar
and
Allan
(2000),
has
been
used
with
some
modification
for
the
assessment
of
setbacks
in
foredune
areas
along
the
PNW
coast
(e.g.,
Allan
and
Priest,
in
press).
A
further
revision
of
Scenario
#3
is
presented
here,
in
effect
a
"worst
case"
event
that
in
effect
represents
the
occurrence
of
a
storm
comparable
to
that
on
2-4
March
1999
but
during
an
El
Nino
winter.
The
resulting
list
of
processes
affecting
the
measured
tide
plus
the
storm
wave
runup
are:
Predicted
Tide:
1.30
meters
NGVD29
Monthly
Mean
Water
Level:
0.50
meters
Storm
Surge:
1.60
meters
Wave
Runup,
R
n
,
5.11
meters
NGVD29
TOTAL
WATER
LEVEL
8.51
meters
NGVD29
The
1.60
m
value
for
the
storm
surge
elevation
is
that
measured
on
the
Washington
coast
during
the
March
1999
storm,
after
the
monthly-mean
water
level
had
been
subtracted
(Allan
and
Komar,
in
press).
In
this
Scenario
the
tide
is
increased
by
0.50
m
to
account
for
the
processes
related
to
an
El
Niiio
(warmer
water
temperatures,
etc.);
actual
measurements
indicate
that
this
could
actually
be
as
great
as
0.70
m
(Komar
et
al.,
2000).
The
R
2%
=
5.11
m
runup
value
is
that
for
the
25-year
storm
event
according
to
Table
1.
Of
interest,
we
now
assess
the
major
storm
of
2-4
March
1999
as
representing
approximately
a
25-
to
50-year
event
based
on
the
14.1-m
deep-water
significant
wave
height
and
calculated
runup
level
which
also
depends
on
the
wave
periods
during
the
storm
(Allan
and
Komar,
2001).
The
total
extreme
water
level
in
this
revised
scenario
comes
to
8.5
m
NGVD29.
There
is
reason
to
believe
that
this
could
happen,
and
would
have
done
so
had
the
March
1999
storm
occurred
a
year
earlier,
during
the
1997-98
El
Nitio.
Unfortunately,
it
is
essentially
impossible
to
assess
its
actual
probability
of
occurrence.
The
tide-level
factors
in
the
Scenario
add
up
to
a
measured
tidal
elevation
of
3.4
m
NGVD29.
According
to
Table
1,
this
is
nearly
1-m
greater
than
the
100-year
projected
extreme
tide.
This
difference
is
due
to
the
extreme
1.6-m
storm
surge
of
the
March
1999
storm
used
in
the
Scenario,
whereas
storm
surges
accompanying
more
common
storms
are
generally
less
that
0.50
m.
The
lower
value
of
the
projected
100-year
extreme
tide
in
Table
1
reflects
the
fact
that
a
storm
surge
greater
than
1.0
m
has
not
occurred
during
an
El
Nitio
winter
during
the
25-year
time
frame
of
the
data
analyzed.
But
we
cannot
be
certain
that
it
will
not
occur
in
the
future.
The
probability
of
the
above
Scenario
occurring
is
also
complicated
by
the
recognition
that
there
are
long-term
cycles
between
the
frequencies
of
El
Nirios
versus
La
Nifias,
defined
as
the
Pacific
Decadal
Oscillation,
the
PDO
(Mantua
et
al.,
1997).
Having
experienced
some
25
years
in
the
El
Nino
dominated
part
of
the
cycle,
evidence
suggests
that
we
are
now
entering
the
La
Nifia
dominated
phase.
If
so,
this
would
decrease
the
probability
of
the
Scenario
presented
here,
at
least
for
the
next
25
years.
The
last
column
in
Table
I
lists
the
extreme
values
for
total
water
levels
E
T
+
1?.
2%
,
derived
from
a
direct
analysis
of
the
day-to-day
combined
measured
tides
and
calculated
wave
runup
levels,
based
on
daily
measurements
of
deep-water
wave
heights
and
periods; the
analysis
spans
the
25-year
period
for
which
we
have
both
measured
tides
and
waves.
The
100-year
extreme
projection
is
7.33
m
NGVD29,
which
again
is
lower
than
that
predicted
by
the
Scenario,
for
the
same
reason
just
discussed
for
the
extreme
measured
tide.
Such
a
direct
analysis
of
E
T
+
is
particularly
versatile
in
820
COASTAL
DISASTERS
'02
that
it
not
only
yields
values
for
extreme
total
water
levels
resulting
from
the
combined
processes,
those
given
in
Table
1,
it
also
yields
predictions
of
average
hours
per
year
during
which
the
total
water
level
reaches
or
exceeds
a
certain
elevation.
