OCR | Wellesley College Digital Repository (2024)

Distinguishing
proximal
and
distal
sources
of
sandstone
by
combining

U-­‐Pb
age,
and
geochemical
signatures
of
detrital
zircon
populations:


An
example
from
the
Permian
Brushy
Canyon
Formation,
West
Texas

Sarah
George

Advisor:
David
Hawkins

Table
of
Contents

ACKNOWLEDGEMENTS


ABSTRACT………………………………………………………………………………………………...…………1

INTRODUCTION…………………………………………………………………………...………………………2

REGIONAL
GEOLOGY……………………………………………………………...…………………………….6

Equatorial
Pangea
and
the
Delaware
Basin…………………………………………………6

Previous
Provenance
Studies
of
the
BCFM…………………………………..……………10

The
Florida
Mountains
and
Rift
Related
Plutons……………………...……………......15

APPROACH………………………………………………………………………………………………………...17


BACKGROUND…………………………………………………………………………………………………...18

U-­‐Pb
Geochronology
and
Geochemistry
of
Zircon……………………………………..18

METHODS………………………………………………………………………………………………………….21

Rock
Sampling
in
the
Field……………………………………………………………………….21

Sample
Preparation…………………………………………………………………………………23

LA-­‐ICPMS………………………………………………………………………………………………..24

RESULTS……………………………………………………………………………………………………………26

Brushy
Canyon
Formation……………………………………………………………………….26

Florida
Mountain
Granite………………………………………………………………………...32

DISCUSSION……………………………………………………………………………………………………….36

REFERENCES
CITED……………….…………………………………………………………...……………..46

APPENDIX

ACKNOWLEDGEMENTS

I
would
like
to
thank
my
crazy
geologist
parents
who
dragged
me
around
the
world

as
a
child,
piquing
my
interest
in
geology.

I
owe
my
Wellesley
friends
a
great
deal
of

gratitude
for
their
tolerance
and
love
throughout
the
process.

I’d
also
like
to
thank

Dr.
Mark
Schmitz
and
Dr.
Jim
Crowley
for
allowing
me
to
use
their
lab
facilities
and

showing
me
the
ropes
of
zircon
geochronology,
as
well
as
Dr.
Jeff
Amato
for

generously
providing
samples
from
the
Florida
Mountain
granite.

Finally,
I
owe

Dave Hawkins
a
huge
thank
you
for
his
patience,
guidance
and
unwavering
support

throughout
the
past
few
years
and
with
this
thesis.

“The
desert
sharpened
the
sweet
ache
of
his
longing,
amplified
it,


gave
shape
to
it
in
sere
geology
and
clean
slant
of
light.”


―
Jon
Krakauer,
Into
the
Wild

ABSTRACT

Due
to
vast
petroleum
reserves,
the
Delaware
Basin
of
west
Texas
has
undergone

extensive
study,
but
the
provenance
of
the
Paleozoic
sediments
has
remained
poorly

constrained.

Approximately
50%
of
detrital
zircons
in
the
oldest
clastic
unit,
the

Brushy
Canyon
Formation,
are
ca.
0.5
Ga;
an
atypical
age
of
zircons
in
North
America.


This
study
uses
U-­‐Pb
geochronology,
trace
element
compositions
and
textural

characteristics
of
detrital
zircons
to
compare
ca.
0.5
Ga
detrital
zircons
in
the
Brushy

Canyon
Formation
to
coeval
zircons
from
a
potential
local
source,
the
Florida

Mountain
granite
of
New
Mexico.


The
Florida
Mountain
granite
is
correlated
with
a

suite
of
rift-­‐related
plutons
and
yields
an
age
of
517
±
11
Ma
(including
internal
and

systematic
error),
which
is
slightly
older
than
the
majority
of
ca.
0.5
Ga
BCFM

detrital
zircons.

Zircon
grains
from
the
Florida
Mountain
granite
also
overlap
in

composition,
for
most
trace
elements,
with
the
ca.
0.5
Ga
BCFM
detrital
zircons,

including
grains
with
textural
characteristics
consistent
with
igneous
origins.

These

results
suggest
that,
on
the
basis
of
age,
trace
element
composition
and
textural

characteristics,
ca.
0.5
Ga
BCFM
detrital
zircons
are
similar
to
zircons
in
the
Florida

Mountain
granite
and
hence,
the
rift
related
plutons
of
the
New
Mexico
Aulacogen

represent
a
likely
source.

The
unusual
occurrence
of
a
high
proportion
of
ca.
0.5
Ga

detrital
zircons
in
the
BCFM
appear
to
reflect
a
local
or
regional
source,
rather
than

a
distal
North
American
provenance.

1

INTRODUCTION

Provenance
studies
allow
geologists
to
link
mountainous
source
regions
with

sediment
deposited
in
basins
at
various
times
in
the
past
(Figure
1)
and
provide

important
evidence
for
paleogeographic
reconstructions
of
ancient
continents.


Modern
provenance
studies
go
beyond
the
traditional
methods
of
looking
at

mineralogy
and
paleocurrent
data
(e.g.,
Gardner
and
Borer,
2000)
by
evaluating
the

distribution
of
U-­‐Pb
ages
from
detrital
zircon
crystals
(e.g.,
Gehrels
et
al.,
2011).


Zircon
is
an
ideal
tracer
of
ancient
sediment
transport
because
it
is
common
in

igneous
and
metamorphic
source
rocks,
it
is
dense
and
highly
resistant
to

weathering
and
erosion,
and
it
is
an
ideal
U-­‐Pb
geochronometer
(Fedo
et
al.,
2003).

Source
Sink

Figure
1.

Mountain
(source)
to
ocean
basin
(sink)
cross-­‐section
showing
sediment

dispersal
system.
Modified
after
Grotzinger
and
Jordan
(2010).

Typically,
provenance
is
evaluated
by
comparing
the
age
distribution
of

measured
U-­‐Pb
ages
on
a
representative
population
of
detrital
zircon
crystals

separated
from
a
clastic
sedimentary
rock
with
age
spectra
from
potential
source

regions
(Fedo,
2003).

For
example,
Gehrels
et
al.
(2011)
used
the
distribution
of

ages
from
orogenic
belts
exposed
in
North
America
to
interpret
the
distribution
of

2

detrital
zircon
ages
from
sedimentary
rocks
exposed
in
the
Grand
Canyon
(Figure

2b).


However,
Thomas
(2011)
noted
that
the
peak
matching
methodology

employed
by
Gehrels
et
al.
(2011)
ignores
important
aspects
of
sediment
systems

such
as
recycling
of
sediments
and
clastic
inputs
from
local
sources.

Specifically,

Thomas
(2011)
found
that
much
of
the
detrital
zircon
age
distribution
in
the
Grand

Canyon
stratigraphy
could
be
accounted
for
by
local
sources
rather
than
the
distal

sources
proposed
by
Gehrels
et
al.
(2011).

Moreover,
because
orogenic
belts
span

thousands
of
kilometers
and
rocks
of
a
single
age
range
crop
out
over
large
areas
of

the
continent
(Figure
2a),
additional
constraints
are
needed
to
more
precisely
match

detrital
zircons
with
their
original
source
region
and
to
distinguish
local
sources

from
distal
sources
(Thomas,
2011).

