Portland State University

PDXScholar
Dissertations and Theses Dissertations and Theses

1985

Provenance study of late Eocene arkosic sandstones in southwest

and central Washington
Mark Edward Byrnes
Portland State University

Let us know how access to this document benefits you.

Follow this and additional works at: http://pdxscholar.library.pdx.edu/open_access_etds
Part of the Geology Commons, and the Stratigraphy Commons

Recommended Citation
Byrnes, Mark Edward, "Provenance study of late Eocene arkosic sandstones in southwest and central Washington" (1985).
Dissertations and Theses. Paper 3405.

This Thesis is brought to you for free and open access. It has been accepted for inclusion in Dissertations and Theses by an authorized administrator of
PDXScholar. For more information, please contact [email protected].
AN ABSTRACT OF THE THESIS OF Mark Edward Byrnes for the Master

of Science in Geology presented June 10, 1985.

Title: Provenance Study of late Eocene Arkosic Sandstones

in Southwest and Central Washington.

APPROVED BY MEMBERS OF THE THESIS COMMITTEE:

Paul E. Hammond, Chairman

The purpose of this study is to compare the sandstone composition

and trace element geochemistry between samples representing the Summit

Creek sandstone, Naches, Chumstick, and Carbonado Formations in order to

determine if these sediments were all derived from the same provenanc~,

and to determine the composition of the source rocks in hopes to

identify the present day location of the source areas.

The light mineral portion of this study reveals that the Summit

Creek sandstone, Naches, Chumstick, and Carbonado Formations are all

 2

arkosic to lithic arkosic in composition, and were derived from the

erosion of a silicic plutonic source terrane. The rather small percen-

tage of K-feldspar (<12%) found in these sandstones suggests that these

sediments were not derived from rocks of true granitic composition, but

rather from plutonic rocks of lower potassium content. The polycrystal-

line quartz and rock fragment content indicate a major low grade meta-

morphic (quartz-biotite schist), and mafic volcanic source rock,

contributing sediments to the finer fractions of these rocks.

All four formations are characterized by the heavy minerals

epidote, apatite, zircon, garnet, sphene, rutile, tourmaline, hornblende

and hypersthene. The Naches, Chumstick, and Carbonado Formations also

contain small percentages of zoisite and augite, and the Carbonado

Formation contains traces of staurolite. This heavy mineral composition

indicates that the coarser fraction of these sediments were derived

chiefly from plutonic source rocks.

The Summit Creek sandstone, Chumstick, and Carbonado Formations

have a similar geochemical composition indicating a common provenance.

The average trace element similarity coefficient existing between these

three formations is 0.99, where a 1.0 indicates perfect similarity. The

younger Naches Formation has a somewhat unique chemistry, and shows only

a 0.72 similarity coefficient with the other three formations. The

significantly higher concentration of Cr, and Sc, and the lower concen-

tration of La in the Naches Formation suggests that it contains a larger

mafic or ultramafic component to its sediments than the other three

formations.

The Mount Stuart area, and the Chelan Mountain terrane are the
 3
most likely source areas for these sediments, since they can account for

all of the rock types needed to comprise these sediments. Derivation of

sediments from these two source areas is also in accordance with south-
westerly current directions.
PROVENANCE STUDY OF LATE EOCENE ARKOSIC SANDSTONES IN
SOUTHWEST AND CENTRAL WASHINGTON

by

MARK EDWARD BYRNES

A thesis submitted in partial fulfillment of the

requirements for the degree of

MASTER OF SCIENCE
in
GEOLOGY

Portland State University

1985
TO THE OFFICE OF GRADUATE STUDIES AND RESEARCH:
The members of the Committee approve the thesis of Mark

Edward Byrnes presented June 10, 1985.

Paul E. Hammond, ~nairman

rvin H. Beeson

APPROVED:

Paul E. Hammond; Head, Department or Geology

Jim Heath, Dean of Graduate Studies and Research

 ~'

ACKNOWLEDGEMENTS

I would like to thank J.H. Fouret, A.V. Lewis, Jr., and George H.
Rudkin from Shell Western, Union 76, and Marathon Oil Companies for
choosing to provide financial support for this thesis.
I would like to thank my advisor Dr. Paul E. Hammond, and Warren
Winters for their guidance, encouragement, and support not only for this
thesis, but also for their support throughout my stay at Portland State
University.
I would also like to thank the other two members of my thesis
committee, Dr. Marvin H. Beeson, and Dr. Robert Van Atta, for always
finding time in their busy schedules to provide me with direction, and
support.
Drs. Michael L. Cummings, and Ansel Johnson pushed me from the
first day I set foot in this geology department. I am deeply indebted
to them both for providing me with the drive, to strive to reach my
limits as a professional in the field of geology.
I would especially like to thank my wife Karen who has always
provided me with love, encouragement, and who has always believed in me.
- I would also like to acknowledge Chris Burke, Gene Pierson, Phil
and Janet Amundson, Krista McGowan, Mike Pollock, Rod Swanson, Matt
McClincy, Tom Ullery and Tom Gable for their support.
I would like to dedicate this thesis to my mom and dad (Elfie and
Frank Byrnes) who have always provided me with emotional, financial, and
loving support.
 TABLE OF CONTENTS

PAGE

ACKNOWLEDGEMENTS. . . • . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . iii

LIST OF TABLES. . . . . . • . . . . • . . • . • . . . . . . . . • . . . . . . . . • • . . . • • . • . . • • . vi

LIST OF FIGURES. . . . . . . . • • . . . . . . . . . . . . • . . . • . . . . • . . . . . . . • . . • . • . . vii

CHAPTER

I INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . 1

Purpose of study. • • . . . • • . . • . . . . . . . . • . • . . . . • • . • . • . • • 1

Methods of investigation. . . . . . . • . • • . . . . • . . . . . . . . . . . 3

Location and access ......•..•..•.......••.......•.. 5

Previous work. . . . • . . • . . . • • . . . . . . . . . . . . . . . . . . . . • . . • . 7

Regional geology ..•..••.....•.....•....•..••.••.•.. 11

II SEDIMENTARY PETROLOGY. . . . . . . . . . • . . • • . • . . . . . • . . . • . • . . . 13

Light minerals .•.••..••......................•.•... 13

Comparative mineralogy with correlative arkosic

sandstone units •.......•.........•..••............. 21

Heavy minerals ••..••.••.•...••.•••••••••••..•••••.. 26

Comparative heavy mineralogy with correlative

arkosic sandstones ...•......•......•.•...•.......•• 26

III GEOCHEMISTRY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
 v
Comparative geochemistry with correlative
Cowlitz Formation •••...•.•.....••.....•.•.•.•.•.... 32

IV DISCUSSION AND CONCLUSIONS ........................... 33

Light r1inerals. . . . . . . • • . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Heavy Minerals. . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . 3S

Geochemistry. . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

V REGIONAL SIGNIFICANCE OF STUDY ....................... 42

Possible source rocks .............................. 47

VI S~RY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S2
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SS
APPENDIX A: TRACE ELEMENT DATA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
APPENDIX B: SAMPLE LOCATION SITES ............................ 64
 LIST OF FIGURES

FIGURE PAGE
1. Location of formations under study ........•............. 2

2. Sample location sites ................................... 6

3. Chronostratigraphic correlation chart between

formations ..•........................................... 9

4. Ternary diagrams showing light mineral composition

of samples from each of the four formations .....•.•..... 15

5. Chronostratigraphic correlation chart between

correlative sandstone units ............................. 22

6. Ternary diagrams showing mean framework grain

composition between correlative sandstone units ......... 23

7. Ratio diagrams showing variation in framework grain

composition between correlative sandstone units ......... 24

8. Cluster analysis showing geochemical similarity

between the formations .................................. 30

9. Geochemical data plots showing variation in chemical

chemical composition between formations ................. 31

10. Map showing location of correlative sedimentary units ... 43

11. Map showing location of possible source areas ...•....... 48

 LIST OF TABLES

TABLE PAGE

I SUMMARY OF THE CHARACTERISTICS OF THE FORMATIONS

UNDER STUDY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
II RELATIVE PERCENTAGES OF LIGHT MINERALS IN SAMPLES
FROM THE FOUR FORMATIONS UNDER STUDY .................. 14
III RELATIVE PERCENTAGES OF HEAVY MINERALS IN SAMPLES
FROM THE FOUR FORMATIONS UNDER STUDY .................. 27
 CHAPTER I

INTRODUCTION

PURPOSE OF STUDY

This study concentrates on the Summit Creek sandstone, Chumstick

Formation, Naches Formation, and Carbonado Formation in central and

southwestern Washington (Figure 1). These deposits are Eocene in age,

light gray to pale brown in color, arkosic to lithic arkosic in

composition, and all except the Carbonado Formation - which was

deposited in a near shore marine environment - were deposited in

nonmarine fluvial and braided stream environments (Snavely and others,

1951; Gard, 1968; Gresens and others, 1981; Clayton, 1983; Tabor and

others, 1984). The purposes of this study are: 1) to compare these

four formations in terms of their light and heavy mineral assemblages

and trace element contents, in order to determine if these sediments

were derived from the same provenance; and 2) to determine the

composition of the source rocks based on the light and heavy mineral

assemblages, in hopes of identifying the sauce areas.

 2

122° 120°
:----1---------------..1-.---------4~
WASHINGTON

48°

N~
~~ ·:~--c~
...
..

~
47°

C
~ &SS Yakima
•
46°

OREGON I
_____._ 4 s0
0 100 Km

Figure 1. Map of the state of Washington showing the location of

the Summit Creek sandstone (SS), Naches (N), Chumstick (CH), and
Carbonado Formations (CB).
 3

METHODS OF INVESTIGATION

The fieldwork entailed collecting samples representative of the

Summit Creek sandstone, Chumstick Formation, Naches Formation, and the

Carbonado Formation. Samples were preferably collected from the basal

1 m of fine-grained, thinly bedded, sandstone units, because Carver (1971)

found heavy minerals to be more concentrated in these types of finer

grained rocks.

These samples were first dissaggregated using a kerosene and hot

water treatment. Samples were soaked in kerosene for several hours,

after which time the kerosene was poured off and replaced with boiling

water. The boiling water causes the kerosene, still held within the

pore spaces of the rock, to expand as it vaporizes. This expansion

forces a separation between matrix and clasts and thus disaggregates the

sample. This treatment works best to disaggregate samples that are

predominantly clay cemented. However, since several samples from each

formation were partly cemented with calcite, they were first soaked in a

dilute HCl solution before being treated with kerosene and hot water.

The disaggregated samples were then split in half. Half of the

sample was prepared for instrumental neutron activation analysis, while

the remainder of the sample was used in the light and heavy mineral

study.

The samples for the petrographic study were sieved to separate out

the 2 to 3 ¢ size fraction, then split into approximately 10 gram

samples. These samples were then weighed, placed in 50 ml test tubes,

and then centrifuged in tetrabromoethane (sp. gravity 2.96 at 25 C) for

20 minutes. The sediment was stirred once during the centrifuging to

 4

make it easier for the finer heavy minerals to settle out. The light and

heavy minerals were then mounted on separate petrographic slides in

Lakeside 70 cement. The light mineral slides were ground with 400 grit

to expose the grains for plagioclase and potassium feldspar staining.

This staining procedure required the slides first to be etched in fumes

of hydrofluoric acid for up to 1 minute. The slides were then dipped

into sodium cobaltinitrate for 30 seconds, dipped in water, and then

dried. They were then dipped into barium chloride for 8 to 10 seconds,

dried, dipped into red dye #3 for 45 seconds, rinsed in water as before

and finally dried. The slides where then point counted using the 'Line

Method' (Carver, 1971).

The samples for the geochemical study were first washed, crushed

to a powder in a mortar and pestle, split and weighed, and then placed

in small plastic polyvials. The samples were then irradiated for 1.0

hours in the TRIGA Mark I reactor at Reed College at 250 kw power. The

samples were then counted twice using gamma ray spectroscopy. The first

count was run as soon as the dead time fell below 10%. This count was

important in detecting elements with short half-lives. The second count

was run one month after irradiation, this count avoids interference from

elements with relatively short half-lives. From the first and second

count data, elemental abundances were determined in terms of percentages

and parts per million. This data was then plotted on graphs, and entered

into similarity and cluster analysis programs.