These
two
aspects
derived
from
the
analysis
can
be
employed
respectively
to
evaluate
the
expected
extent
of
dune
erosion
and
the
susceptibility
of
sea
cliffs
to
wave
attack
and
erosion,
in
both
cases
leading
to
assessments
of
setback
distances.
In
the
case
of
foredune
erosion,
the
extreme
values
developed
either
in
the
above
Scenario
where
the
processes
are
considered
or
in
the
listing
of
E
T
+
R.
T
,
total
water
levels
in
Table
I
become
the
WL
water
level
depicted
in
Figure
5
and
used
in
Equation
(2)
to
calculate
the
associated
dune
erosion
using
the
geometric
model.
Important
to
the
resulting
extent
of
the
retreat
of
the
dunes
is
the
water-level
WL
compared
with
the
elevation
to
the
beach/dune
junction,
E.
This
is
also
the
case
for
sea
cliffs,
but
unlike
foredunes
it
is
not
possible
to
simply
predict
the
resulting
extent
of
cliff
erosion
since
it
depends
on
the
resistance
of
the
cliff
to
wave
attack,
factors
that
need
to
be
examined
in
a
site
visit
by
a
trained
coastal
scientist
or
engineer.
From
a
survey
of
the
elevation
of
the
toe
of
the
cliff,
our
analysis
methodology
provides
an
evaluation
of
the
average
number
of
hours
per
year
tides
plus
waves
can
reach
or
exceed
that
elevation,
that
is,
the
average
wave-impact
hours
per
year.
We
have
demonstrated
along
specific
areas
of
the
Oregon
coast
that
this
assessment
correlates
well
with
actual
measurements
of
cliff-
erosion
rates,
the
greater
the
wave-impact
hours
the
higher
the
rate
of
cliff
erosion
(Ruggiero
et
al.,
2001).
Therefore,
although
it
is
not
possible
to
provide
a
direct
prediction
of
cliff
erosion
rates
due
to
their
variable
compositions
and
resistance,
our
approach
does
provide
an
assessment
of
the
frequency
of
wave
attack,
and
thus
for
the
susceptibility
of
the
cliff
to
wave-induced
erosion.
It
needs
to
be
recognized
that
the
wave
runup
assessments
presented
here,
including
the
extreme
values
listed
in
Table
1,
are
based
on
calculations
of
R
2%
assuming
a
beach
slope
S
=
0.04.
Although
this
slope
is
fairly
representative
of
PNW
beaches,
in
applications
it
would
be
necessary
to
revise
the
values
for
sites
where
beach
slopes
are
different.
We
also
want
to
point
out
that
we
are
still
mainly
using
the
NGVD29
elevation
datum
in
applications
along
the
Oregon
coast,
because
the
Statutory
Vegetation
Line
established
by
State
law
is
based
on
that
datum.
The
values
relative
to
the
NAVD'88
datum
for
Oregon
and
Washington
are
obtained
approximately
by
subtracting
1.0
m
from
the
NGVD29
elevations.
CONCLUSIONS
AND
DISCUSSION
The
unusually
severe
storms
and
accompanying
beach
and
property
erosion
experienced
along
the
PNW
coast
in
recent
years
has
led
to
re-examinations
of
the
wave
climate,
erosion
processes,
and
recommended
setback
distances.
Table
1
presented
the
results
of
the
extreme-value
analyses,
important
to
the
evaluation
of
potential
major
storms
and
the
erosion
that
might
result.
Based
on
wave
data
collected
up
through
1996,
Tillotson
and
Komar
(1997)
had
projected
that
the
100-year
storm
would
generate
deep-water
significant
wave
heights
of
about
10
m.
With
the
occurrence
of
six
storms
above
that
level
during
the
following
three
winters,
the
revised
assessment
now
places
the
value
at
15.0
m,
a
substantial
change.
Both
the
extreme-value
analysis
and
development
of
Scenarios
presented
here
adopt
a
rather
static
view
of
the
wave
climate
and
nearshore
processes,
one
that
does
not
account
for
the
discovery
that
wave
heights
and
related
nearshore
processes
have
been
increasing
for
at
least
the
past
25
years
(Allan
and
Komar,
2001).
At
present
we
do
not
have
a
firm
understanding
of
the
underlying
cause
for
this
increase,
but
it
must
be
related
in
some
way
to
a
change
in
the
Earth's
climate.
Without
this
understanding
we
cannot
predict
whether
the
increase
will
continue
into
the
future,
a
prediction
that
is
important
COASTAL
DISASTERS
'02
821
to
the
placement
of
setbacks.