Sandstones
of
the
late
Paleozoic
Delaware
Basin
provide
an
excellent

opportunity
to
distinguish
local
sediment
sources
from
distal
sediment
sources.
The

Brushy
Canyon
Formation
(BCFM)
is
an
early
to
middle
Permian
delta
deposit
in
the

Delaware
Basin
of
West
Texas
(Figure
3)
(Carr
and
Gardner,
2000).

Based
on

paleoflow
indicators
and
the
geometry
of
sand
channels,
Gardner
and
Borer
(2000)

concluded
that
the
source
of
sediment
for
the
BCFM
lay
to
the
north-­‐northwest
and

inferred
the
source
area
to
be
the
Ancestral
Rocky
Mountains.

However,
the
U-­‐Pb

age
distribution
of
detrital
zircons
from
the
lower
and
middle
BCFM
requires

additional
source
areas
(Cantine
et
al.,
2013).

More
than
50%
of
the
detrital
zircons

are
ca.
0.5
Ga,
and
therefore,
younger
than
rocks
thought
to
be
exposed
in
the

Ancestral
Rockies
during
the
Late
Paleozoic
(Cantine
et
al.,
2013;
Duncan
et
al.,

3

2013).

Cantine
et
al.
(2013)
found
that
granitic
and
sedimentary
rocks
such
as
those

exposed
today
in
the
Florida
Mountains
of
New
Mexico
(Amato
and
Mack,
2012)
are

very
similar
in
age
to
zircons
deposited
in
the
BCFM
during
the
Late
Paleozoic.

The

goal
of
this
study
is
to
test
whether
a
local
source
area
such
as
the
ancient
Florida

Mountains
can
account
for
ca.
0.5
Ga
sediment
in
the
BCFM
by
supplementing
U-­‐Pb

ages
of
detrital
zircon
with
additional
proxies
such
as
the
trace
element

compositions
and
textural
characteristics
of
those
crystals.

Figure
2.
(a)
Proterozoic
crustal
age
provinces
of
Laurentia
and
surrounding

continents
at
1.7
Ga.

Each
color
represents
rocks
of
different
ages.

Note
the
lateral

extent
of
each
age
province.

From
Karlstrom
et
al.
(2012)

(b)
Detrital
age
spectra

from
Grand
Canyon
(dark
grey)
compared
to
the
age
distribution
of
rocks
from

orogenic
belts
(light
grey)
that
represent
potential
source
regions
for
Grand
Canyon

sediment.

Note
that
although
there
are
similarities,
no
single
source
can
account
for

all
peaks
in
the
Grand
Canyon
spectrum.

From
Gehrels
et
al.
(2011).

4

Figure
3.
Map
of
Delaware
Basin
and
surrounding
Permian
aged
basins.

Grey

represents
deep-­‐water
deposits.

From
Olszewski and Erwin (2009).

5

REGIONAL
GEOLOGY

Equatorial
Pangea
and
the
Delaware
Basin


The
ancient
supercontinent
Pangea
was
an
amalgamation
of
modern
day

North
America,
South
America,
Antarctica,
India,
Eurasia
and
Australia
that
was
fully

assembled
during
the
early
Permian
and
persisted
until
the
start
of
the
Jurassic

(Figure
4;
Blakey,
2013).

Through
evidence
like
paleomagnetism,
shared
fossil

records
and
the
continuity
of
mountain
ranges,
geologists
have
a
global

understanding
of
the
distribution
of
continents
during
the
Permian
(e.g.,
Blakey,

2013).
However,
local
paleogeography
is
not
well
constrained
and
is
under
constant

revision
(e.g.,
Blakey,
2013).

Provenance
of
sediment
in
basins
can
be
used
to
help

constrain
local
uplifts
and
sediment
dispersal
pathways
and
hence
can
inform
more

accurate
paleogeographic
reconstructions
(Fedo,
2003).

In
the
case
of
the
Delaware

Basin,
the
sediments
deposited
during
the
Permian
are
preserved
in
the
Guadalupe

Mountains
of
West
Texas
offering
some
insight
into
the
regional
paleogeography
of

equatorial
Pangea.

NA

DB

PANTHALASSA

SA

AP-­‐OM

AF

Figure
4.
Global
paleogeographic
reconstruction
of
Pangea
at
280
Ma.

DB=

Delaware
Basin;
AP-­‐OM=
Appalachian
and
Ouachita-­‐Marathon
Mountains;
AF=

future
Africa,
Antarctica
and
India;
NA=
future
North
America;
SA=
future
South

America.

Note
that
the
Delaware
Basin
was
located
on
the
western
margin
of

equatorial
Pangea.

Map
from
Blakey
(2013).

6

DB

Figure
5.
Zoomed
in
paleogeographic
reconstruction
of
western
United
States
at

280
Ma.

Note
the
paleohighs
near
the
Delaware
Basin.

DB=
Delaware
Basin.

Modified
from
Blakey
(2013).

During
the
Permian,
the
Delaware
Basin
was
one
of
the
southernmost
basins

of
equatorial
Pangea
filled
with
an
epicontinental
sea
(Figure
5)
(Gardner
and
Borer,

2000;
co*cks
and
Torsvik,
2011).

The
Guadalupe
Mountains
and
surrounding
hills

expose
the
Permian
delta
deposit,
revealing
sequences
of
bedded
siltstone

punctuated
by
sandstone
lenses
representing
submarine
sand
channels
during
sea

level
lowstands
(Carr
and
Gardner,
2000).

Carbonates
that
were
deposited
along

the
basin
margin
during
the
early
Permian
impacted
the
depositional
patterns
of

younger
clastic
units
such
as
the
conformable
Brushy
Canyon
Formation;
an
early
to

middle
Permian
unit
hence
referred
to
as
the
BCFM
(Gardner
and
Borer,
2000;

Soreghan
and
Soreghan,
2013).
The
BCFM
along
with
the
overlying
units,
the
Cherry

7

Canyon
Formation
and
the
Bell
Canyon
Formation,
make
up
the
Delaware
Mountain

Group
(Figure
6).

Figure
6.
Schematic
lithostratigraphy
of
the
Delaware
Basin,
TX
from
NW
to
SE.

Yellow=
Sandstone,
Blue=
Limestones
and
Dolostones,
Red=
Dolostones.

Note
that

the
Brushy
Canyon
Formation
is
the
oldest
clastic
unit
in
the
Delaware
Mountain

Group.
From
Soreghan
and
Soreghan
(2013).

The
Delaware
Mountain
Group
is
a
laterally
extensive
(approximately
110

km
by
220
km)
clastic
body
containing
an
approximately
7,600
m
thick
package
of

Paleozoic
sediment
(Payne,
1976).

The
sediment
in
the
Delaware
Mountain
group
is

comprised
of
two
predominant
lithologies;
large
siliciclastic
sand
channel

complexes
that
were
deposited
during
sea
level
lowstands
and
laminated
siltstones

that
were
deposited
during
highstands
(e.g.
Carr
and
Gardner,
2000).

The
channel

complexes
are
characterized
by
well-­‐sorted,
well-­‐rounded,
clast
supported,
medium

grained
quartz
arenites
to
arkosic
sandstones.

Due
to
the
roundness
of
grains,
the

relatively
hom*ogeneous
grain
size
distribution
and
dunes
along
the
shelf
edge,
it
is

inferred
that
sediment
was
transported
to
the
shelf
edge
via
eolian
transport
and

8

was
brought
into
the
basin
during
high-­‐energy
flow
events
when
sea
level
dropped

(Soreghan
and
Soreghan,
2013;
Gardner
and
Borer,
2000).