 5
LOCATION AND ACCESS

The four study areas can be found on the Packwood, Leavenworth,

Easton, and Morton 15' quadrangle maps of the State of Washington

(Figure 2).

The Summit Creek sandstone is exposed along Summit Creek which is

located approximately 12 miles northeast of the town of Packwood. This

section is approximately 900 m thick and can be found in the NEl/4 of

section 14 and the NWl/4 and NEl/4 of section 13, T.14 N., R.10 E., of

the Packwood 15' quadrangle (Clayton, 1983). Although the base of this

section is not exposed, the sandstones overlie Eocene basaltic lava

flows. The Summit Creek sandstone is conformably overlain by rocks of

the Ohanapacosh Formation.

The Chumstick Formation is well exposed along highway US 2 between

the towns of Wenatchee and Winton. This section is 5,800 m thick

(Whetten, 1976); it is unexposed at its base and is overlain uncon-

formably by rocks of the Wenatchee Formation.

The upper portion of the Naches Formation is exposed along logging

roads in the Quartz Creek drainage, located approximately 25 miles

northwest of the town of Naches. This section is approximately 1,500 m

thick (Tabor and others, 1984), and is exposed in the NEl/4 of section

21 and the SWl/4 and SEl/4 of section 22, T.18 N., R.14 E., of the

Easton 15' quadrangle. This section is underlain by the Rhyolite of

Mount Clifty (Tabor and others, 1984), and is overlain by the upper tuff

member of the Naches Formation which may be correlative with the Steven

Ridge Formation (Hammond, 1980).

The Carbonado Formation is well exposed along Snow Creek which can
 7

be found in the SWl/4 of section 12, the SEl/4 and SWl/4 of section 11,

and in the NEl/4 of Section 10, T.13 N., R.4 E.. The total exposed
thickness of this formation is more than 1200 m. It is unexposed at its

base and is overlain by volcanic rocks of the Northcraft Formation.

PREVIOUS WORK

Previous workers have mapped a number of Eocene arkosic units in

southwest and central Washington, including the Puget Group (Gard, 1968;

Buckovic, 1979), Roslyn Formation (Bressler, 1951), Mcintosh Formation

(Snavely and others, 1958), Carbonado Formation (Gard, 1968), Chumstick

Formation (Gresens, 1981), Naches Formation (Frizzell, 1979; Hammond,

1980), Skookumchuck Formation (Snavely and others, 1958), Summit Creek

sandstone (Clayton, 1983), Renton Formation (Vine, 1962), Tiger Mountain

Formation (Vine, 1962), and the Spiketon Formation (Gard, 1968).

A summary of the characteristics of the four formations is shown

in Table 1. The chronostratigraphic relationships between these

formations is shown in Figure 3.

Buckovic (1979) summarized the stratigraphic relationships between

various Eocene arkosic sedimentary units mapped by previous workers in

west-central Washington, including the Puget Group, Renton, Tiger

Mountain, Spiketon, Carbonado, Skookumchuck and Mcintosh Formations. He

believes that the similar arkosic to lithic arkosic composition of these

sediments, as well as the dominant westerly current directions collected

from the channel sandstone deposits, indicates that these formations

represent different facies of one large 'Puget Deltaic System'. He

found the majority of the Puget Group to consist of an extensive

 TABLE I

DESCRIPTION OF FORMATIONS

FORMATION SOURCE AGE TH LI1'110LOGY

SUMMIT CREEK HAMMOND (1980)1 LATE 900m (+IInterstratified gray to pale brown
SANDSTONE CLAYTON (1983)~0CENE micaceous, arkosic sandstone, siltstone,
1'0 shale, sporadic thin coal beds, and
~ARLY rhyolitic tuff beds.
)LlG-.
OCENE

NACHES FRIZZELL (1979) LATE

I 11400 m Dominantly arkosic sandstones and
FORMATION HAMMOND (1980) EOCENE siltstones near the top, with volcano-
clastic sandstones, siltstones, basaltic
and rhyolitic lava flows and rhyolitic
tuffs interstratified with arkosic.rocks in
the middle and lower portions of the
formation.

CHUMSTICK BUZA (1979) MIDDLE f'5800 m Arkosic conglomeratic sandstone, sandstone,

FORMATION GRESENS (19 8 l) TO and shale with tuffaceous interbeds.
(1983) J..ATE
EOCENE

CARBONADO GARD (1968) EARLY ~370 m Interstratif ied gray to pale brown
FORMATION Hl\MMOND (1980) 1'0 nicaceous arkosic sandstone, siltstone, and
MIDDLE ~hale with thin coal beds, interfingered
EOCENE towards tile top with volcanic rocks of the
Northcraft Formation.

00
 9

n.y.
Pierce and King summit Crk. Naches River Leavenworth
Counties r- 35
Area Area Area SERIES

<
I I
I
I
!
I
I
i
I
I I Wenatchee
,..1
I al
c
I I 1 Formation al I ll
> I ~I'
tJ
0
I ...... O'
( 7l ?----...f I
I
·~
......

)
I ·~·111 I c
Ohanapecosh I Ohanapecosh
Fm.
( Fm. I I I i i
,..

µ?~
I al
I Q.
) I
Q.
:s
t- 40
7 I ? I

Spiketo11 !Renton Summit Naches

Fm. Fm. Chums tick
~I
I '1'
Creek Fm. Fm.
·~ sandstone ll
c
.~
'1'
I al
tJ
=
:s
I
0
tz:I
...
g. I QJ

:ti'1' I I"" 45
,..
0
C.!> I ·"1
a
.j.J

~
Fm.
I Tiger
,5! !Mountain
i Fm.
I--;~;ed.- -r
base not
"'-?
Rhyolite
of
~r
Riv~r ;m.,
-- - -?---- I
Mount
Clifty
50

Figure 3~ Chronostratigraphic correlation chart (modified after

Winters, 1984). Dashed vertical lines between columns indicate
lithostratigraphic equivalence; solid line indicates uncertain
lithostratigraphic relationship. References for local columns
are: 1, Armentrout and others, 1983; 2, Buckovic~ 1974; 3, Gard,
1968; 4, Mullineaux, 1970; 5, Vine,1969; 6, Clayton, 1983; 7,
Tabor and others, 1984.
 10

facies of nonmarine sandstone, siltstone, mudstone, and coal. Marginal

marine and marine facies are less common, particularly in the lower

portions of the Puget sequence. Alternating sandstone, siltstone,

mudstone, and coal sequences are interpreted as distributary and

interdistributary channel deposits of the delta plain facies. The thin-

bedded marginal marine strata of the lower Carbonado Formation and upper

Skookumchuck Formations are considered to be lagoon, or bay~like, inter-

distributary deposits (Buckovic, 1979). The upper Mcintosh Formation

and lower Skookumchuck Formation are interpreted as delta front facies,

while the lower portion of the McintQsh and the middle part of the

Skookumchuck Formations may represent prodelta facies of the Puget

Deltaic system. These interpretations were supported with mega- and

micro-faunal data.

Frizzell (1979) studied the framework grains of Paleogene non-

marine arkosic sandstone units in western Washington. He collected

samples from the Puget Group, Chuckanut, Swauk, Chumstick, Manastash,

and Naches Formations, and found all six of these sandstone units to be

either feldspathic or lithofeldspathic subquartzose in composition

(terminology after Dickinson, 1970). Relating his findings to previous

work by Dickinson (1970), Frizzell concluded that the total quartz

content in these sediments indicates a combination of plutonic and

magmatic arc provenances. Winters' (1984) study of the framework

grains of the Chambers Creek beds in southwest Washington revealed that

these sandstones range in composition from arkose to lithic arkose, and

are dominantly composed of subangular monocrystalline quartz and

feldspar. Rock fragments comprise from 3 to 19 percent of these rocks,

 11

and are dominantly of the metamorphic, and volcanic variety, with

smaller amounts of sedimentary and granitic rock fragments. Plotting

the results from this study along with results collected from previous

workers - on the composition of the Cowlitz, Chumstick, Naches, Renton,

Roslyn, Spiketon, Skookumchuck, and Spencer Formations - Winters

attempted to compare the provenances between these nine sedimentary

units. The results revealed two compositionally distinct,

geographically separated groups. The Chumstick, Roslyn, and Naches

Formation contain a significant amount of coarsely polycrystalline

quartz (Qp/Q=.30) and are relatively low in potassium feldspar

(P/F=.85). In contrast the Chambers Creek beds, and the Skookumchuck,

Cowlitz, and Spencer Formations all contain very little polycrystalline

quartz (Qp/Q<.30), and notable amounts of K-feldspar (P/F=.60).

REG ION AL GEOLOGY

The Cascade Range of Washington can be subdivided into two main

geologic provinces~ separated along the northwest-trending Olympic-

Wallowa lineament near Snoqualmie Pass, east of Seattle. The northern

Cascade province consists primarily of a metamorphic-plutonic crystal-

line core, which is fault-bounded on the east and west by pre-Tertiary

marine and continental sedimentary rocks. The crystalline core rocks

consist of low to high ranking regionally metamorphosed rocks of pre-

middle Devonian to Jurassic age, and plutonic intrusions of late

Triassic to Tertiary age (Misch, 1966).

The southern Cascade province consists primarily of Cenozoic vol-

canic and fine-grained intrusive rocks, Tertiary continental and marine

 12

sedimentary and volcanoclastic deposits, and Tertiary granitic plutons.

Additional Tertiary continental and marine deposits. along with

Holocene glacial deposits and alluvium, are found west of the Cascade

Range in the Puget Lowlands.

The volcanic and sedimentary strata in the four study areas are

deformed chiefly along northwest trends. Dips are as steep as 78

degrees overturned in the Summit Creek sandstone area; generally dips

range from 20 to 35 (+S) degrees.

 CHAPTER II

SEDIMENTARY PETROLOGY

LIGHT MINERALS

The light minerals, which are defined in this study as those min-

erals which have specific gravities less than tetrabromoethane (specific

gravity 2.96 at 25 C), were point counted using the 'Line Method' of

Carver (1971). The results from this analysis are listed in Table 2,

and are plotted in ternary diagrams in Figures 4A-E.

These results indicate that all four of the formations are charac-

terized by sandstones that are arkosic to lithic arkosic in composition,

with feldspar being more abundant than quartz. Plagioclase comprises

between 30 and 50 percent of these rocks, and thus is the most abundant

mineral in all of the formations. Plagioclase is most abundant in the

Naches and Chumstick Formations.