Lacking
this
certainty
we
suggest
the
implementation
of
two
setbacks,
one
for
small
developments
that
uses
the
values
given
in
Table
1
or
a
Scenario
based
on
those
values,
and
a
second
setback
for
more
substantial
developments
such
as
hotels
or
public
facilities,
one
that
factors
in
an
assumption
that
wave
heights
and
nearshore
processes
will
continue
to
increase
for
at
least
another
25
years,
yielding
a
greater
setback
distance
between
the
constructed
development
and
the
erosive
forces
of
the
waves.
Additional
research
is
required
to
improve
our
models
and
methodologies.
An
important
erosive
process
on
PNW
beaches
is
the
occurrence
of
rip
currents
and
the
embayments
they
cut
into
the
beach,
becoming
the
focal
areas
of
maximum
property
loss
(Komar,
1997).
More
survey
information
is
required
concerning
the
effects
of
these
embayments
locally
on
the
beach
morphology,
knowledge
of
which
is
required
in
assessments
of
the
beach-level
change
ABL
in
the
geometric
model
(Fig.
5)
and
in
equation
(2).
On
a
larger
scale,
we
still
do
not
have
a
systematic
methodology
to
include
the
"hot
spot"
erosion
effects
of
El
Nitios,
which
locally
require
greater
setbacks
than
predicted
by
the
geometric
model.
This
is
being
investigated
using
the
LIDAR
surveys
that
spanned
the
1997-98
El
Nino,
documenting
the
extent
of
erosion
in
the
"hot
spot"
zones
(Revell
et
al.,
in
press).
The
focus
of
this
paper
has
been
on
the
coast
of
the
U.S.
Pacific
Northwest,
the
States
of
Oregon
and
Washington.
It
is
our
hope
that
the
models
and
methodologies
developed
for
the
PNW
will
be
adopted
for
use
on
other
coasts.
The
models
are
sufficiently
general
for
such
an
adoption,
but
on
other
coasts
they may
also
need
to
consider
such
things
as
dune
overwash
as
well
as
dune
undercutting
[e.g.,
Sallenger
et
al.
(1999)].
Of
course,
all
of
the
wave
climate
and
water-level
values,
like
those
in
Table
1
derived
for
application
to
the
PNW
coast,
will
require
research
programs
to
provide
comparable
values
for
use
on
other
coasts.
Ultimately
the
results
should
permit
a
more
rational
foundation
for
the
placement
of
setbacks,
one
that
better
accounts
for
the
erosion
processes.
REFERENCES
Allan,
J.C.,
and
Komar,
P.D.
2000.
Long-Term
and
Climate-Related
Increases
in
Storm
Wave
Heights
in
the
North
Pacific.
EOS,
81
(47),
pp.
561
and
566-567.
Allan,
J.C.,
and
Komar,
P.D.
2001.
Wave
Climate
Change
and
Coastal
Erosion
in
the
US
Pacific
Northwest.
Proceedings
of
WAVES
2001,
Amer.
Soc.
Civil
Engrs.
Allan,
J.C.,
and
Komar,
P.D.
(in
press)
Extreme
Storms
on
the
Pacific
Northwest
Coast
during
the
1997-98
El
Nino
and
1998-99
La
Nina.
Journal
of
Coastal
Research.
Allan,
J.C.,
and
Komar,
P.D.
(in
review)
The
Wave
Climate
of
the
Eastern
North
Pacific:
Long-Term
Trends
and
El
Nino/La
Nina
Dependence:
Journal
of
Coastal
Research.
Allan,
J.C.,
and
Priest,
G.R.,
in
press.
Evaluation
of
Coastal
Erosion
Hazard
Zones
along
Dune
and
Bluff
Backed
Shorelines
in
Tillamook
County,
Oregon.
Oregon
Dept.
of
Geology
and
Mineral
Industries
Technical
Report,
Portland,
Oregon.
Bruun,
P.
1962.
Sea-Level
Rise
as
a
Cause
of
Shore
Erosion.
Journal
Waterways
and
Harbors
Division,
Amer.
Soc.
Civil
Engrs.,
88
(WW1),
117-130.
Graham,
N.E.,
and
Diaz,
H.F.
2001.
Evidence
for
Intensification
of
North
Pacific
Winter
Cyclones
since
1948.
Bulletin
of
the
American
Meteorological
Society,
82,
1869-1893.
Holman,
R.A.
1986.
Extreme
Value
Statistics
for
Wave
Runup
on
a
Natural
Beach.
Coastal
Engineering,
9,
527-544.
822
COASTAL
DISASTERS
'02
Kaminsky,
G.,
and
others.
1999.
Synthesizing
Geological
Observations
and
Process-
Resonse
Data
for
Modeling
Coastal
Change
at
Management
Scale.
Coastal
Sediments
'99,
Amer.
Soc.