This
interpretation
is

consistent
with
the
strong
trade
winds
and
monsoonal
transport
that
contributed
to

large
amounts
of
sediment
movement
during
the
Permian
in
southwestern
Pangea

as
shown
in
Figure
7
(Soreghan
and
Soreghan,
2013).


The
Paleozoic
sediment
in
the
Delaware
Basin
is
exposed
along
Tertiary
aged

normal
faults
associated
with
Basin
and
Range
extension
(Gardner
and
Borer,
2000).



A
combination
of
faulting
and
nonhorizontal
depositional
surfaces
has
led
to
a

moderate
(2
to
6
degree)
dip
to
the
East
(Payne,
1976).

Figure
7.
Inferred
fluvial
and
eolian
transport
during
the
middle
Permian.

DB=

Delaware
Basin,
MB=
Midland
Basin.

Note
that
the
monsoonal
transport
direction

near
the
Delaware
Basin
is
from
the
paleo-­‐Northwest
which
is
the
location
of
the

Florida
Mountains
and
other
rift
related
0.5
Ga
plutons.

From
Soreghan
and

Soreghan
(2013).

9

Previous
Provenance
Studies
of
the
BCFM

The
Delaware
Basin
is
one
of
three
Permian
depocenters
which
together

make
up
the
Permian
Basin.

Over
the
past
30
years,
the
area
has
undergone

extensive
study
due
to
petroleum
reserves
within
the
basin.

Despite
the
well-­‐
characterized
stratigraphy
and
extensive
field
study,
the
source
of
sediment
has

remained
poorly
constrained
(Soreghan
and
Soreghan,
2013).

Based
on
paleoflow

indicators
and
the
geometry
of
sand
channels,
Gardner
and
Borer
(2000)
concluded

that
the
source
of
sediment
for
the
BCFM
lay
to
the
north-­‐northwest
and
inferred

the
source
area
to
be
the
Ancestral
Rocky
Mountains
(Figure
8).

However,
Cantine

et
al.
(2013)
found
that
less
than
10%
of
the
detrital
zircons
in
the
BCFM
had
ages

and
trace
element
compositions
that
could
be
compatible
with
an
Ancestral
Rocky

Mountain
source.

Moreover,
Cantine
et
al.
(2013)
found
a
distinctive
peak
at
ca.
0.5

Ga
comprising
approximately
50%
of
the
zircons
in
the
lower
to
middle
BCFM.



500
Ma
rocks
are
uncommon
in
North
America
and
thus
there
are
limited

sources
that
could
have
provided
the
ca.
0.5
Ga
detrital
zircons
found
in
the
BCFM.


Figures
9a
to
9e
show
ca.
0.5
Ga
zircons
in
the
BCFM
compared
to
published
zircon

probability
distributions
from
potential
sources
with
ca.
0.5
Ga
zircons
(Figure
8

shows
a
map
of
potential
sources).

Based
on
ages
alone,
the
Ouachita
Mountains

and
the
Appalachian
Mountains
(Figure
9a),
and
the
Acatlan
magmatic
and

metasedimentary
rocks
of
present
day
Mexico
(Figure
9b
and
9c)
can
be
excluded
as

sources
for
the
ca.
0.5
Ga
zircons
in
the
BCFM.


Figure
9d
and
9e
show
the
Bliss
Sandstone,
a
unit
derived
from
the
FM

granite
(Amato
and
Mack,
2012),
and
the
FM
granite
respectively
compared
to
ca.

10

0.5
Ga
zircons
from
the
BCFM.

Based
on
ages
published
by
Amato
and
Mack
(2012),

the
Florida
Mountains
have
the
same
age
(within
error)
as
the
BCFM
ca.
0.5
Ga

zircons
and
hence
the
Florida
Mountain
granite,
or
another,
similarly
aged
pluton,

are
a
likely
source
of
ca.
0.5
Ga
zircons
in
the
BCFM.



Soreghan
and
Soreghan
(2013)
obtained
detrital
zircon
age
spectra
to

identify
the
sources
of
Paleozoic
sediment
in
the
Delaware
Basin,
including
the

BCFM.

Soreghan
and
Soreghan
(2013)
suggested
the
Ouachita-­‐Marathon
belt
in

present
day
Mexico
as
a
potential
source
based
on
the
direction
of
inferred
trade

winds
and
roughly
similar
ages.

However,
as
Cantine
et
al.
(2013)
showed,
ages

alone
preclude
the
Ouachita
Mountains
from
being
the
source
of
ca.
0.5
Ga
zircons
in

the
BCFM.

Figure
8.
Sources
considered
by
Cantine
et
al
(2013)
at
280
Ma.

The
Delaware

Basin
is
circled
in
red
and
potential
sources
are
circled
in
black.

Note
the
distances

(and
regional
extent)
between
potential
sources
and
the
Delaware
Basin.

The

Mexican
Blocks
includes
the
Acatlan
province.

Map
modified
from
Blakey
(2013).

11

A

B

12

C

D

13

E

F

Figure
9.
Probability
distribution
plot
of
BCFM
ca.
0.5
Ga
zircons
compared
to
ca.

0.5
Ga
rocks
from:
(A)
the
Ouachita-­‐Southern
Appalachian
Mountains
(B)
Acatlan

metasedimentary
rocks
(C)
Acatlan
igneous
rocks
(D)
the
Bliss
Sandstone
derived

from
the
Florida
Mountain
granite
(E)
the
Florida
Mountain
granite
and
(F)

Paleozoic
sediment
in
the
Grand
Canyon.

Published
data
is
shown
with
systematic

errors
to
allow
comparison
to
Cantine
et
al.
(2013)
although
the
data
was
originally

published
with
only
internal
errors.

14

The
Florida
Mountain
Granite
and
Rift
Related
Plutons


The
Florida
Mountains
(FM)
are
a
roughly
10
km
long,
north-­‐south-­‐trending

mountain
range
in
present
day
south-­‐central
New
Mexico

(Brown
and
Clemons,

1983).

The
range
is
divided
by
a
WNW-­‐striking
high
angle
reverse
fault
into
a

northern
portion
dominated
by
Precambrian
igneous
rocks
and
a
southern
portion

dominated
by
Paleozoic
sediments
(Brookins,
1974).

The
FM
primarily
consist
of
a

large
granitic
pluton
emplaced
around
510
±
5
Ma
(Amato
and
Mack,
2012).

The

overlying
Bliss
Sandstone
contains
granite
clasts
identical
to
the
FM
granite

indicating
a
nonconformable
contact
relationship,
a
conclusion
supported
by
U-­‐Pb

ages
from
detrital
zircons
(e.g.,
Figure
9d,
9e;
Amato
and
Mack,
2012).



The
FM
granite
is
one
of
a
series
of
Cambro-­‐Ordovician
granite
to
syenite

plutons
that
occur
in
a
broad
band
from
southern
New
Mexico
to
central
Colorado

(Figure
10;
McMillan
and
McLemore,
2004).
Due
to
similarities
in
ages
and

distinctive
geochemical
compositions
that
suggest
a
perturbed
mantle
consistent

with
rifting,
this
series
of
rift
related
plutons
has
been
correlated
with
the
Oklahoma

Aulacogen
(McMillan
and
McLemore,
2004).