Quartz i~ present in both a monocrystalline ~nd polycrystalline

form. Monocrystalline quartz comprise between 15 and 38 percent of

these rocks, while the polycrystalline form comprises between 1 and 9

percent. The monocrystalline grains are generally equant in shape and

often show strain shadows through 5 degrees of stage rotation. The

 14

TABLE II

RELATIVE PERCENTAGES OF LIGHT MINERALS

total total total

Sample Qm PQl PQ2 ~Q3 Q p K F LV Lm Ls L Cht B M
Sn3 36 2 2 3 43 25 10 35 8 4 2 14 2 2 l
~ Snl9 30 l 2 6 39 36 5 41 7 8 l 16 l 3 T
0
'O
IO
c
Sn23 21 1 3 3 28 38 ll 49 s 14 T 19 1 2 -
.s 5n38 26 l 2 3 32 33 5 38 10 8 T 18 T 7 2
M
IO
u
Sn47 14 2 l - 17 48 12 60 7 8 T 15 1 6 l
Sn58 18 2 4 l 25 42 12 S4 6 4 T 10 l ll T
• 54 30 3 2 1 36 36 3 39 7 13 T 20 T 3 T
en
ui SS 49 l l - Sl 29 2 31 6 10 T 16 - 2 T
~ S6
CV
29 l 2 1 33 50 - so 4 7 l 12 T 4 1
~ 510 27 T 2 T 29 so 2 S2 1 10 T ll l 7 T
.., Sll
·~ 29 1 1 l 32 51 l 52 2 7 1 10 - 5 1
I Sl2
en
21 2 2 - 2S 48 3 Sl 2 13 l 16 l 6 2
Sl2TK 22 l 4 1 28 49 l so 2 ll 1 14 - 6 2
N3 21 4 3 T 28 38 ll 49 s ll l 17 T 5 l
N7 16 2 T T 18 46 11 S7 4 8 1 13 - 12 T
~ Nl2 15 1 1 l 18 48 4 S2 5 12 2 19 - 9 T

!
112
Nl4 20 4 l - 25 37 9 46 2 10 3 lS - 13 l
~ NlS 19 2 l T 22 43 5 48 4 14 2 20 T 9 T
z
r1122 19 l 4 l 25 47 5 52 3 ll l 15 - 7 T
~i23 18 4 3 l 26 40 10 so 4 13 2 19 - s T
Chl 19 l 4 2 26 44 3 47 9 8 l 18 - 6 2
l: Ch8 19 T 3 - 22 48 6 54 3 5 l 9 - 13 2
~
tJ
.....
Chl6 24 - l T 25 37 10 47 5 13 l 19 - 7 1
~ Chl8 18 l 4 l 24 S4 3 S7 4 6 l 11 - 5 l
] Ch26 30 T s 2 37 36 ll 47 2 10 l 13 - 2 T
u
~h29 21 T 1 1 23 43 7 so 5 11 l 17 - 9 T
 15

am a

A
A£J
AA
A

p F L
Lm P01

Lv Ls PQ2 P03

( c ) = mean composition

Figure 4A. Ternary diagrams showing the composition of framework

grains for seven samples from the Summit Creek sandstone (A).
Qm, quartz, monocrystalline; P, plagioclase; F, total feldspar;
Lm, lithic fragments, metamorphic; Lv, lithic fragments,
volcanic; Ls, lithic fragments, sedimentary; PQl, quartz,
polycrystalline, containing greater than 10 crystals/grain; PQ2,
quartz, polycrystalline, containing less than 10 and greater than
3 crystals/grain; and PQ3, quartz, polycrystalline, containing 3
crystals/grain or less.
 16

Qm Q

.:a.•
p K F L
Lm PQ1

Lv Ls PQ2 PQ3

( c) = mean composition

Figure 4B. Ternary diagrams showing the composition of framework

grains for seven samples from the Naches Formation (•). (refer
to Figure 4A for abbreviations)
 17

Om
a

0
0
ooo o

p K F
L
Lm
PQ1

Lv Ls
PQ3

( c ) = mean composition

Figure 4C. Ternary diagrams showing the composition of framework

grains for six samples from the Chumstick Formation (o). (refer
to Figure 4A for abbreviations)
 18

Om a

AA
~A

A
•

p K F
L
Lm
P01

Lv Ls PQ2 PQ3

(c ) = mean composition

Figure 4D. Ternary diagrams showing the composition of framework

grains for six samples from the Carbonado Formation (•). (refer
to Figure 4A for abbreviations)
 19

Qm Q

p K F L
Lm P01

Lv Ls PQ2 PQ3

Figure 4E. Ternary diagrams showing the mean composition of

framework grains for the Summit Creek sandstone (A), Naches (e),
Chumstick (o), and Carbonado Formations (•). (refer to Figure 4A
for abbreviations)
 20
Polycrystalline grains generally show crystal units with polyhedral

outlines, smooth crystal-crystal boundaries, and interfacial angles of

120 degrees at triple junctions of crystal boundaries. Only one poly-

crystalline quartz grain was found to contain distinctly elongated

crystal units. PQl, PQ2, and PQ3 are three varieties of polycrystalline

quartz defined by the author. PQl and PQ2 identify those polycrystal-

line quartz grains which contain greater than ten and those less than

ten and greater than three crystals/grain, while PQ3 identifies grains

which contain 3 crystals/grain or less. PQ2 is the most abundant

variety of polycrystalline quartz found in all four formations, followed

by PQl and PQ3. PQl is found most abundant in the Naches formation,

whereas PQ2 and PQ3 are found most abundant in the Chumstick and

Carbonado Formations. The overall total quartz percentages are slightly

higher in the Carbonado Formation, and the Summit Creek sandstone than

in the Naches and Chumstick Formations.

Potassium feldspar is found in all of the formations, generally

making up between 1 and 11 percent of these rocks, and is dominantly of

the orthoclase variety. The Summit Creek sandstone contains the

smallest percentage of this mineral at 3 percent or less, where the

others contain near equal amounts.

Rock fragments are found to comprise as much as 21 percent of

these rocks. The most abundant rock fragments are of the metamorphic

quartz-biotite-schist variety. These fragments show a well defined

foliation, are fine grained, and can be distinguished from polycrystal-

line quartz by containing greater than 5 percent mica.

The next most abundant rock fragments are the mafic volcanic
 21

variety, which comprise between 1 and 10 percent of these rocks. These

fragments have been determined to be mafic on the grounds that they

contain no potassium feldspar or quartz. Sedimentary siltstones and

mudstones are rare. Biotite is the most abundant phyllosilicate

mineral comprising as much as 13 percent of these rocks. Muscovite and

chlorite are considerably less abundant making up less than 2 percent

and less than 1 percent of these rocks respectively. The biotite found

in the Carbonado Formation, and the Summit Creek sandstone is medium-

brown to dark-brown in color, while in the Naches and Chumstick

Formations, the boitite ranged in color from dark-brown to dark-brown-

green to yellow-green in color.

COMPARATIVE MINERALOGY WITH CORRELATIVE ARKOSIC SANDSTONE UNITS

In order to better understand the provenance, the overall size of

the depositional basin, and the depositional setting of the four forma-

tions under study, the results from this analysis were compared to

published modal data from the correlative middle to upper Eocene

Chambers Creek beds, Roslyn, and Renton Formations (Winters, 1984;

Frizzell, 1979), the Cowlitz, and Spencer Formations of northwest Oregon

(Jackson, 1983; Al-Azzaby, 1980), and the Skookumchuck Formation of

western Washington (Buckovic, 1974).

The stratigraphic relationships between these formations are shown

in Figure 5.

The framework grain compositions of the above mentioned formations

were added to ternary and secondary ratio diagrams previously construc-

ted by Winters (1984) (Figures 6 and 7).

 Clobd CMono-
at.ratl9caphic anlta
ISal udor •. lff l)
01l ll lon year a YAIQllLL COLIJHBIA so1m1 ru1ac CDITAAUl rJ EJICE MIO ltl HG .JOKHSOtl ~£5 UVER llOSLYN UAVDfNOtml
lunu"M CRIC.
be fol e pre 'e"~ COUii rt COIJllTf llllL'rA HIU.S AJILA COUNTIES CN. AllF.A AJILA AllEA Mn S£1UES
A•••
OR. OR. NA. Ill. ~. WA. WA.
lS ""· ""· ""· JS
I •.r.
(
rlttsLon9 Llncoln Creek Jover W••tern ) lovH' lfeatern
aluft ror . . tlon Caecade Group faecade Group: JI l
rornatlon IOh.tn•p•c:oah <.nn••1 (ot.- napeco d li! c::
fon••tlonl ;-rlne ro .... tlonl -H I I
•ar In• Iud.
seJl .. entu· (rocka I
rocke Ohen•peaoelh
) I ,.......
40 I ., .__ I co
l-1 ?--
-u
1 1 I ' Ch-bere M10che1 lloalyn C,,_•tlck
! I llenton Cre•k roa•atloa ror-mation roe.at.ion
I! I ••· bed• Su11urtll
c •••• i
I l 1 , ... delono

Ii I b;!..._ , _ TunaverCJI
H It; I '5
not roc.atlon
••p<>•ed
=I It••• not
'f••navar
.
ta .. hlll l• ro.-.atlon
!.
ro1aat.lon •
ltto~!::~n ---?---
........
I ""·

so Creec•nt
.......!
roraadon -so 50
m~.......ll~t• Vo lean lee ror ... tion
12,&,11 (111 (11) (II (II (II
n ..n:11.a1C1:S1 IJ, l) 12) 11.101 u. J ,4, ' · ' '

Fig11n~ l). Chrouosln.ltigraphic correlation chart (moclHiecl after

Winters, 19B4). Dashed vertical lines between coJunms indicate
lithostraligr·aphic equivalence; solid line jndicates uncertain
I itlioslrnl igraphic: relatjouship. Refen~nces for local columns
an~: I, Al-Azzahy, l<JBO; L, Armentrout and others, 1983; 3,
Buckovic, 1971•; la, C:anl, l'J()H; 5, M11llinea11x, 1970; 6, Rau, 1981;
7, Snavely and othen->, l<Vi8; 8, Tabor and others, 1984; 9, Vine,
l'Jt)'>; 10, Wei Is, 1981; 11, Winters, 198]; 12, Clayton, 1983.
N
N
 23

01'N

cow
0
ca
c£ N
C>-to

I ~ !_
F L Lv Ls
Om

F Lt p K

Figure 6. Ternary diagrams showing the mean composition of

framework grains between the Chambers Creek beds (CC),.Summit
Creek sandstone (SS), Carbonado (CB), Chumstick (CH), Cowlitz
(COW), Naches (N), Renton (RN), Roslyn (RO), Skookumchuck (SK),
and Spencer Formations (SP) (modified after Winters, 1984). Qm,
quartz, monocrystalline; Qp, quartz, polycrystalline; Q, total
quartz; P, plagioclase; K, potassium feldspar; F, total
feldspar; Lv, lithic fragments, volcanic plus metavolcanic; Ls,
lithic fragments, sedimentary plus metasedimentary; L, Lv + Ls;
Lt, Lv + Ls + Qp.
 24

CH N SS ca SP cow SK cc AN AO
I .so

+.29 t.36
.40

A I
+.16
+.21
1.12 + .20
+.15
~ .20
I .!.os !.06
0
Qp/Q

+.89 + + t
.83 .95
t~,
.. 1.0

.so

t
.79 ~

B I +.63
+ t.59
.56
.60 t.64 ~ .60

.40

P/F

t" f t
79
l .12
1'• ~ ~o
1.0

c I
.73 .75
1 I.oS
~ .60

f .38 +.30 t.3• t.36

~

~
.40

.20

. 0

Figure 7. Ratio of polycrystalline quartz to total quartz

(Qp/Q), plagioclase to total feldspar (P/F), and volcanic lithic
fragments to total lithic fragments (Lv/L) between formations
identified on preceding figure. Bars show 80% confidence
interval for data (modified after Winters, 1984).
 25
Although the framework grain compositions of the ten formations

show a considerable amount of variation between formations, only the

QpLvLs diagram in Figure 6 shows an obvious grouping pattern. In this

diagram, it appears that the Roslyn, Renton, and Chambers Creek beds

contain a significantly larger polycrystalline quartz concentration than

the other five formations. On the other hand, the Summit Creek

sandstone, Naches, Chumstick, Carbonado, and Chambers Creek beds contain

a significantly larger concentration of sedimentary and metasedimentary

rock fragments than the other three formations. The Chambers Creek beds

and the Skookumchuck Formation are unique in that they respectively

contain a considerably larger concentration of total quartz, and total

lithic fragments than any of the other eight formations. Plot C, in

Figure 7, shows by far the best grouping of formations based on

framework grain composition. This plot reveals three distinct popula-

tions of samples, The first population, which represents the Spencer,

Cowlitz, Skookumchuck, Renton, and Roslyn Formations, is characterized

by a relatively high volcanic lithic to total lithic fragment ratios

(averaging 0.72). The second population, which represents the Summit

Creek sandstone, Chumstick, Naches, and Carbonado Formations, is charac-

terized by intermediate volcanic to total lithic fragment ratios

(averaging 0.35), and finally the third population, which represents the

Chambers Creek beds, is characterized by very low volcanic to total

lithic fragment ratios (averaging 0.06). Plot A, in Figure 7 also

reveals three distict populations of samples, although these populations

are not as well defined as in Plot C. The first population in Plot A

includes the Renton and Roslyn Formations, this population is charact-

 26
erized by a relatively high polycrystalline quartz to total quartz ratio

(averaging 0.32 + 0.15). The second population includes the Summit

Creek sandstone, Chumstick, Naches, Carbonado, and Spencer Formations,

it is characterized .by an intermediate polycrystalline quartz to total

quartz ratio (averaging 0.16 + 0.1). The third population includes the

Cowlitz Formation and Chambers Creek beds, and is characterized by the

smallest polycrystalline quartz to total quartz ratio (averaging 0.06 +

0.05). Plot Bin Figure 7 shows only two populations of samples. One

population contains a relatively high plagioclase to total feldspar

ratio, which includes the Summit Creek sandstone, Chumstick, Naches,

Carbonado, and Roslyn Formations (averaging 0.85 + 0.1), and the other

contains a lower plagioclase to total feldspar ratio (averaging 0.60 +

0.2), and includes the Chambers Creek beds, Spencer, Cowlitz,

Skookumchuck, and Renton Formations.