Civil
Engrs.,
1708-1723.
Komar,
P.D.
1986.
The
1982-93
El
Nitio
and
Erosion
on
the
Coast
of
Oregon.
Shore
&
Beach,
54,
3-12.
Komar,
P.D.
1997.
The
Pacific
Northwest
Coast:
Living
with
the
Shores
of
Oregon
and
Washington:
Duke
University
Press,
Durham,
NC,
195
pp.
Komar,
P.D.,
1998a.
Beach
Processes
and
Sedimentation:
2nd
edition,
Prentice-Hall,
Upper
Saddle
River,
New
Jersey,
544
pp.
Komar,
P.D.,
1998b.
The
1997-98
El
Nifio
and
Erosion
on
the
Oregon
Coast.
Shore
&
Beach,
66
(3),
33-41.
Komar,
P.D.,
McDougal,
W.G.,
Marra,
J.J.,
and
Ruggiero,
P.
1999.
The
Rational
Analysis
of
Setback
Distances:
Applications
to
the
Oregon
Coast.
Shore
&
Beach,
67
(1),
41-49.
Komar,
P.,
Allan,
J.,
Diaz-Mendez,
G.M.,
Marra,
J.J.,
and
Ruggiero,
P.,
2000.
El
Nino
and
La
Nina
-
Erosion
Processes
and
Impacts:
Proceeding
27th
International
Conf
on
Coastal
Engineering,
Amer.
Soc.
Civil
Engrs.,
2414-2427.
Komar,
P.D.,
Diaz-Mendez,
G.M.,
and
Marra,
J.J.
2001.
Stability
of
the
New
River
Spit,
Oregon,
and
the
Position
of
the
Beach-Zone
Line:
Journal
of
Coastal
Research,
17
(3),
625-635.
Kriebel,
D.L.,
and
Dean,
R.G.
1985.
Numerical
Simulation
of
Time-Dependent
Beach
and
Dune
Erosion.
Coastal
Engineering,
9,
221-245.
Larson,
M.,
and
Kraus,
N.C.
1989.
SBEACH:
Numerical
Model
for
Simulating
Storm
Induced
Beach
Change:
Technical
Report
CERC-89-9,
U.S.
Army
Corps
of
Engineers.
Mantua,
N.J.,
Hare,
S.R.,
Zhang,
Y.,
Wallace,
J.M.,
and
Francis,
R.C.
1997.
A
Pacific
Inderdecadal
Oscillation
with
Impacts
on
Salmon
Production.
Bulletin
of
the
American
Meteorological
Society,
78,
1069-1079.
Nairn,
R.B.,
and
Southgate,
H.N.
1993.
Deterministic
Profile
Modeling
of
Nearshore
Processes.
Part
2,
Sediment
Transport
and
Beach
Profile
Development.
Coastal
Engineering,
19,
57-96.
Ruggiero,
R.,
Komar,
P.D.,
McDougal,
W.G.,
and
Beach,
R.A.
1996.
Extreme
Water
Levels,
Wave
Runup
and
Coastal
Erosion.
Proceedings
25th
Coastal
Engineering
Conference,
Amer.
Soc.
Civil
Engrs.,
2793-2805.
Ruggiero,
P.,
Komar,
P.D.,
McDougal,
W.G.,
Marra,
J.J.,
and
Beach,
R.A.,
2001.
Wave
Runup,
Extreme
Water
Levels
and
the
Erosion
of
Properties
Backing
Beaches.
Journal
of
Coastal
Research,
17
(2),
407-419.
Sallenger,
A.,
and
others.
1999.
Scaling
Winter
Storm
Impacts
on
Assateague
Island,
MD,
VA.
Coastal
Sediments
'99,
Amer.
Soc.
Civil
Engrs.,
1814-1825.
Seymour,
R.J.
1996.
Wave
Climate
Variability
in
Southern
California.
Journal
of
Waterway,
Port,
Coastal
and
Ocean
Engineering,
Amer.
Soc.
Civil
Engrs.,
122,
182-186.
Seymour,
R.J.
1998.
Effects
of
El
Nino
on
the
West
Coast
Wave
Climate.
Shore
&
Beach,
66,
3-6.
Tillotson,
K.,
and
Komar,
P.D.
1997.
The
Wave
Climate
of
the
Pacific
Northwest
(Oregon
&
Washington):
A
Comparison
of
Data
Sources:
Journal
of
Coastal
Research,
13,
440-452.
Vellinga,
P.
1982.
Beach
and
Dune
Erosion
during
Storms.
Coastal
Engineering,
6,
361-387.
Wood,
W.L.,
and
others.
1990.
Managing
Coastal
Erosion:
National
Academy
Press,
Washington,
D.C.,
182
pp.