McMillan
and
McLemore
(2004)

attempted
identify
plutons
related
to
the
New
Mexico
Aulacogen,
however,
their

attempts
were
limited
by
the
paucity
of
published
ages
in
the
region.

15

FM

Figure
10.
Map
showing
the
interpreted
link
between
New
Mexico-­‐Colorado
ca.
0.5

Ga
plutons
and
Oklahoma
Aulacogen.

FM=
Florida
Mountains;
Black
stars
represent

known
Cambro-­‐Ordovician
igneous
rocks;
White
stars
represent
suspected
Cambro-­‐
Ordovician
rocks.
Note
the
NE-­‐SW
trend
of
the
rift
related
plutons.

McMillan
and

McLemore
(2004).

16

APPROACH


This
study
evaluates
whether
ca.
0.5
Ga
detrital
zircons
in
the
Permian
BCFM

were
derived
from
the
Florida
Mountain
granite.

To
accomplish
this
goal,
I
(1)
used

CL
images
on
all
zircon
grains
to
compare
textural
characteristics
from

compositional
zoning;
(2)
obtained
U-­‐Pb
ages
and
trace
element
compositional
data

for
zircons
from
the
FM
granite
using
the
LA-­‐ICPMS
at
Boise
State
and
(3)
conducted

additional
analyses
on
detrital
zircon
samples
(by
LA-­‐ICPMS)
from
the
middle
to

upper
BCFM
to
further
evaluate
the
detrital
zircon
population
of
the
BCFM
over
time.


These
data
provide
three
independent
criteria
–
(1)
textural
characteristics,
(2)
U-­‐
Pb
age
and
(3)
trace
element
compositions–
for
comparing
zircon
crystals
in
the

Florida
Mountain
granite
to
ca.
0.5
Ga
detrital
zircons
in
the
BCFM.

17

BACKGROUND


U-­‐Pb
Geochronology
and
Trace
Element
Compositions
of
Zircon


U-­‐Pb
geochronology
is
a
robust
dating
technique
and
the
mineral
zircon

(ZrSiO4)
is
an
ideal
U-­‐Pb
chronometer.

The
U-­‐Pb
isotopic
system
is

geochronologically
powerful
because
two
isotopes
of
U
decay
to
two
different

isotopes
of
Pb
(238U206Pb
and
235U207Pb),
providing
two
independent

radioactive
clocks
(Gehrels,
2012).

The
two
isotopes
of
U
have
different
half-­‐lives

(235U207Pb
has
a
half-­‐life
of
704
million
years;
238U206Pb
has
a
half-­‐life
of
4.47

billion
years)
making
the
U-­‐Pb
system
useful
over
most
of
Earth’s
history
(Harley

and
Kelly,
2007).



The
mineral
zircon
is
an
ideal
U-­‐Pb
geochronometer
because
when
a
crystals

grows,
U
is
incorporated
into
the
zircon
structure,
but
Pb
is
excluded.


The

substitution
of
uranium
into
the
structure
makes
the
initial
U-­‐Pb
ratio
extremely

high,
and
measured
ratio
of
U-­‐Pb
isotopes
is
directly
related
to
time

(Harley
and

Kelly,
2007).

In
addition,
zircon
is
an
abundant
accessory
mineral
in
both
igneous

and
metamorphic
rocks
and
resistant
to
physical
and
chemical
weathering,
making

it
an
ideal
tracer
of
sediment
transport
(Fedo
et
al.,
2003).

The
trace
element
composition
of
a
zircon
crystal
is
related
to
its
atomic

structure,
the
bulk
composition
of
the
magma
or
fluid
from
which
the
crystal
grew

and
the
crustal
conditions
(T,
P)
during
growth.

Zircon
is
a
tetragonal
orthosilicate,

characterized
by
the
structural
formula:

VIIIZrIVSiO4.

18

When
zircon
crystallizes
from
a
melt,
the
crystal
incorporates
cations
other
than

zirconium
and
silicon.
The
substitutions
that
occur
are
dictated
primarily
by

Goldschmidt’s
Rules
(size,
charge,
electronegativity
of
competing
cations)
and
the

pressure-­‐temperature
conditions
during
growth.

Thus,
the
relative
abundances
of

specific
trace
elements
can
be
used
as
proxies
for
the
source
rocks
from
which

detrital
zircons
originated.


As
mentioned
previously,
large
cations
with
a
4+
charge,
such
as

U4+,
Th4+,

and
particularly
Hf4+
readily
substitute
into
the
VIII-­‐fold
site
and
are
thus
relatively

abundant
in
typical
zircons.

Ti4+
typically
occurs
in
lower
concentrations
because
of

its
smaller
ionic
radius,
but
the
concentration
of
Ti4+
is
T-­‐dependent
and
serves
as
a

geothermometer.

Most
rare
earth
elements
(REE)
occur
in
the
crust
as
3+
cations

and
their
radius
varies
systematically
with
atomic
number
(Harley
and
Kelly,
2007).


The
heavy
REE
readily
substitute
for
VIIIZr4+,
but
the
light
REE
tend
to
be

incompatible.
To
compensate
for
the
charge
imbalance
created
when
a
REE

substitutes
for
VIIIZr4+,
a
coupled
substitution
occurs
with
the
following
scheme:

VIIIZr4+
IVSi4+



VIII(Y3+,
REE3+)
IVP5+.

The
coupled
substitution
involving
Y
and
P
is
common
and
forms
an
incomplete

binary
solid
solution
between
zircon
and
xenotime
(YPO4),
which
are
isostructural

despite
their
contrasting
chemical
composition.

Cations
with
a
5+
charge,
like
Nb5+

and
Ta5+,
also
substitute
for
Zr4+
in
the
VIII-­‐fold
site,
but
these
substitutions
must
be

coupled
to
substitution
of
a
3+
cation
(e.g.,
Fe
3+)
into
the
tetrahedral
site.

When
cations
substitute
into
zircon,
zones,
which
represent
potions
of
the

zircon
that
are
enriched
or
depleted
in
certain
cations,
are
often
created
within
the

19

crystal.

The
type
of
zoning
that
occurs
within
a
crystal
can
be
helpful
in
interpreting

a
zircons
crystallization
history
(Harley
and
Kelly,
2007).

For
example,
fine

oscillatory
zoning
is
typically
diagnostic
of
igneous
source.
These
zones
can
be

imaged
using
cathodoluminescence
(CL)
and
backscattered
electron
(BSE)

microscopies.

Zones
that
appear
bright
on
CL
images
tend
to
be
enriched
in
heavy

rare
earth
element
(HREE)
concentrations
and
relatively
low
U
and
Th

concentrations
because
HREE
emit
energy
while
U
and
Th
absorb
energy
(Hanchar

and
Westrenen,
2007).



Mineral
inclusions
can
impact
the
age
or
trace
element
composition
obtained

during
an
analysis
and
hence,
avoiding
inclusions
is
an
important
consideration

when
choosing
grains
and
locations
on
the
grain
to
analyze.

Minerals
that
typically

occur
in
zircon
crystals
include
apatite
[Ca5(PO4)3(F,Cl,OH)],
monazite
[(Ce,La)PO4],

rutile
(TiO2),
xenotime
(YPO4),
and
Fe
oxides.