HEAVY MINERALS

The heavy mineral assemblages were characterized by the presence

of epidote, apatite, zircon, garnet, sphene, rutile, tourmaline,

zoisite, hornblende, augite, hypersthene, and staurolite. The relative

abundances of these minerals are shown in Table III.

The results from this analysis indicate that all four formations

are basically characterized by the same heavy mineral populations, with

only minor variations in percentages between formations.

COMPARATIVE HEAVY MINERALOGY WITH CORRELATIVE ARKOSIC SANDSTONES

The most extensive heavy mineral study done on Eocene sandstones

 27
TABLE III

RELATIVE PERCENTAGES OF HEAVY MINERALS

Q)
Q) Q) c::
-g. Q)

Q)
0
:2
~ca
..
c::
0
0
a;
...c::ca
Q)
c::
.!0. =-;
Q)
:i
ca
E
3
s
c;; ~
...
Q)

:aca
.t::
c;
...
Q)
0.
-s
0
~
ca ~0
...en
0.
Q)
0.
ca N 0 Cl> ... g 0
N ~
:;:, >.
.t:: (0

7 16 20 8 2 2 6 - ~ 2 2 l 34 ~
I-
SN3

- - -
~
4 3 19 l4 2 6 lO 2 l 39 SN19

10 - 2S 23 - ' - - 4 l ..- 31 SN22

18 26 3 l - - - - - - 3 - 49 8 SN38

10 l8 14 lO - - - 4 T 8 - - 36 ~
Ill
SN47
- - - - - - ~
l9 24 12 s -
- - 88 0 SNSS
19 7 19 9 - - - - - 2 10 - 34
a;
•$4

6 27 13 ll - lS - - -9 - - 19 en SS
s s - - - - -
~
lS 20 lO 8 3 34 w S6
w·
a:
ll 21 23 l l s - - -
2 6 - - 16 0
I-
sio
- l - -
10

7
22

24
17 l l

29 a
l
6
3

- s - - -
6 l

- -
28

21
! ..
s11

S12TK
42 12 7 6 - - 4 6 l 6 - - 16
~
N3

37 3 16 3 6 - - 4 3 10 4 - l4
I-
<
~
N7

37 6 ll 8 8 4 3 - lS u.
l l s l a:
0
N12
46 3 3 3 ll 4 l lO - - - - 19 Cl'J
~·
N1S
33 s 2 9 l4 - - - - - - z<
2 35 0 N22
45 s 2 12 ll 2 l - - - -
l 21 N23.
26 - 10 l l lS 3 - s - - ~
2 2 23 CH1
19 - 10 17 ll l - - - ·~
l ll 3 27
I-
CHS
34 l4 3 4 10 3 s - - -
3 lS u. 9 CH16

30 l4 - 40 7 l - - - - l l 6
~
~
I-
CH18'
34 lO 6 23 13 l - - - - - ~3 9
Cl'J
.CH26

45 4 4 s 22 3 - s - - - - 12 0 Ch29
 28

in southwest and central Washington, was a study done by Bressler (1951)

on the Roslyn Formation. Heavy minerals were also noted in

Karachewski's (1983) study of the Oligocene Lincoln Creek Formation in

southwest Washington, and in Van Atta's (1971) study of the Cowlitz

Formation in northwest Oregon.

Bressler's (1951) petrographic study of the middle to upper Eocene

Roslyn Formation in central Washington revealed these sedimentary units

to contain heavy mineral assemblages which are dominated by the minerals

epidote, clinozoisite, apatite, garnet, sphene, tourmaline, zircon,

actinolite, and least abundant rutile. Karachewski (1983) found zircon,

garnet, sphene, and apatite in his study of the Lincoln Creek Formation

in southwest Washington, while Van Atta (1971) noted apatite, sphene,

epidode, tourmaline, garnet, hornblende, augite, actinolite, kyanite,

staurolite, rutile, andalusite, and least abundant glaucophane in his

study of the Cowlitz Formation of northwest Oregon.

 CHAPTER III

GEOCHEMISTRY

Samples for this portion of the study were analyzed using gamma

ray spectroscopy. The results from this analysis can be found in

Appendix A. The concentrations of the twelve trace elements Ce, Co, Cs

Eu, Fe, Hf, La, Na, Sc, Sm, Ta and Th were then entered into a similar-

ity matrix, and then plotted in the cluster analysis shown in figure 8.

The results from this cluster analysis indicate that there are two

distinct populations of samples. The first population includes samples

from the Summit Creek sandstone, Chumstick, and Carbonado Formations.

The samples within this population show trace element compositional

similarities averaging 0.99, where 1.0 indicates samples of perfect

similarity. The samples within the second population, which represents

the samples from the Naches Formation, show an average similarity of

0.95. However, the similarity between the two populations is only 0.72.

The results from this analysis were also plotted graphically in Figure

9, in hopes to better illustrate the compositional variation between

formations. From Figure 9 and Appendix A, it is evident that relative

to the Summit Creek sandstone, Chumstick, and Carbonado Formations, the

Naches Formation contains a significantly lower concentration of Ce, La,

 30
0.70 0.80 0.90 0.990 0.995 1.000
. I . 1 CH

"'
-· 18 SS
"'

.-·
4 CB

~
-· 5 CB

- 14 SS

--· 15 SS

..
-· 2 CH

-i- 3 CH

-
.
-· e CB

I
-· 13 SS

-· 7 CB

-~ 12 N

11 N

a N

9 N

.. 10 N
SIMILARITY

Figure 8. Cluster analysis showing the geochemical similarity

between various samples from the Summit Creek sandstone (SS),
Naches (N), Chumstick (CH), and Carbonado Formations (CB).
 31

6-t ~
3
•
Oc
51 I
•
La/Sm i • K
( %)
-\

I
~ 0

0
I •
31 l

10 20 30 40 l 2 3
Sc Na
(ppm) I (%)

200 ~ 7

~
6
•
1601
• Cl
5 0

0
(~::) i La/sm
•
I
801
.~

c
3

4 8 l 2 K 3 4
Th
(ppm) ( %)

Figure 9. Plots showing the geochemical variation between

the Summit Creek sandstone (~), Naches (e), Chumstick (0),
Carbonado (.&), and Cowlitz Formations (O).
 32

Ta and Th, and a higher concentration of Co, Cr, Fe and Sc. Of the

three more geochemically similar formations, the Chumstick Formation

contains a slightly higher concentration of Cr, Sm, and Hf then either

the Summit Creek sandstone, or the Carbonado Formation. The Summit

Creek sandstone, and the Carbonado Formation are nearly identical in

composition.

COMPARATIVE GEOCHEMISTRY WITH CORRELATIVE ARKOSIC SANDSTONES

The trace element data from Kadri's (1982) study of the Cowlitz

Formation was also plotted in Figure 9 against the trace element compo-

sition of the Summit Creek sandstone, Naches, Chumstick, and Carbonado

formations.

The results indicate that the Cowlitz Formation shows similar

La/Sm ratios to both the Chumstick, and Carbonado Formations. It also

shows similar concentrations of Sc and K to the other four formations.

However, the Cowlitz Formation shows considerably larger concentrations

of both Na and Th, and a slightly smaller concentration of Cr than do

the other three formations.

 CHAPTER IV

DISCUSSION AND CONCLUSIONS

LIGHT MINERALS

The results from the light mineral analysis indicate that the

Summit Creek sandstone, Naches, Chumstick, and Carbonado Formations are

virtually identical in composition in terms of their monocrysta1line

quartz, feldspar, and lithic fragments (Figure 4). There is some

degree of variation in the concentration of the three varieties of

polycrystalline quartz between formations; however, this variation may,

in part, be due to the rather small number of grains to which this plot

is based.

The arkosic to lithic arkosic composition of these sediments

suggests either a silicic plutonic or metamorphic source terrain. The

significantly higher percentage of feldspar than quartz in these rocks,

may suggest that the source areas were uplifted and eroded rapidly so as

to not allow sufficient time for a significant amount of these minerals

to either be physically or chemically broken down.

The rather small percentage of K-f eldspar in these rocks leads me

to conclude that the majority of these sediments were not derived from
 34
rocks of true granitic composition, since granitic rocks typically

contain greater than 30 percent K-feldspar (Blatt and others, 1980).

This rather low percentage of K-feldspar is more indicative of sediments

derived from tonalites, quartz-diorites, or high grade metamorphic

gneisses and schists. The lack of twinning in the feldspars may be due

in a large part to the fact that in grain mounts, plagioclase grains

generally lie on the (010) surface, which is the most common twin plane.

The rather small ratio of polycrystalline quartz to total quartz, is

common in sedimentary rocks since the numerous grain boundaries in the

polycrystalline quartz grains cause them to be less stable. The major-

ity of the polycrystalline quartz grains appear to be of the upper

greenschist facies metamorphic variety, since the majority of the poly-

crystalline quartz contains greater than three crystals per grain (Young,

1976; Blatt and others, 1980).

The rock fragment data also indicates a major medium to low grade

quartz-biotite schist, and mafic volcanic source area contributing

sediments to the finer fractions of these sedimentary rocks. The small

percentage of sedimentary rock fragments is typical of most sedimentary

rocks, since they generally are more easily broken down than metamorphic

and volcanic rock fragments.

The abundance of brown biotite (titanium-rich biotite), as opposed

to green biotite (titanium depleted biotite), in these formations

indicates that the majority of the biotite from these formations is

derived from a mafic volcanic, as opposed to a silicic plutonic source,

since titanium tends to be enriched in these types of rocks. The higher

concentration of green biotite in the Naches, and Chumstick Formations

 35

may indicate that these sediments are closer to a silicic plutonic

source than the Summit Creek sandstone and Carbonado Formation

In summary the light mineral data indicates a tonalite, or quartz-

diorite plutonic, or a high grade metamorphic source rock contibuting to

the coarser fraction of the sediments, with a medium to low grade meta-

morphic and mafic volcanic source rock contributing to the finer

fraction of these sediments.

HEAVY MINERALS

The results from this portion of the study indicate all four

formations under study to be characterized by the presence of the heavy

minerals epidote, apatite, zircon, garnet, sphene, rutile, tourmaline,

hornblende, and hypersthene. The Naches, Chumstick, and Carbonado

Formations also contain small percentages of zoisite and augite, and the

Carbonado Formation contains traces of staurolite.

The minerals apatite and zircon are common minerals of plutonic

igneous rocks, and granite pegmatites. They are also present although

slightly less common in contact and regional metamorphic rocks, and

volcanic rocks (Deer and others, 1966).

Rutile and sphene are both common accessory minerals of interme-

diate to acidic plutonic rocks, metamorphic schists and gneisses rich in

ferromagnesium minerals, as well as calc-silicate rocks and skarns (Deer

and others, 1966).

Hornblende is one of the most widely distributed minerals in

nature, being found in igneous rocks from ultramafic to salic in compo-

sition, regional metamorphic schists and gneisses, and volcanic rocks.

 36

However, this mineral is most characteristic of intermediate plutons

(Phillips and Griffen, 1982).

Augite is the common pyroxene of mafic igneous rocks, and is

especially characteristic of gabbros. This mineral is less common in

intermediate and silicic igneous rocks. Augite is an antistress mineral

and thus is generally not characteristic of metamorphic rocks (Phillips

and Griffen, 1982).

Hypersthene is a common constituent of ultramafic and mafic

plutonic igneous rocks, volcanics, and high-grade metamorphic rocks,

both regional and contact (Phillips and Griffen, 1982).

The tourmaline found in all four formations was predominantly of

the schorl-dravite (iron-rich) variety, evidenced by its common blue and

black color. This variety of tourmaline is most common of granitic

plutonic rocks, and granite pegmatites (Phillips and Griffen, 1982).

Epidote, staurolite and zoisite are all predominantly metamorphic

minerals. Staurolite is a common mineral of medium grade regionally

metamorphosed argillaceous sediments, and carbonate rocks. Epidote and

zoisite also principally occurs in regional metamorphic rocks where they

mark the change from the greenschist facies to the amphibolite facies.

They may also form as the result of the hydrothermal alteration of

plagioclase (Deer and others, 1966; Phillips and Griffen, 1982).

The garnet found in these rocks is predominantly of the iron-rich

almandine variety, evidenced by its reddish-pink color (Phillips and

Griffen, 1982). Garnet is predominantly a metamorphic mineral found in

upper greenschist through the granulite facies metamorphic rocks. It is

typically found in association with staurolite, and kyanite in upper

 37
amphibolite and granulite facies rocks. Garnet in most sedimentary

rocks is of the almandine variety (Phillips and Griffen, 1982).

From the heavy mineral data it appears that the coarser fraction of

these sediments was derived predominantly from plutonic as opposed to

high grade metamorphic source rocks. This is evidenced by the absence

of strictly high grade upper amphibolite and granulite facies meta-

morphic minerals such as sillimanite, andalusite, or kyanite. The only

minerals found in these sediments which are common in high grade

metamorphic rocks are garnet, hornblende, and hypersthene. However,

garnet can also be derived from metamorphic rocks with as low of a grade

as upper greenschist facies. Hornblende and hypersthene, although they

can be derived from high-grade metamorphic rocks, can just as easily

be derived from mafic and ultramafic plutonic, or volcanic rocks. The

heavy minerals zircon, rutile, sphene, and apatite are common heavy

minerals which can be derived from either plutonic, pegmatitic, or

greenschist to lower-amphibolite facies metamorphic rocks, and therefore

tell us little about the source rocks other than that they were not

high-grade metamorphic.

Staurolite and zoisite are both common heavy minerals of regional

metamorphic rocks. These minerals are derived from lower amphibolite

facies metamorphic rocks. Zoisite may also be derived from upper

• greenschist facies metamorphic rocks .

In summary absence of strictly high grade metamorphic minerals

such as sillimanite, andalusite, or kyanite may indicate that the garnet

found in these sediments is derived from rocks of lower metamorphic

grade, suggesting then that the coarser fraction of these sediments was
 38

derived predominantly from plutonic as opposed to high grade metamorphic

source rocks.

GEOCHEMISTRY

The two populations of samples which the cluster analysis clearly

defined in figure 8 suggest that the trace element composition of the

Naches Formation is significantly different than that of the other three

formations. The very strong geochemical similarity, averaging 0.99,

between the Summit Creek sandstone, Chumstick, and Carbonado Formations,

strongly supports these sediments being derived from the same

provenance. The somewhat unique chemistry of the Naches Formation on

the other hand may suggest one of several things. Since the Naches

Formation is the youngest of the four formations, its unique chemistry

may suggest: 1) a changing chemistry of the source rock with time; 2)

the unroofing of a new source; 3) some local source rock contributing

more sediment to this formation than to the others; or 4) the Naches

Formation is part of an entirely different drainage system than the

other three formations. The author tends to support the third

hypothesis that the rather unique chemistry of the Naches Formation is

as a result of some local source rock contributing more sediments to the

Naches Formation than to the other three formations, for reasons to be

discussed.

Recent studies have shown trace elements to be more concentrated

in some minerals as opposed others. By knowing which minerals tend to

concentrate which trace elements it is possible to draw generalizations

in terms of rock composition from trace element data. This technique is

 39
especially useful in attempting to determine the provenance of sedi-

mentary rocks. Important studies by Kaskin and Paster (1979) found that

rare earth elements appear to be transferred almost quantitatively into

elastic sediments, and that common sedimentary and metamorphic processes

do not significantly affect the rare earth element distribution in

sedimentary rocks. It has also become accepted that the average rare

earth element pattern in sedimentary rocks basically reflects the rare

earth element distribution in the continental crust (McLennan and

others, 1980).

The concentrations of the various trace elements in the

sedimentary rocks in this study reveals a significantly lower concentra-

tion of the Ce, La, Ta, and Th, and a significantly higher concentra-

tions of Co, Cr, and Sc in the Naches Formation as compared to the other

three formations. The Summit Creek sandstone, Chumstick and Carbonado

Formations, as revealed in the cluster analysis, show very similar

concentrations of the various trace elements, with the Chumstick

Fo~mation containing only a slightly higher concentration of Cr, Sm, and

Hf than the near geochemically identical Summit Creek sandstone, and

Carbonado Formation. Of these nine elements which show the most varia-

tion in concentration between samples, the elements Cr, La, and Sc were

selected for this portion of the study because they show the largest

amount of variation in concentration between rock types, and they are

less mobile than some of the other elements.

The element chromium, Cr, occurs as a trivalent ion, readily

substitutes for Fe3+ and Al3+, and does not go into solution easily. It

is typically precipitated from magmas at an early stage either in the

 40

form of spine!, or clinopyroxene, which are minerals common in ultrama-

fic rocks. Cr is typically depleted in less mafic rocks (Smith, 1972).

The element lanthanum, La, occurs in a tripositive oxidation

state, and occurs in highest concentration in rare earth minerals that

favor light rare earths such as monazite (Kay, 1972). The La concentra-

tion in average basalts is 17 ppm, and ranges over a factor of ten.

Alkaline oceanic basalts and most continental basalts have more La than

oceanic tholeiites and some continental diabases. The La concentration

in granites average 84 ppm, a value which is considerably higher

although more variable than for basalts (Kay, 1972).

The element scandium, Sc, is also a trivalent ion, which is cap-

tured widely as a trace constituent of silicate minerals. The process

of magmatic differentiation apparently leads to a maximum concentration

of Sc of about 30 ppm in rocks of basaltic composition. In more silicic

rocks the Sc content drops off to quite low levels (Curtis (1972)).

From this information, the significantly higher concentration of

Cr, and Sc, and a significantly lower concentration of La in the Naches

Formation appear to indicate that the Naches Formation has a significa-

ntly larger maf ic component to its sediments than the other three

formations. Although the light mineral study did not reveal this to be

the case, it is possible that the finer silt and clay sized fractions of

the Naches samples may contain a larger maf ic volcanic component than

the other three formations. This may explain the slightly darker color

of the Naches samples.

In summary, the strong geochemical similarity that exist between

samples from the Summit Creek sandstone, Chumstick, and Carbonado

 41

Formations strongly supports these sediments being derived from the same

provenance. The somewhat unique chemistry of the Naches Formation may

suggest some local source rock is contributing more sediments to the

Naches Formation than the other three formations. The significantly

higher concentration of Cr, and Sc, and lower concentration of La in the

Naches Formation indicates that this formation has a larger maf ic compo-

nent to its sediments than the other three formations .

 CHAPTER V

REGIONAL SIGNIFICANCE OF STUDY

In order to better understand the provenance, the overall size of

the depositional basin, and the depositional setting of the four forma-

tions under study, the results from this study were compared with

published data from the time correlative Chambers Creek beds, Roslyn,

Renton, Cowlitz, Spencer, and Skookumchuck Formations (Figure 10).

The seven framework grain plots in Figures 6 and 7, when looked at

as a whole, appear to indicate that there are three distinct populations

of formations. The first population, which shows the strongest similari-

ty in framework grain composition, includes the Summit Creek sandstone,

Chumstick, Naches, Carbonado, and Roslyn Formations. These formations

all contain relatively large mean concentrations of lithic fragments,

relatively small mean concentrations of sedimentary rock fragments, and

K-feldspar (Figure 6), and relatively high mean plagioclase to total

feldspar ratios (Figure 7). The Summit Creek sandstone, Chumstick,

Naches, and Carbonado Formations also contain very similar intermediate

mean polycrystalline quartz to total quartz, and volcanic to total

lithic fragment ratios. The Roslyn Formation contains a significantly

higher mean polycrystalline quartz to total quartz, and volcanic to

 43

122° i20°
49°

WASHINGTON

48°
I\ I
Seattle
•• o Q
R~

N~"' ~ 47°

JS V'SS
~cc
Yakima
•
I 46°
~

N
I

OREGON \ I
45°
0 100 Km

Figure 10. Map showing the location of the Summit Creek sandstone
(SS), Naches (N), Chumstick (CH), and Carbonado Formations (CB)
from this study, along with the time correlative Chambers Creek
beds (CC), Roslyn (RO), Renton (RN), Cowlitz (COW), Spencer (SP),
and Skookumchuck Formations (SK). (modified after Winters, 1984)
 44

total lithic fragment ratios than the other four formations; however, the

Roslyn Formation also shows the largest variation in concentration of

these constituents between samples within this formation.

The second population, which includes the Cowlitz, Skookumchuck,

Renton, and Spiketon Formations, is characterized by sediments with

relatively small concentrations of lithic fragments, relatively large

concentrations of K-feldspar (Figure 6), relatively small ratios of

plagioclase to total felspar, and a relatively high ratio of volcanic to

total lithic fragments (Figure 7). The only significant difference in

composition between these four formations is the relatively large

concentration of polycrystalline quartz in the Renton Formation.

The third population contains only the Chambers Creek beds. This

population is unique in its relatively large percentage of total quartz,

and sedimentary rock fragments, and its relatively small ratios of

polycrystalline quartz to total quartz, and volcanic to total lithic

fragments.

These three populations appear to indicate that the formations

that occur within a population are all derived from the same provenance,

while formations between populations are either derived from an entirely

different provenances, or there is at least a major source rock contrib-

uting sediments to one population and not to the other.

Comparing the heavy mineral data collected in this study with

Bressler's (1951) petrographic study of the Roslyn Formation, indicates

that these five formations show virtually identical heavy mineral

populations. Although the relative percentages of the heavy minerals

vary slightly between formations, they are all basically characterized

 45

by the presence of epidote, apatite, zircon, garnet, sphene, rutile,.and

tourmaline. The only minerals found in the Roslyn Formation which are

not noted in the other four formations are clinozoisite, and a small
amount of actinolite. On the other hand, the only minerals found in the

four formations in this study which are not likewise found in the

Roslyn Formation are hornblende, hypersthene, augite, zoisite, and a

trace of staurolite.

Actinolite, and clinozoisite are both metamorphic minerals common

in upper greenschist to lower amphibolite facies metamorphic rocks (Deer

and others, 1966). Since these minerals occur in such small concentra-

tions in the Roslyn Formation, it is very possible that these minerals

are also present in the Summit Creek sandstone, Naches, Chumstick, and

Carbonado Formations, and were overlooked. These two minerals are at

least consistant with the metamorphic grade of the minerals zoisite and

staurolite from this study.

The Oligocene Lincoln Creek Formation, in southwestern Washington,

also shows a nearly identical heavy mineral assemblage to the four

formations in this study (Karachewski, 1983). The zircon, garnet,

sphene, and apatite noted in Karachewski's (1983) study, are all

abundant in the Summit Creek sandstone, Naches, Chumstick, and Carbonado

Formations. Since garnet is the only predominantly metamorphic mineral

noted in Karachewski's study, it is not possible to determine the grade

of metamorphic rocks contributing sediments to the Lincoln Creek

Formation.

The heavy mineral assemblage of the Cowlitz Formation (Van Atta,

1971) is different from the Summit Creek sandstone, Naches, Chumstick,

 46

Carbonado, Roslyn, or Lincoln Creek Formations, in that it contains

small percentages of kyanite, and andalusite which are high grade upper

amphibolite to granulite facies metamorphic minerals. The presence of

these two minerals indicates that a high grade metamorphic source rock

is contributing to these sediments. The presence of apatite, sphene,

epidote, tourmaline, garnet, staurolite, and rutile in the Cowlitz

Formation, however, is consistent with the other six formations. This

observation may indicate that the Cowlitz Formation is receiving

sediments from the same source areas as the Summit Creek sandstone,

Naches, Chumstick, Carbonado, Roslyn, and Lincoln Creek Formations,

along with sediments from some high grade metamorphic source.