If
the
laser
ablates
an
inclusion
during

analysis
of
a
zircon
crystal,
the
elements
associated
with
that
mineral
are
enriched

beyond
typical
values
for
zircon.

So
while
it
is
important
to
try
to
avoid
inclusions

prior
to
analysis,
analyses
contaminated
by
mineral
inclusions
are
simple
to
identify

and
reject
after
analysis.

20

METHODS

Rock
Sampling
in
the
Field

Samples
from
the
BCFM
were
collected
by
Cantine
et
al.
(2013)
from
the

locations
shown
in
Appendix
A.


Stratigraphically,
the
samples
span
the
oldest

portion
of
the
BCFM
to
the
top
of
the
Middle
BCFM
(Figure
11).

All
samples
were

taken
from
large
sand
channels
as
seen
in
Appendix
A.

The
sand
channels
sampled

are
composed
of
well
sorted,
clast
supported,
medium
to
coarse
grained
quartz

arenites.

The
sand
channels
represent
high-­‐energy
flow
events
during
sea
level
low

stands
and
hence
include
rip-­‐up
clasts
from
older
siltstone
units.
High
energy
flow

events
likely
produced
well-­‐mixed,
hom*ogenous
sediment
populations.

Our

sampling
strategy,
collecting
and
analyzing
two
samples
per
channel
(e.g.,
BCMC12-­‐
3A
and
3B)
was
designed
to
evaluate
the
hom*ogeneity
of
the
detrital
zircon

populations
and
whether
our
laboratory
procedures
resulted
in
sampling
bias.


Samples
BCMC12-­‐1A,
1B,
and
4A
were
analyzed
by
Cantine
et
al.
(2013).

Samples
BCMC12-­‐3A
and
3B
were
analyzed
for
this
study
to
provide
greater

stratigraphic
coverage
of
the
BCFM
and
to
better
evaluate
the
temporal
persistence

of
the
ca.
0.5
Ga
source.

Zircon
separates
from
one
sample
of
the
Florida
Mountain
granite
(sample

08FM-­‐10
in
Amato
and
Mack,
2012)
in
southeastern
New
Mexico
were
generously

provided
by
Jeff
Amato
at
New
Mexico
State
University.

21

Stratigraphic
Height

UPPER

3A
&
3B
(this
study)
MIDDLE

LOWER

1A
&
1B
(Cantine
et
al.,
2013)
4A
(Cantine
et
al.,
2013)

Figure
11.

Schematic
stratigraphic
column
showing
Lower,
Middle
and
Upper

BCFM
with
relative
locations
of
Cantine
et
al.
(2013)
samples.

Table
1.
Sampling
locations
for
this
study
and
Cantine
et
al.
(2013).

Sample

Latitude


Longitude


BCMC12-1A

31°50'53.52"

104°50'24.70"

BCMC12-1B

31°50'52.10"

104°50'23.84"

BCMC12-3A

31°51'45.72"

104°50'20.70"

BCMC12-3B

31°51'46.14"

104°50'18.42"

BCMC12-4A

31°50'32.51"

104°50'33.50"

22

Sample
Preparation

All
sample
processing
and
analytical
methods
were
conducted
at
Boise
State

University
in
the
laboratory
of
Dr.
Mark
Schmitz
and
in
collaboration
with
Dr.
Jim

Crowley.

Zircon
separates
were
obtained
from
samples
of
the
BCFM
by
standard

crushing
(jaw
crusher
and
disc
mill),
water
table,
heavy
liquid
(bromoform,
ρ=
2.89

g/cm3)
and
magnetic
(Frantz)
separation
techniques.
The
least
magnetic
zircon

separate
was
annealed
at
900°C
and
1
atm
for
60
hours.
Annealing
the
grains
serves

two
purposes;
it
repairs
radiation
damage
to
the
crystal
lattice
and
increases
the

amount
of
luminescence
which
helps
when
the
zircons
are
imaged
using

cathodoluminescence
(Mattinson,
2005).




Approximately
140
zircon
crystals
per
sample
were
hand
selected,
to
ensure

representation
of
all
grain
types
by
the
largest,
most
inclusion-­‐
and
crack-­‐free

crystals.

Selected
grains
were
separated
into
two
size
groups
prior
to
mounting
in

epoxy.

By
mounting
grains
of
similar
size
in
the
same
mount,
we
ensured
that
when

the
mount
was
ground
down
and
polished,
similar
depths
were
reached
within
each

crystal.

Therefore,
textural
characteristics
reflecting
more
of
the
grain’s
growth

history
were
exposed
for
imaging.


Polished
mounts
were
imaged
on
a
JEOL
scanning
electron
microscope
using

both
back-­‐scattered
electron
(BSE)
and
cathodoluminescence
(CL)
detectors.

The

utility
of
imaging
zircon
crystals
prior
to
LA-­‐ICPMS
is
twofold.

First,
BSE
and
CL

images
can
be
used
to
document
multiple
growth
domains,
cracks
and
inclusions

which
helps
geologists
avoid
sampling
spots
that
will
lead
to
an
erroneous
ages
or

unreliable
chemical
compositions
(Harley
and
Kelly,
2007).

Secondly,
BSE
and
CL

23

images
provide
textural
information
about
crystals
which
allows
first
order

conclusions
to
be
drawn
on
whether
the
detrital
zircons
were
derived
from
an

igneous
source
rock
or
a
metamorphic
source
rock
(Gehrels,
2012).

Every
zircon

crystal
was
imaged
prior
to
analysis
to
locate
areas
of
each
grain
that
were
devoid
of

inclusions,
cracks
or
multiple
growth
domains.

In
addition,
the
polished
grain

mounts
were
placed
under
a
petrographic
microscope
and
each
grain
was
observed

in
both
reflected
and
transmitted
light
to
locate
areas
and
volumes
that
were

inclusion-­‐
and
crack-­‐free.

This
step
was
not
performed
for
the
samples
analyzed
by

Cantine
et
al.
(2013)
and
resulted
in
a
higher
yield
of
usable
analyses
for
this
study.


After
analysis,
the
grains
were
re-­‐imaged
to
record
exactly
where
each
grain
was

analyzed.

LA-­‐ICPMS

Zircons
were
analyzed
by
laser
ablation
inductively
coupled
plasma
mass

spectrometry
(LA-­‐ICPMS)
using
a
Nb:YAG
laser
(1024
cm-­‐1)
and
a
quadropole
mass

analyzer
to
obtain
U-­‐Pb
ages,
rare
earth
element
(REE)
concentrations
and

concentrations
of
other
elements
dominant
in
zircon’s
structure.

Zircons
were

ablated
with
a
spot
size
of
30
μm
(fluence=
5
J/cm2
and
pulse
rate=
10
Hz).

Ablated

material
was
carried
to
the
argon
torch
by
a
helium
gas
stream.

Prior
to
each

analysis,
background
counts
were
taken
and
were
subtracted
from
the
total
counts

during
analysis.
Additional
analytical
details,
particularly
for
trace
element
analyses,

are
presented
in
Appendix
B.



Each
analysis
was
evaluated
for
quality.

Plots
of
analytical
sweeps
were

inspected
for
the
presence
of
inclusions
(count
rate
spikes
in
P,
Ti,
Nb,
Y,
Ce,
etc.)

24

and
reverse
fractionation.

The
data
were
then
evaluated
for
unusual
REE
patterns,

anomalous
concentrations
of
individual
trace
elements
and
U-­‐Pb
discordance.