Comparing the trace element composition of the Summit Creek

sandstone, Naches, Chumstick, and Carbonado Formations, with geochemical

data from the Cowlitz Formation (Kadris, 1982), reveals similar La/Sm

ratios, as well as Sc and K concentrations in these five formations

(Figure 9). However, the Cowlitz Formation shows a significantly higher

concentration of Th, and smaller concentration of Cr than the other four

formations. The element Th is concentrated in the late stages of magma-

tic crystallization (Moore, 1972), while Cr is concentrated in the early

stages (Smith, 1972), suggesting that the Cowlitz Formation has the

smallest mafic and ultramafic component to its sediments.

The results from this portion of the study indicates that

the Summit Creek sandstone, Naches, Chumstick, Carbonado, and Roslyn

Formations are compositionally very similar and thus are probably de-

rived from the same provenance. The Chambers Creek beds, Renton,

Cowlitz, Spencer and Skookumchuck Formations, on the other hand, are

 47

compositionally unique from the above formations, indicating that they

were at least partially derived from a different provenance.

POSSIBLE SOURCE ROCKS

The Mount Stuart area, the Chelan Mountain terrane, the Pasayten

River terrane, and the Idaho batholith are the four most probable source

terrains which contributed sediments to the four formations studied

(Figure 11).

The Mount Stuart batholith area lies just to the west of

Leavenworth, Washington. This batholith intruded the surrounding upper

Jurassic ophiolite melange of the Ingalls tectonic complex, and the

Chiwaukum Schist, in late Cretaceous time (Miller, 1982; Tabor and

other, 1984). Tonalite and granodiorite dominate the composition of

this batholith (Erikson, 1977). The southern end of the batholith is

surrounded by ultramafic rock of the Ingalls tectonic complex. The

contact between these two rock units is marked by a discontinuous zone

of chiefly biotite and hornblende schists. In contact with the Mount

Stuart batholith to the north is the Chiwaukum Schist which is composed

predominantly of garnet and staurolite-bearing graphite-biotite-quartz

schist with lesser amounts of siliceous hornblende schist, amphibolite,

calc-silicate schist and marble (Page, 1939).

The Chelan Mountain terrane is located in the vicinity of Lake

Chelan. This terrain is mainly characterized by migmatitic and

tonalitic plutonic rocks, but also contains amphibolite, biotite schist,

siliceous schist, and rare light-colored tonalite sills and dikes (Cater

and Wright, 1967). Work by Mattinson (1972) revealed a 220 m.y. age for
 48

PR~Ys;J; I 10 I MO

~·~~H
Spokane
•
• MS ···.
:: :·: Missoula
I "\ •
catJJ . Ill.N~
SS Yakima
• WA
-
--
•
OR

Portland
AT r _,,,,, _,,,,, 1
N

~
r
I
0 100Km

Figure 11. Map showing the location of the Summit Creek sandstone
(SS), Naches (N), Chumstick (CH), and Carbonado Formations (CB), in
relation to the four most probable source areas: Mount Stuart
batholith (MS), Chelan Mountain terrane (CM), Pasayten River Area (PR),
and the Idaho batholith (Bitterroot (BR) and Atlanta (AT) lobes).
 49

the plutonic rocks in this area. The Entiat pluton, as described by

Waters (1930), represents a facies of the Chelan batholith, and is

characterized by medium-grained rocks of granitoid composition. These

rocks have a much higher K-f eldspar content than other plutonic rocks of

the Chelan area. K-Ar age dates on hornblende in the Entiat pluton

reveal a 60 to 73 m.y. age for this pluton (Mattison, 1972).

The Pasayten River area studied by Tabor and others (1968) is

located in the north central part of the state of Washington. This area

contains granitic rocks of the Monument Peak stock, hornblende-biotite

granodiorite of the Castle Peak stock, hornblende-biotite-granodiorite

and quartz monzonite of the Lost Peak stock, biotite-hornblende quartz

diorite to granodiorite of the Pasayten and Rock Creek dikes, along with

gneiss, schist and granitoid plutons of the Okanogan Highland crystal-

line block. These rocks range from 46 to 86 m.y. in age.

The Idaho batholith is located in east-central Idaho. This batho-

lith can be broken down into two lobes, the northern Bitterroot lobe,

and the southern Atlanta lobe. The Bitterroot lobe is both the smaller

and the younger of the two lobes. It ranges in age from 85 to 66 m.y.,

and is slightly over half the size of the Atlantic lobe. The Bitterroot

lobe is predominantly granodioritic and granitic in composition, and was

emplaced into metamorphosed Belt Group and pre-Belt basement rocks

during late Cretaceous time (Hyndman and others, 1977). The Atlanta

lobe is 63 to 117 m.y. in age and the rocks grade in composition from

west to east across the western half of the batholith from fine grained

migmatitic gneisses, to biotite-quartz schist, to leucocratic quartz

dioritic migmatite, to quartz dioritic gneiss, to leucocratic qµartz

 so
diorite, to granodiorite, to quartz monzonite, and finally granite at

the interior of the batholith (Hyndman and others, 1977).

Comparing the rock composition from the four possible source areas

mentioned above, to the framework grain composition of the Summit Creek

sandstone, Naches, Chumstick, and Carbonado Formations, it appears that

the Mount Stuart batholith area, and the Chelan Mountain terrane are the

most likely source rocks for these sediments. This conclusion is based

on the fact that the four formations under study contain no more than

12% K-feldspar. The Mount Stuart batholith area, and the Chelan

Mountain terrane both contain plutonic rocks which are predominantly

tonalitic in composition (<10% K-feldspar), with smaller amounts of

granodiorite (<35% K-feldspar). The Pasayten River Area, and the Idaho

batholith, on the other hand, all contain abundant granite (>35% K-

feldspar) and granodiorite (<35% K-feldspar) plutonic rocks. If these

sediments had been derived from the Pasayten River Area, or the Idaho

batholith area, one would expect higher concentrations of K-feldspar.

The area immediately surrounding the Mount Stuart batholith,

and the Chelan Mountain terrane can both account for the rather large

percentage of quartz-biotite-schist metamorphic rock fragments found in

the four formations. The Mount Stuart batholith area could have sup-

plied these metamorphic rock fragments from the Chiwaukum Schist located

just north of the batholith, while the Chelan Mountain terrane could

have supplied these metamorphic rock fragments from schists of Twenty-

Fi ve Mile Creek.

The Mount Stuart area can also account for the rather large per-

centage of mafic volcanic rock fragments found in these sediments. These

 51

mafic fragments could be derived from the ultramafic Ingalls tectonic

complex just south of Mount Stuart. The Ingalls complex contributing to

these sediments could account for the rather high Cr concentrations in

these sediments. The Naches Formation, which is the closest formation

to the Ingalls Complex, has a significantly higher Cr concentration than

the other formations, as one would expect.

Paleocurrent direction data collected by previous workers reveals

dominantly southwesterly current directions (Winters, 1983; Buckovic,

1979; Buza, 1979) for most of the Eocene sedimentary rocks in the area.

These current directions are consistent with a Mount Stuart and Chelan

Mountain terrane source rock.

 CHAPTER VI

SUMMARY

The purpose of this study is to compare the sandstone composition

and trace element geochemistry between samples representing the Summit

Creek sandstone, Naches, Chumstick, and Carbonado Formations in hopes of

identifying if these sediments were all derived from the same provenance,

and to determine the composition of the source rocks in hopes to identi-

fy the present day location of the source areas.

The light mineral portion of this study revealed the Summit Creek

sandstone, Naches, Chumstick, and Carbonado Formations to be arkosic to

lithic arkosic in composition, indicating a silicic plutonic source

area. The rather small percentage of K-feldspar found in these sediments

indicates that these sandstones were not derived from rocks of true

granitic composition, since granitic rocks typically contain greater

than 30% K-feldspar, but were more likely derived from tonalites,

quartz-diorites, or granodiorites. The majority of the polycrystalline

quartz appears to be derived from upper greenschist facies metamorphic

rocks. The rock fragment data parallels the polycrystalline quartz data

by indicating a major low grade quartz-biotite-schist metamorphic source

rock, as well as a mafic volcanic source rock, contributing sediments to

 53

the finer fractions of these rocks.

All four formations are characterized by the heavy minerals

epidote, apatite, zircon, garnet, sphene, rutile, tourmaline, hornblende

and hypersthene. The Naches, Chumstick, and Carbonado Formations also

contain small percentages of zoisite and augite, and the Carbonado

Formation contains traces of staurolite. These heavy minerals appear to

indicate that the coarser fraction of the four formations is derived

predominantly from plutonic as opposed to high grade metamorphic source

rocks, which is evidenced by the absence of upper amphibolite and

granulite facies metamorphic minerals such as kyanite, sillimanite, and

andalusite.

The cluster analysis constructed from the trace element data

reveals two populations of samples. The first population consists of

samples representing the Summit Creek sandstone, Chumstick, and

Carbonado Formations, while the second population consists of samples

from the Naches Formation. The strong geochemical similarity that

exists between the Summit Creek sandstone, Chumstick, and Carbonado

Formations supports the interpretation that these sediments were derived

from the same provenance. The somewhat unique chemistry of the Naches

Formation appears to indicate that a local source rock is contributing

more to the Naches Formation than to the other three formations.

The significantly higher concentration of Cr, and Sc, and the signific-

antly lower concentration of La in the Naches Formation as opposed to the

Summit Creek sandstone, Chumstick, and Carbonado Formations suggests

that the Naches Formation has a significantly larger mafic or ultramafic

component to its sediments than the other three formations.

 54

Comparing the light and heavy mineral and geochemical data in this

study with published data on time correlative units reveals three dis-

tinct populations of formations. The first population, which shows the

strongest compositional similarity includes the Summit Creek sandstone,

Chumstick, Naches, Carbonado, and Roslyn Formations. The second popula-

tion includes the Cowlitz, Skookumchuck, Renton, and Spiketon

Formations, and the third population includes the Chambers Creek beds.

The formations within each of the three populations were most likely all

derived from the same provenance, while formations between populations

are likely being influenced at least partially by unique source rocks.

The Mount Stuart batholith area, and the Chelan Mountain terrane

are assumed to be the major source rocks for the Summit Creek sandstone,

Naches, Chumstick, Carbonado, and Roslyn Formations based on their light

and heavy mineral assemblages, and current directions.

 REFERENCES

Al-Azzaby, F.A., 1980, Stratigraphy and Sedimentation of the

Spencer Formation in yamhill and Washington Counties,
Oregon (M.S. thesis): Portland, Oregon, Portland State
University, 104 p.

Armentrout, J.M., Hull, D.A., Beauliau, J.D, and Rau, W.W., 1983,
Correlation of Cenozoic stratigraphic units of western
Oregon and Washington: Oregon Dept. of Geol. and Min. Ind.
Oil and Gas Investigation 7, 90 p.

Blatt, H., Middleton, G., and Murray, R., 1980, Origin of

sedimentary rocks: Englewood Cliffs, New Jersey, Prentice-
Hall, 782 p.
Bressler, C.T., 1951, The petrology of the Roslyn arkose, central
Washington (Ph.D. thesis): University Park, Pennsylvania,
Penn. State Univ., 147 p.

Buckovic, W.A., 1974, The Cenozoic stratigraphy and structure of

a portion of the west Mount Rainier are, Pierce County,
Wahington (M.S. thesis): Seattle, University of
Washington, 123 p.

Buckovic, W.A., 1979, The Eocene deltaic system of west-central

Washington: Soc. Econ. Paleo. Min., Pacific Coast
Paleogeography Symposium 3, p.147-163.

Buza, J.W., 1979, Dispersal patterns and paleogeographic

implications of lower and middle Tertiary fluviatile
sandstones in the Chiwaukum Graben, east-central Cascade
Range, Washington: Soc. Econ. Paleo. Min., Pacific Coast
Paleogeography Ssymposium 3, p. 63-73.

Carver, R.E., 1971, Procedures in sedimentary petrology: New

York, Interscience Pub., 635 p.
 56
Cater, F.W., and Wright, T.L., 1967, Geologic map of the Lucerne
quadrangle, Chelan County, Washington: USGS Geol. Quad.
Map GQ-647.

Clayton, G.A., 1983, Geology of the White Pass area, south-

central Cascade Range, Washington (M.S. thesis): Seattle,
University of Washinton, 212 p.

Crooke, K.A., 1960, Classification of Arenites: American Journal

of Science, v. 258, no. 6, p. 417-428.

Curtis, C.D., 1972, Scandium: in Fairbridge, R.W., The

encyclopedia of Geochemistry and Environmental Sciences,
Van Nostrand Reinhold Comp., New York, p. 1061-1062.

Davis, G.D., Monger, J.W., and Burchfiel, B.C., 1978, Mesozoic

construction of the Cordilleran 'Collage', central British
Columbia to central California: Soc. Econ. Paleo. Min.,
Pacific Coast Paleogeography Ssymposium 2, p. 33-70.

Deer, W.A., Howie, R.A., and Zussman, J., 1966, An introduction

to the rock forming minerals: London, England, Longman,
347 p.

Dickinson, W.R., 1970, Interpreting detrital modes of graywacke

and arkose: Jour. Sed. Pet., v. 40, no. 2, p. 695-707.

Dickinson, W.R., and Suczek, C.A., 1979, Plate tectonics and

sandstone compositions: Amer. Assoc. Pet. Geol. Bull.,
v.62, p. 2164-2182.

Erikson, E.H., 1977, Petrology and petrogenesis of the Mount

Stuart Batholith: v. 60, p.183-207.

Frizzell, V.A., Jr., 1979, Petrology and stratigraphy of

Paleogene nonmarine sandstones, Cascade Range, Washington:
U.S.G.S. Open File Report 79-1149, 151 p.

Gard, L.M., Jr., 1968, Bedrock geology of the Lake Tapps

quadrangle, Pierce County, Washington: U.S.G.S. Prof. Paper
388-B, 33 p.

Grant, A.R., 1982, Summary of economic geology data for the

Glacier Peak Wilderness, Chelan Snohomish and Skagit
Counties, Washington: U.S.G.S. Open File Report 82-0408,
41 p.

Gresens, R.L., Naeser, C.W., and Whetten, J.T., 1981,

Stratigraphy and age of the Chumstick and Wenatchee
Formations: Geol. Soc. Amer. Bull., Part I, v. 92, p.233-
236.
 57

Gresens, R.L., Naeser, C.W., and Whetten, J.T., 1981,

Tertiary fluvial and lacustrine rocks, Chiwaukum Graben,
Washington: Geol. Soc. Amer. Bull., Part II, v. 92, p. 841-
876.

Hammond, P.E., 1980, Reconnaissance geologic map and cross-

sections of southern Washington Cascade Range: Portland,
Oregon: Publication of the Department of Earth Sciences,
Portland State University, 31 p.

Hampel, C., 1972, Hafnium: in Fairbridge, R.W., The Encyclopedia

of Geochemistry and Environmental Sciences, Van Nostrand
Reinhold Comp., New York, p. 488-489.

Hyndman, D.W., Alt, D., 1977, The Idaho Batholith and its
.tectonic setting: Northwest Geology, v. 6-1, p. 1-16.

Hyndman, D.W., 1983, The Idaho Batholith and associated plutons,

Idaho and western Montana: Geol. Soc. Amer. Memoir 159, p.
213-240.
Jackson, M.K., 1983, Stratigraphic relationships of the Tillamook
Volcanics and the Cowlitz Formation in the Upper Nahalem
River - Wolf Creek area, northwestern Oregon (M.S. thesis):
Portland, Oregon, Portland State University, 109 p.

Kadri, M.M., 1982, Structure and influence of the Tillamook

uplift on the stratigraphy of the Mist Area, Oregon (M.S.
thesis): Portland, Oregon, Portland State University, 105
p.

Karachewski, J.A., 1983, Paleomagnetism, sedimentary petrology,

and paleogeography of the Oligocene Lincoln Creek Formation
in Grays Harbor basin, southwestern Washington (M.S.
thesis): Bellingham, Western Washington University, 180 p.

Kay, R., 1972, Lanthanum and Samarium: in Fairbridge, R.W., The

Encyclopedia of Geochemistry and Environmental Sciences,
Van Nostrand Reinhold Comp., New York, p. 641-642, and
1060-1061.

Kienle, C.F., Sheriff, S.D., and Bentley, R.D., 1978, Tectonic

significance of the paleomagnetism of the Frenchman Springs
Basalt, Oregon and Washington: Geol. Soc. Amer. Abst. with
Programs, v. 10, p. 111-112.

Mattinson, J.M., 1972, Ages of zircons from the north Cascade

Mountains, Washington: Geol. Soc. America Bull., v. 83, p.
3769-3784.
 58

McLennan, S.M., Nance, W.B., and Taylor, S.R., 1980, Rare earth
element-thorium correlations in sedimentary rocks, and the
composition of the continental crust: Geochimica et
Cosmochimica Acta, v. 44, p. 1833-1839.

Miller, R.B., 1982, Geology of the Rimrock Lake pre-Tertiary

inlier, southern Washington Cascades: Geol. Soc. Amer.
Abst. with programs, v.14, no. 4, p. 217.

Misch, P., 1966, Tectonic evolution of the northern Cascades of

Washington State - a west-Cordilleran case history, in
symposium on the tectonic history, mineral deposits of the
western Cordillera in British Columbia and in neighboring
parts of the U.S.A.: Canadian Institute of Mining and
Metallurgy, special v. 8, p. 101-148.

Misch, P., 1977, Dextral displacements at some major strike

faults in the North Cascades: Geol. Assoc. of Canada
Programs with Abstracts, p. 37.

Moore, W.S., 1972, Thorium: in Fairbridge, R.W., The

encyclopedia of geochemistry and environmental sciences,
Van Nostrand Reinhold Comp., New York, p. 1183-1188.

Mullineaux, D.R., 1970, Geology of the Renton, Auburn, and Black

Diamond quadrangles, King County, Washington: U.S.G.S.
Prof. Paper 672, 92 p.

Newcomb, R.C., 1970, Tectonic structure of the main part of the

basalt of the Columbia River Group, Washington, Oregon, and
Idaho: U.S. Geol. Survey Misc. Geol. Inves. Map I 587.

Page, B.M., 1939, Geology of a part of the Chiwaukum quadrangle

(Ph. D. thesis): Stanford University, 203 p.

Phillips, W.R., Griffen, D.T., 1982, Optical Mineralogy: W.H.

Freeman and Company, San Francisco, 677 p.

Rau, W.W., 1981, Pacific Northwest Tertiary benthic foraminiferal

biostratigraphic framework - a review: in Armentrout,
J.M., ed., Pacific Northwest Cenozoic biostratigraphy,
Geol. Soc. Amer. Special Paper 184, p. 67-84.

Smith, C.H., 1972, Chromium: in Fairbridge, R.W., The

Encyclopedia of Geochemistry and Environmental Sciences,
Van Nostrand Reinhold Comp., New York, p. 167-170.

Smith, G.O., and Calkins, F.C., 1906, Description of the

Snoqualmie quadrangle, Washinton: U.S. Geol. Survey Geol.
Atlas, Snoqualmie Folio 139, 44 p.
 59

Snavely, P.D., Rau, W.W., Hoover, L.Jr., Roberts, A.E., 1951,

Mcintosh Formation, Centralia-Chehalis Coal District,
Washington: Amer. Ass. of Pet. Geol. Bull. v. 35, no. 5,
p. 1052-1061.

Snavely, P.D., Brown, K.D., Jr., Robert, A.R., and Rau, W.W.,
1958, Geology and coal resources of the Centralia district,
Washington: U.S. Geol. Sur. Bull., 1053, 159 p.

Tabor, R.W., Engels, J.C., Staatz, M.H., 1968, Quartz diorite,

quartz monzonite, and granite plutons of the Pasayten River
Area, Washingtqn - petrology, age, and emplacement: U.S. Geol.
Sur. Prof. Paper 604, 67 p.

Tabor, R.W., ·Frizzell, V.A. Jr., Vance, J.A., and Naeser, C.W.,
1984, Ages and stratigraphy of lower and middle Tertiarty
sedimentary and volcanic rocks of the central Cascades,
Washington: Application to the tectonic history of the
Straight Creek Fault: Geol. Soc. Amer. Bull., v. 95, p.
26-44.

Van Atta, R.O., 1971, Sedimentary petrology of some Tertiary

formations, upper Nehalem River basin, Oregon (Ph.D.
thesis): Corvallis, Oregon, Oregon State University,
245 p.

Vine, J.D., 1962, Stratigraphy of Eocene rocks in part of King

County, Washington: Washington Div. of Mines and Geol.
Rept. Inv. 21, 20 p.

Vine, J.D., 1969, Geology and coal resources of the Cumberland,

Hobart, and Maple Valley quadrangles, King County,
Washington: U.S. Geol. Survey Prof. Paper 624, 67 p.

Waters, A.C., 1930, Geology of the southern half of the Chelan

quadrangle (Ph.D. thesis): New Haven, Conn, Yale
University, 265 p.

Wells, R.E., 1981, Geologic map of the eastern Willapa Hills,

Cowlitz, Lewis, Pacific, and Wahkiakum Counties,
Washington: U.S. Geol. Survey Open-File Report 81-674.

Whetten, J.T., 1976, Tertiary sedimentary rocks in the central

part of the Chiwaukum Graben, Washington: Geol. Soc. Amer.
Abst. with programs, v.8, p. 420-421.

White, C.A., 1888, On the Puget Group of Washington Tertiary:

Am. Jour. Science, v. 36, p.443-450.
 60
Winters, W.J., 1984, Stratigraphy and sedimentology of Paleogene
arkosic and volcanoclastic strata, Johnson Creek-Chambers
Creek Area, southern Cascade Range, Washington (M.S.
thesis): Portland, Oregon, Portland State University,
134 p.

Wolfe, J.A., 1968, Paleogene biostratigraphy of nonmarine rocks

in King County, Washington: U.S. Geol. Survey Prof. Paper
571, 33 p.