25

RESULTS

Brushy
Canyon
Formation

CL
images
for
zircon
grains
from
the
BCFM
show
a
wide
variety
of
textures

including
fine
oscillatory
and
sector
zoning,
patchy
resorption
features,
multiple

growth
domains,
inherited
cores,
metamorphic
overgrowths
and
recrystallization

(Figure
12).

These
textures
provide
clues
into
the
igneous
or
metamorphic
origin
of

detrital
zircon
grains
and
thus
provide
a
basis
of
comparison
to
zircons
from
the
FM

granite.

As
expected,
the
BCFM
contains
grains
exhibiting
a
variety
of
textural

characteristics
and
zoning
types
in
BSE
and
CL
images.

30
microns

Figure
12.
CL
images
of
8
BCFM
detrital
zircons
from
sample
BCMC12-­‐3A.

Note
the

wide
variations
in
textures
and
the
presence
of
cracks
in
the
grains
in
the
top
right

and
bottom
left
corners.

The
circular
features
visible
on
five
grains
are
the
pits

created
by
LA-­‐ICPMS
analysis.

These
pits
show
the
portion
of
the
grain
analyzed.

26

U-­‐Pb
ages
were
measured
on
295
BCFM
from
samples
BCMC12-­‐3A
and

BCMC12-­‐3B,
47
of
which
were
rejected
during
quality
control.
Both
visual

examination
and
Kolmogorov-­‐Smirnov
(K-­‐S)
analysis
indicate
that
both
samples

yield
statistically
indistinguishable
age
spectra.
The
K-­‐S
test
is
a
statistical
measure

of
the
similarity
of
spectral
datasets,
such
as
DZ
age
probability
spectra.
The
K-­‐S
test

produces
P-­‐values
that
provide
a
basis
of
comparison,
such
that
P-­‐values
<
0.05

indicate
two
spectra
statistically
different
at
the
95%
confidence
interval
(Guynn

and
Gehrels,
2006).

The
age
distribution
spectra
for
samples
BCMC12-­‐3A
and
3B

yield
a
P
value=0.571.

This
value
is
well
above
the
threshold
value
of
0.05,

suggesting
hom*ogeneity
of
the
zircon
population
within
the
sand
channel
sampled,

affirming
the
laboratory
methods
employed
to
reduce
sampling
bias,
and
permitting

treatment
of
these
results
as
a
single
population
of
zircons
(n=248).
The
combined

age
spectrum
for
samples
BCMC12-­‐3A
and
3B
are
statistically
indistinguishable

from
the
age
distribution
of
detrital
zircons
from
samples
lower
in
the
stratigraphic

sequence
(samples
BCMC12-­‐1A,
1B
and
4A;
Hawkins
unpublished),
yielding
a
K-­‐S

test
P-­‐value
of
0.654.

These
statistical
results
permit
treating
the
results
from
all

five
samples
from
the
BCFM
as
a
single,
hom*ogeneous
population
(n=551)
of

detrital
zircon
ages
(figure
13).




The
age
spectrum
for
the
BCFM
is
characterized
by
prominent
peaks
at
ca.

0.5
Ga,
1.1
Ga,
1.4
Ga,
1.7
Ga,
1.9
Ga
and
2.7
Ga.

Peaks
at
ages
greater
that
0.8
Ga
are

typical
of
North
American
continental
crust
(Gehrels
et
al.
2011;
Soreghan
and

Soreghan,
2013).

However,
the
dominant
peak
at
ca.
0.5
Ga,
representing
about

48%
of
all
grains,
is
unusual
for
North
America
and
likely
represents
a
local
or

27

regional
source
rather
than
sediment
transported
from
distal
North
American

sources.

Figure
13.
Normalized
probability
age
distributions
for
samples
from
the
BCFM.


Note
the
close
similarity
of
the
age
and
magnitude
of
peaks
in
each
sample
and
the

similarity
of
the
combined
spectrum
to
the
spectra
of
samples
from
different

stratigraphic
levels
in
the
BCFM.

28

Since
the
goal
of
this
study
is
to
evaluate
the
FM
granite
as
a
source
for
BCFM

sediment,
in
this
section
I
describe
the
zoning
patterns
and
compositional

characteristics
of
ca.
0.5
Ga
zircon
grains
from
all
five
BCFM
samples
analyzed
by
the

Hawkins
research
group
over
the
past
18
months.

Figure
14
shows
a
sampling
of
ca.

0.5
Ga
BCFM
detrital
zircons
with
a
variety
of
zoning
patterns
including
fine

oscillatory,
coarse
oscillatory,
resorption
features,
potential
sector
zoning
and

potential
patchy
zoning.

Of
the
n=246
ca.
0.5
Ga
BCFM
zircons,
at
least
20%
are

consistent
with
an
igneous
origin
based
on
the
oscillatory
zoning
patterns.

An

additional
40
to
60%
of
grains
show
potential
sector
zoning
and
may
also
be

igneous
in
origin.

Figure
14.
Nine
BCFM
ca.
0.5
Ga
zircons
displaying
the
variations
in
zoning
patterns.


Note
the
significant
variations
in
luminescence.

The
4+
cations
that
freely
substitute
into
the
zircon
structure
show
order-­‐of-­‐
magnitude
variations
in
concentrations.

Specifically,
the
U
concentrations
range

29

from
27
to
700
ppm
with
most
values
falling
between
100-­‐350
ppm,
Th

concentrations
range
from
9
to
308
ppm
with
most
values
falling
between
40-­‐110

ppm
and
Ti
concentrations
range
from
2
to
25
ppm
with
most
values
falling

between
5-­‐15
ppm.
Th/U
ratios
vary
from
0.17
to
1.08.

The
5+
cations
Nb
and
Ta,

which
require
coupled
substitution,
also
show
order
of
magnitude
variations,
but

the
Nb/Ta
ratio
for
the
BCFM
zircons
shows
limited
variation.
Phosphorus,
which
is

coupled
to
variations
in
large
3+
cation
concentrations
(e.g,
HREE
and
Y),
ranges

from
65
to
357
ppm
with
most
falling
between
130-­‐200
ppm.
Yttrium

concentrations
range
from
275
to
4767
ppm
with
most
zircons
falling
between
700-­‐
1900
ppm.


Figure
15
shows
the
chondrite-­‐normalized
REE
patterns
for
ca.
0.5
Ga
BCFM

zircons.

The
steep
REE
patterns,
enriched
in
HREE,
exhibit
(Lu/Nd)cn
ratios

ranging
from
52
to
3739,
with
most
values
falling
between
100-­‐400.
Concentrations

of
Lu
range
from
14
to
179
ppm
with
most
grains
falling
between
30-­‐80
ppm.

The

chondrite
normalized
europium
anomaly
ranges
from
0.01
to
0.50
with
most
falling

between
0.05
and
0.15.


The
trace
element
composition
of
the
BCFM
detrital
zircons
is
limited

compared
to
the
compositional
variation
of
zircon
from
continental
crust,
a
topic

which
is
expanded
upon
in
Discussion.

30

Figure
15.
REE
plot
of
all
ca.
0.5
Ga
BCFM
grains.

Note
the
order
of
magnitude

ranges
in
concentrations
as
well
as
significant
variations
in
slopes
and
in
shape
of

anomalies.

31

Florida
Mountain
Granite

The
FM
granite
zircon
grains
are
euhedral
and
exhibit
distinctive
sector

zoning
in
both
BSE
and
CL
images
(Figure
16).
Typical
sector
zoned
grains
are

characterized
by
bright
hour-­‐glass-­‐shaped
domains
in
sharp
contact
with
darker

domains.

Most
grains
contain
abundant
inclusions,
especially
devitrified,
irregularly

–shaped
melt
inclusions,
acicular
needles
of
apatite,
and
equant
monazite
grains.

30
microns

Figure
16.
Post
laser
ablation
CL
image
of
FM
granite
zircons.

Note
the
distinctive

sector
zoning
and
clear
presence
of
inclusions.

U-­‐Pb
ages
were
obtained
on
77
laser
spots
on
FM
granite
zircons,
of
which,

32
were
rejected
during
quality
control.

The
35
analyses
that
passed
quality
control

yield
an
age
of
517
±
3/11
Ma
(random
errors/systematic
errors)(figure
17).

This

32

age
is
indistinguishable
from
the
ages
determined
Amato
and
Mack
(2012)
on
two

samples
analyzed
by
SHRIMP
and/or
LA-­‐ICPMS
in
(Figure
18).

The
trace
element
composition
of
FM
granite
zircon
is
more
restricted
than

that
of
the
BCFM
detrital
zircons,
as
expected
for
a
sample
of
igneous
rock.


Cations

with
a
4+
charge
show
much
more
consistency
than
in
the
BCFM
ca.
0.5
Ga
zircons

with
the
U
concentration
ranging
from
59
to
192
ppm,
the
Th
concentration
ranging

from
36
to
172
ppm
and
the
Ti
concentration
ranging
from
6
to
10
ppm.

Th/U

ratios
vary
from
0.6
to
0.94.

The
5+
cations
also
show
tight
clusters
with
the
Nb

concentration
ranging
from
6
to
20
ppm
and
the
tantalum
concentration
ranging

from
3
to
8
ppm.
Phosphorus
ranges
from
137
to
236
ppm.


Figure
19
shows
chondrite-­‐normalized
REE
patterns
for
all
FM
granite

zircons.

Although
the
(Lu/Nd)cn
ratio
varies
from
125
to
656
ppm
,
the
Yb/Gd
ratio

shows
limited
variation
ranges
from
9.7
to
16.0.

Lu
concentrations
range
from
36
to

116
ppm.

There
is
a
suggestion,
particularly
in
the
LREE
portion
of
the
patterns,

that
there
may
be
two
compositionally
distinct
groups
of
zircons,
although
this
may

be
an
artifact
of
the
small
number
of
analyses.

33

Figure
17.
Age
distribution
of
Florida
Mountain
granite
(this
study).

Note
the
mean

age
of
517
±
11
Ma
including
both
internal
and
external
error
and
the
MSWD.

Figure
18.
Florida
Mountain
granite
ages
from
this
study
compared
to
FM
granite

ages
from
Amato
and
Mack
(2012).

Note
the
agreement
in
ages
between
all
samples.

34

Figure
19.
REE
plot
of
Florida
Mountain
granite.

Note
that
the
concentrations
do

not
vary
much,
particularly
when
compared
to
the
variations
in
the
BCFM
REE
plot.

35

DISCUSSION

The
BCFM
age
spectrum
is
characterized
by
prominent
peaks
at
ca.
0.5
Ga,

1.1
Ga,
1.4
Ga,
1.7
Ga,
1.9
Ga
and
2.7
Ga.
Peaks
older
than
0.8
Ga
are
typical
of
North

American
zircon
populations
derived
from
the
Grenville
orogen
(1.1
Ga),
the

Yavapai
orogen
(1.7
Ga),
the
Laurentian
province
(>2.7
Ga)
and
from
anorogenic

granites
(1.4
Ga)
intruded
throughout
North
America
(Figure
2).

However,
the
age

of
the
dominant
peak
at
ca.
0.5
Ga,
comprising
48%
of
the
BCFM
zircons,
is
not

typical
of
North
American
crust.

In
addition,
the
magnitude
and
the
limited
age

range
of
ages
beneath
the
0.5
Ga
peak
suggest
that
these
grains
were
derived
from
a

proximal
source.

A
more
distal
source
would
have
a
broader,
lower
magnitude
peak

because
long
transport
distances
lead
to
mixing
(averaging)
of
sediments
derived

from
large
regions
and
varied
source
rocks.

As
shown
in
Figure
9f,
granite
plutons

in
the
New
Mexico
Aulacogen
(described
by
McMillan
and
McLemore,
2004),
such
as

the
FM
granite,
are
similar
in
age
to
detrital
zircons
from
the
BCFM.


The
FM
granite
and
the
McClure
Mountain
syenite
are
two
plutons
from
the

New
Mexico
Aulacogen
for
which
U-­‐Pb
ages
are
available,
and
will
therefore
be
used

to
represent
this
suite
of
plutons.

The
FM
granite
yields
an
age
of
517
±
11
Ma

including
both
internal
and
external
error
(this
study)
which
is
nominally
older
than

the
ca.
0.5
Ga
peak
in
the
BCFM
age
spectrum,
although
they
overlap
within
error

(Figure
20).

The
McClure
Mountain
Syenite
yields
an
ID-­‐TIMS
age
of
524.0
±
0.1
Ma

(Schoene
and
Bowring,
2006)
which
is
statistically
indistinguishable
from
the
FM

granite
(this
study),
but
older
than
the
FM
granite
ages
determined
by
Amato
and

Mack
(2012).

Moreover,
the
highly
precise
age
of
the
McClure
Mountain
syenite

36

falls
on
the
older
shoulder
of
the
age
distribution
for
BCFM
detrital
zircon
(Figure

20).

These
results
demonstrate
that
the
age
of
the
FM
granite
and
McClure

Mountain
syenite,
as
exposed
today,
are
within
analytical
uncertainty
of
the

apparent
ages
of
ca.
0.5
Ga
detrital
zircon
from
the
BCFM.

However,
despite
the

similarity
of
ages,
the
FM
granite
and
the
McClure
Mountain
syenite
do
not
account

for
the
entire
distribution
of
ages
observed
in
BCFM
detrital
zircons,
particularly

those
defining
the
younger
portion
of
the
age
distribution
peak.

The
overlap
in
ages
between
two
plutons
from
the
New
Mexico
Aulacogen

and
the
ca.
0.5
Ga
zircons
in
the
BCFM
suggests
a
genetic
link,
but
also
highlights
the

need
to
consider
additional
criteria
when
identifying
the
provenance
of
sediments.


Trace
element
compositional
data
provide
an
additional
point
of
comparison.

Given

the
relationship
between
the
zircon
crystal
structure
and
the
substitution
schemes

of
cations,
cation
ratios
provide
a
useful
basis
for
comparing
different
zircon

populations.
Since
it
is
difficult
to
fractionate
cations
with
the
same
charge
and

similar
size,
ratios
between
them
(e.g.,
Th/U
and
Nb/Ta)
provide
potential

fingerprints
for
source
rocks
of
zircons,
and
a
basis
for
comparing
zircon

populations.



The
trace
element
composition
of
zircons
from
the
FM
granite
overlap
with

the
composition
of
ca.
0.5
Ga
detrital
zircons
from
the
BCFM
(Figure
21).