Young, S.W., 1976, Petrographic textures of detrital polycrystal-

line quartz as an aid to interpreting crystalline source
rocks: Journal of Sed. Pet., v.46, p.595-603.
 APPENDIX A

TRACE ELEMENT DATA

MEB TIIESIS lST CT

SAMPLE BA CE a> FE
--
CHS 1000.00 ± 200.00 75.00 + 9.00
-
18.00 + 2.00 4.52
---
+ 0.20
CH16 900.00 200:00 54.00 7.00 14.00 2.00 5.40 0.20
CH29 900.00 200;00 67.00 8.00 15.00 3.00 4.40 0.20
SN17T 610.00 150.00 54.00 7.00 12.20 1.80 5.34 0.18
SN19 680.00 160.00 63.00 7.00 11.20 1.60 4.42 0.15
SN32 780.00 170.00 57.00 6.00 5.40 1.50 5.74 0.16
SN47 640.00 170.• 00 38.00 6.00 17.80 2.00 5.36 0.18
N3 900.00 200.00 40.00 6.00 20.00 2.00 6.10 0.20
N7 610.00 150.00 37.00 s.oo 27 .oo 2.00 7.50 0.20
NlO 470.00 160~00 19.00 5.00 23.00 2.00 7.90 0.20
Nl2 720.00 170.00 39.00 5.00 15.60 1. 70 5.33 0.17
Nl5 740.00 180.'00· 37.00 5.00 25.00 2.00 6.01 0.18
S4 600.00 140.00 41.00 5.00 5.70 1.10 6.05 0.16
SlO 720.00 170.00. 71.00 8.00 10.70 1.60 3.61 0.14
Sll 620.00 150.00 65.00 7.00 8.30 1.50 3.53 0.14
Sl2 750.00 170.00 67.00 7.00 12.80 1.60 . 4.58 0.16
BQM4 960.00 190.00 72.00 7.00 13.30 1.20 4.38 0.14
GSPl 620.00 120.00 201.00 18.00 3.70 1.10 1.57 0.05
BCRl 680.00 170.00 54.00 7.00 38.00 3.00 13.40 0.30

HEB TIIESIS lST CT

SAMPLE HF K LA NA
CHS 6.70 + 1.50 0.00 ± o.oo 31.90 +
-
0.60
--
2.74 + 0.01
CH16 5.10 1.30 2.30 a.so 24.60 0.60 2.70 0.01
CH29 7.20 1.60 3.40 0.50 26.80 0.60 2.64 0.01
SN17T 4.80 1.10 1.30 0.40 25.00 0.50 1.64 0.01
SN19 3.90 1.00 2.40 0.50 26.70 o.so 1. 70 0.01
SN32 3.30 0.90 1.50 o.so 25.60 0.50 1.58 0.01
SN47 3.80 1.00 o.oo 0.00 16.50 a.so 2.35 0.02
N3 4.00 1.10 0.00 0.00 16.50 0.50 2.74 0.02
N7 3.70 0.90 0.00 0.00 18.20 o.so 1. 75 0.01
NlO 4.00 1.00 3.20 0.70 12.70 0.40 2.02 0.02
Nl2 6.10 1.30 2.00 0.60 17.40 0.40 2.29 0.02
Nl5 3.90 1.00 0.00 0.00 18.00 0.50 2.32 0.02
S4 3.60 0.80 1.60 o.so 20.00 0.40 1.10 0.01
SlO 5.10 1.10 ·2.so 0.80 31.90 0.60 2.04 0.02
Sll 5.00 1.10 2.40 0.60 26.70 0.60 2.15 0.02
512 6.30 1.30 3.00 0.70 30.90 0.60 2.06 0.02
BQM4 4.60 0.90 3.80 0.60 32.80 0.50 3.10 0.01
GSPl 4.60 0.80 2.90 0.40 70.30 0.80 1.05 0.01
BCRl 4.70 1.10 1. 70 0.20 26.00 0.40 3.27 0.01
MEB THESIS lST CT
62
SAMPLE SC SM YB
CHS 11.18
---------------------------------
± 0.18 5.47 ± 0.06 1.70 ± 0.30
CH16 13.40 0.20 5.01 0.06 2.20 0.30
CH29 12.90 0.20 5.22 0.06 2.50 0.30
SN17T 10.15 0.15 4.52 a.as 2.00 0.20
SN19 10.37 0.16 4.76 0.05 2.30 0.30
SN32 10.28 0.16 4.33 0.05 2.00 0.20
SN47 12.43 0.19 3.40 0.05 1.60 0.30
N3 16.50 0.20 4.18 0.05 2.30 0.30
N7 13.81 0.19 4.11 0.05 2.20 0.30
NlO 22.20 0.30 3.79 0.05 1.80 0.30
Nl2 15.65 0.20 4.28 0.05 2.20 0.30
Nl5 15.63 0.20 4.23 0.05 2.30 0.30
S4 9.08 0.13 3.61 0.04 1.42 0.19
SlO 9.51 0.15 5.08 0.06 2.00 0.20
Sll 7.98 0.14 4.32 0.05 2.00 0.20
Sl2 9.09 0.15 4.97 0.06 2.00 0.20
BQH4 11.00 0.10 4.05 0.03 1.83 0.18
GSPl 2.39 0.04 9.66 0.06 0.65 0.09
BCRl 33.00 0.20 6.60 0.05 3.40 0.30

MEB THESIS 2ND CT

SAMPLE BA CE (l) CR

CHS 730.00 :!: 100.00 8S.OO + 3.00 12.10 +

-----------------
0.03 130.00 :!: 20.00
CH16 610.00 100.00 75.00 2.00 11.70 0.30 140.00 20.00
CH29 760.00 110.00 73.00 2.00 11.40 0.30 114.00 lS.00
SN17T 430.00 80.00 77.00 2.00 9.50 0.30 74.00 12.00
SN19 490.00 80.00 73.00 2.00 10.40 0.30 96.00 15.00
SN32 460.00 80.00 66.00 2.00 6.60 0.20 86.00 13.00
SN47 540.00 100.00 . 44.10 1. 70 16.60 0.40 170.00 30.00
N3 630.00 120.00 34.00 1.70 lS.80 0.40 170.00 30.00
N7 440.00 90.00 33.50 1.60 23.70 0.40 125.00 19.00
NlO 460.00 110.00 21.00 1.90 24.00 0.40 150.00 20.00
N12 500.00 100.00 47.90 1.80 12.30 0.30 170.00 30.00
Nl5 640.00 120.00 50.90 2.00 19.90 0.40 160.00 20.00
S4 450.00 80.00 49.30 1. 70 5.00 0.19 71.00 11.00
SlO 470.00 90.00 92.00 3.00 8.10 0.20 74.00 12.00
SU 480.00 90.00 73.00 2.00 7.60 0.20 70.00 11.00
B<1f4 720.00 110.00 91.00 3.00 10.70 0.30 23.00 4.00
GSPl 460.00 70.00 251.00 6.00 2.22 0.11 6.80 1.60
BCRl 670.00 130.00 53.90 1.90 3S.OO 0.40 18.00 4.00

HEB THESIS 2ND CT

SAMPLE cs EU FE HF
- +
-------------
6.10 + 0.02
CHS
CH16
CH29
4.40
4.30
7.90
-
+ O.ll>
0.80
1.40
1.26
1.21
1.20
+ 0.07
0.07
0.07
4.63
4.93
4.45
0.04
0.05
0.05
5.10
6.10
0.20
0.20
SN17T 3.50 0.70 1.04 0.06 5.23 0.05 4.26 0.18
SN19 3.10 0.60 1.10 0.06 4.23 0.04 4.63 0.19
SN32 2.60 0.50 0.88 0.06 5.70 0.05 4.11 0.18
SN47 3.10 0.60 0.88 0.06 5.37 0.05 4.25 0.20
N3 2.50 0.50 1.08 0.09 6.13 0.06· 3.90 0.20
N7 3.ll> 0.70 1.06 0.07 7.13 0.06 3.43 0.17
NlO 2.60 0.60 1.13 0.08 7.62 0.07 3.20 0.20
Ml2 2.10 0.40 1.08 0.07 5.03 0.05 4.80 0.20
Nl5 3.50 0.60 1.02 0.07 5.86 0.06 3.74 0.19
S4 2.40 0.50 0.88 0.06 5.74 0.05 2.90 0.15
SlO 3.10 0.60 0.91 0.06 3.47 0.04 5.80 0.20
Sll 2.60 0.50 0.81 0.06 3.55 0.04 3.84 0.17
BQH4 8.20 1.50 1.01 0.06 4.46 0.05 4.90 0.20
GS Pl 0.69 0.15 0.83 0.04 1.56 0.02 6.00 0.20
BCRl 1.00 0.20 2.06 0.07 13.40 0.06 4.70 0.19
 63

MEB THESIS 2ND CT TH

SAMPLE RB SC TA

CHS
-------------------------------------------
140.00± 30.00 10.59 + 0.06 1.52 + 0.11 a.so + 0.30
CH16 90.00 20.00 13.25 0.07 1.25 0.10 6.40 0.20
190.00 40.00 12.53 0.01 1. 77 0.12 10.10 0.30
CH29 0.10 7.60 0.30
SN17T 110.00 20.00 10.03 0.05 1.40
SN19 110.00 20.00 10.30 0.06 1.22 0.09 7.40 0.20
SN32 90.00 20.00 10.22 o.os 1.19 0.09 6.60 0.20
79.00 20.00 12.27 0.07 0.91 0.08 4.45 0.19
SN47 3.80 0.20
N3 69.00 19.00 16.84 0.08 0.79 0.08
N7 51.00 16.00 13.84 0.06 0.85 0.08 2.97 0.16
70.00 20.00 22.04 0.10 0.71 0.08 2.50 0.20
NlO 0.20
N12 43.00 14.00. 15.81 0.07 0.81 0.07 4.17
NlS 69.00 18.00 16.16 0.07 0.91 0.08 3.62 0.19
S4 100.00 20.00 8.46 0.04 1.26 0.09 5.07 0.19
510 120.00 30.00 9.34 o.os 1.54 0.11 9.40 0.30
Sll 90.00 20.00 8.03 0.05 1.25 0.09 8.10 0.30
BQM4 180.00 40.00 . 10.60 0.06 1.16 0.09 20.90 0.60
GSPl 130.00 30.00 2.25 0.02 0.52 0.05 43.20 1.10
BCRl 47.00 13.00 33.00 0.09 0.91 0.07 6.00 0.20
 APPENDIX B

SAMPLE LOCATION SITES

FORMATION SAMPLE LOCATION QUADRANGLE ELEV. (FT)

-----------------------------------------------------------------------
CARBONADO SN3 SWl/4,SEl/4,SWl/4, MORTON 15' 1320
SEC.12,Tl3N,R4E
SN19 SEl/4,NWl/4,SWl/4 MORTON 15' 1520
SEC .11 , Tl3N, R4E
SN22 SWl/4,NEl/4,SWl/4 MORTON 15' 1560
SEC.ll,Tl3N,R4E
SN38 NEl/4,SEl/4,SEl/4 MORTON 15' 1780
SEC.10,T13N,R4E
SN58 NWl/4,NEl/4,SWl/4 MORTON 15' 2320
SEC.10,Tl3N,R4E
-----------------------------------------------------------------------
SUMMIT CREEK S4 SEl/4,NEl/4,NEl/4 WHITE PASS 15' 3160
SANDSTONE SEC.13,T14N,RlOE ..
SS NEl/4,SEl/4,SWl/4 WHITE PASS 15' 3140
SEC.13,Tl4N,R10E
S6 NEl/4,SEl/4,SWl/4 WHITE PASS 15' 3120
SEC.13,Tl4N,RlOE
SlO NEl/4,SWl/4,SWl/4 WHITE PASS 15' 3100
SEC.13,T14N,R10E
S12 NWl/4,SEl/4,SEl/4 WHITE PASS 15' 3080
SEC.13,T14N,RlOE
Sl2TK NWl/4,SEl/4,NWl/4 WHITE PASS 15' 3070
SEC.13,Tl4N,RlOE
---------------------------------------------------------------~-------
NACHES N3 NEl/4,NWl/4,SEl/4 EASTON 15' 3840
SEC.21,T18N,Rl4E
 65
N7 NEl/4,SEl/4,NEl/4 EASTON 15' 3760
SEC.21,T18N,R14E
N12 SWl/4,NWl/4,NEl/4 EASTON 15' 3720
SEC.22,T18N,R14E
N14 SWl/4,NEl/4,SWl/4 EASTON 15' 3800
SEC.22,Tl8N,Rl4E
Nl5 SWl/4,NEl/4,SEl/4 EASTON 15' 3840
SEC.22,Tl8N,Rl4E
N22 SWl/4,SEl/4,SWl/4 EASTON 15' 3720
SEC.22,Tl8N,Rl4E
N23 SWl/4,SWl/4,SEl/4 EASTON 15' 3640
SEC.22,Tl8N,Rl4E
--------------------------------~-----------------------------------
CHUMSTICK CH! SWl/4,SEl/4,NWl/4 CHIWAUKUM 820
SEC.33,T24N,Rl9E . 4SE, 7. 5'
CHS SEl/4,SWl/4,NEl/4 CHIWAUKUM 1240
SEC.32,T24N,Rl9E 4SE, 7 .5'
CH16 SEl/4,SEl/4,SEl/4 CHIWAUKUM 990
SEC.26,T24N,Rl8E 4SE, 7 .5'
CH18 NWl/4,SWl/4,NEl/4 CHIWAUKUM 1580
SEC.9,T24N,Rl8E 4SE, 7 .5'
CH26 SEl/4,NEl/4,SEl/4 CHIWAUKUM 1440
SEC.24,T25N,R17E 4NW, 7 .5'
CH29 NWl/4,NEl/4,SWl/4 CHIWAUKUM 1620
SEC.18,T25N,Rl8E 4NW, 7 .5'