For
all

elements
and
elemental
ratios
considered,
spanning
all
cation
charges
(3+,
4+,
5+)

and
a
wide
range
of
ionic
radii,
the
composition
of
FM
granite
zircons
is

indistinguishable
from
the
composition
of
ca.
0.5
Ga
detrital
zircons
from
the
BCFM.


Moreover,
these
plots
demonstrate
that
both
the
ca.
0.5
Ga
BCFM
detrital
zircons

37

and
the
FM
granite
zircons
are
compositionally
similar
to
>0.8
Ga
zircons
derived

from
North
American
crust.

Figure
20.
Comparison
between
ca.
0.5
Ga
zircons
from
BCFM,
McClure
Mt.
Syenite

and
the
FM
granite.
Note
that
U-­‐Pb
ages
determined
by
LA-­‐ICPMS
for
the
FM
granite

from
Amato
and
Mack
(2011)
and
from
this
study
include
both
random
and

systematic
errors.

The
McClure
Mt.
Syenite
U-­‐Pb
age,
determined
by
ID-­‐TIMS,

overlaps
with
the
BCFM
U-­‐Pb
age
but
is
offset
from
the
peak
of
the
BCFM
detrital

zircon
ages.

38

A

B

39

C

D

40

E

Figure
21
a-­‐e.
Y
is
held
constant
as
the
x-­‐axis
as
it
behaves
similarly
to
the
HREE.


Note
the
consistent
overlap
between
BCFM
ca.
0.5
Ga
grains
and
the
FM
granite
in

all
plots.

Also
note
that
most
points
fall
within
the
North
American
zircon
bubble.


All
data
points
encompass
analytical
uncertainty.

North
American
zircon
data

comes
from
Crowley
(unpublished)
and
Hawkins
(unpublished)
which
were
all

analyzed
at
Boise
State
University
by
LA-­‐ICPMS.
Additional
references
in
(a)
are

modified
after
Grimes
et
al.
(2007).

The
substantial
overlap
in
compositions
suggests
compositional
similarities

between
ca.
0.5
Ga
BCFM
detrital
zircons
and
FM
granite
zircons
which
can
be

further
evaluated
by
examining
the
textural
characteristics
revealed
in
BSE
and
CL

images.

The
Nb/Ta
and
the
Th/U
were
used
as
proxies
to
cull
ca.
0.5
Ga
BCFM

zircons
that
have
similar
trace
element
compositions
to
the
FM
granite
for
further

evaluation
using
CL
images.

Approximately
20%
of
ca.
0.5
Ga
BCFM
zircons
overlap

with
the
FM
granite
for
both
the
Th/U
and
Na/Ta
ratios.

Of
the
grains
that
overlap,

60
-­‐
80%
exhibit
textural
characteristics
that
could
be
consistent
with
those
of
the

41

FM
granite.
The
broad
range
between
60
-­‐
80%
reflects
the
uncertainty
of

characterizing
the
texture
of
detrital
grains,
which
only
preserve
a
portion
of
their

growth
history
due
to
abrasion
and
fragmentation
during
sediment
transport..


Figure
22
shows
examples
of
inconsistent
and
potentially
consistent
zoning
patterns

in
BCFM
detrital
zircons
compared
to
FM
granite
zircons.

Figure
22.
(A)
Shows
FM
granite
zircons
for
comparison
to
(B)
ca.
0.5
Ga
BCFM

zircons
with
similar
trace
element
compositions
and
similar
zoning
to
the
FM

granite
and
(C)
ca.
0.5
Ga
BCFM
zircons
with
similar
trace
element
compositions
but

different
zoning
than
the
FM
granite.
Note
that
these
grains
are
rounded
due
to

sediment
transport,
and
that
growth
zones
are
truncated
at
the
margins
of
the

grains.
All
grains
are
shown
at
the
same
scale.

42

Out
of
all
ca.
0.5
Ga
zircons
from
the
BCFM,
approximately
20%
have
fine

oscillatory
zoning
that
typically
reflects
growth
from
silicate
melt
and
therefore
an

igneous
source
rock.

An
additional
40
-­‐
60%
of
the
BCFM
detrital
grains
exhibit

textures
that
could
be
consistent
with
an
igneous
source.
Circa
0.5
Ga

BCFM
zircons

exhibiting
fine-­‐oscillatory
zoning
span
the
compositional
range
of
all
ca.
0.5
Ga

BCFM
zircons
(Figure
23).

Moreover,
the
composition
of
these
grains
overlaps
with

the
composition
of
grains
from
the
FM
granite
.

These
results
indicate
that
the
ca.

0.5
Ga
detrital
zircons
from
the
BCFM
are
closely
allied
to
the
FM
granite
and

perhaps
other
plutons
from
the
New
Mexico
Aulacogen
igneous
suite.

A

43

B

Figure
23
a-­‐b.

Trace
element
plots
comparing
FM
granite
to
ca.
0.5
Ga
BCFM

zircons
with
fine
oscillatory
zoning
indicating
an
igneous
source
and
to
all
BCFM
ca.

0.5
Ga
zircons.
Note
that
the
FM
granite
falls
within
the
bound
of
ca.
0.5
Ga
igneous

BCFM
zircons.

Together
the
U-­‐Pb
ages,
trace
element
compositions
and
textural

characteristics
suggest
that
although
the
FM
granite
cannot
account
for
all
ca.
0.5
Ga

zircons
in
the
BCFM,
the
source
of
the
ca.
0.5
Ga
zircons
was
similar
to
the
FM

granite
in
both
age
and
composition.

Moreover,
the
Ouachita
Mountains,
the

Appalachians,
and
the
Acatlan
igneous
provinces
are
the
only
other
areas
that

contain
ca.
0.5
Ga
rocks
in
the
region,
all
of
which
can
be
ruled
out
as
the
source
of

ca.
0.5
Ga
zircons
in
the
BCFM
based
on
ages
alone
(Figure
9).

When
taken
together,

these
results
suggest
that
the
rift
related
plutons
of
the
New
Mexico
Aulacogen
and

the
correlated
Oklahoma
Aulacogen,
or
clastic
sediments
derived
from
these
sources,

44

are
the
likely
source
of
ca.
0.5
Ga
zircons
in
the
BCFM.

These
results
are
consistent

with
the
eolian
transport
directions
of
the
Permian
as
discussed
by
Soreghan
and

Soreghan
(2013)
with
the
New
Mexico
Aulacogen
lying
to
the
northwest
and
the

Oklahoma
Aulacogen
lying
to
the
east
of
the
Delaware
Basin
(Figure
7).

The
conclusion
that
ca.
0.5
Ga
zircons
in
the
BCFM
were
likely
derived
from

the
New
Mexico
Aulacogen
supports
the
initial
hypothesis
that
the
ca.
0.5
Ga
zircons

can
be
accounted
for
by
local
sources
while
the
remainder
of
the
BCFM
age
spectra

can
be
accounted
for
by
the
orogenic
belts
of
North
America.

Taken
together,
these

results
demonstrate
that
the
peak-­‐matching
methodology
can
be
enhanced
by

considering
trace
element
compositions
and
textural
characteristics
as
additional

criteria,
particularly
for
resolving
local
versus
more
distal
sediment
provenance.


Ultimately,
by
precisely
pinpointing
the
sources
of
sediment,
better
constraints
can

be
placed
on
the
regional
paleogeography
of
ancient
continents.