Proterozoic crustal history of the western United States

as determined by neodymium isotopic mapping

VICTORIA C BENNETT 1
DONALD J DEPAOLO J department of Earth and Space Sciences, University of California, Los Angeles, California 90024

ABSTRACT were more effectively screened, so that the tionships among rock units, and the crystalliza-
most distal province (3) is derived almost en- tion ages have been overprinted or obscured by
Initial Nd isotopic ratios of crystalline rocks tirely from Proterozoic mantle. later geologic events.
from an area oi ~ 1.5 x 10 6 km 2 of the west- The province boundaries are subparallel to The first major synthesis of continental crustal
ern United States have been determined in the crystallization age trends determined by ages as determined by studies of natural radio-
order to map Precambrian age province other workers. An exception to this is the genic isotope systems was presented by Hurley
boundaries and thus document the growth Mojavia region of province 1, which cross- and Rand (1969). Since that time, the methods
and modificatio n of the North American con- cuts and truncates the other provinces in the available for mapping of crustal age provinces
tinent in the Proterozoic. The use of three region of the lower Colorado River. This re- have been considerably expanded. Zartman
representative rock suites of different ages— gion appears to be displaced relative to other (1974) showed that crustal age provinces in the
Mesozoic and Tertiary peraluminous granitic areas of the North American basement that western United States were reflected in the Pb
rocks, middle Proterozoic (ca. 1.4 Ga) "an- have similar isotopic characteristics. This isotopic compositions of Mesozoic and Ceno-
orogenic" granitic rocks, and lower Protero- suggests the presence of a previously unrec- zoic igneous rocks and associated ore deposits.
zoic (ca. 1.7 G a) igneous and metamorphic ognized large-scale, left-lateral, north-south- Kistler and Peterman (1973,1978) were able to
rocks—allows ithe ages of the provinces to be trending basement offset of Proterozoic age use Sr isotopic compositions of Mesozoic and
distinguished o n the basis of different Nd iso- in the vicinity of the California-Arizona Cenozoic igneous rocks in California and Ne-
topic evolution paths rather than solely on the border. vada to map the western edge of the Precam-
basis of model ages. Three age provinces brian basement of North America and put
have been delineated, each generally northeast- INTRODUCTION constraints on Cenozoic tectonic movements.
southwest trending, having decreasing crys- Recently, it has been demonstrated that the Sm-
tallization age; and increasing initial e^d The continental crust preserves a detailed rec- Nd isotopic method provides a particularly
values with increasing distance southeast- ord of the Earth's evolution over some 3.8 b.y. promising addition to the available methods be-
ward from the Archean craton. Province 1 is One of the most important and fundamental as- cause the crustal ages are generally better pre-
composed of crustal rocks of central Utah pects of this record is its chronological frame- served through later geologic disturbances
and northeastern Nevada, which are charac- work, Continental crust, which is composed of (McCulloch and Wasserburg, 1978; DePaolo,
terized by average values of «^(1.7 Ga) ~ 0 rock material that was derived from, and is 1980; Nelson and DePaolo, 1985). Farmer and
and T d m ~ 2.0-2.3 Ga. Province 2 covers chemically fractionated with respect to, the DePaolo (1983, 1984) used the Nd isotopic
Colorado, southern Utah, and northwestern upper mantle, has been forming throughout the patterns in young granitic rocks to demonstrate
Arizona and has e Nd (1.7 Ga) = +3 and Tdm "" past 3.8 b.y., that is, different parts of the conti- the existence of a previously unrecognized age
1.8-2.0 Ga. Province 3, which comprises the nents have different ages. Geochronological province in the western United States. Patchett
basement rocks of New Mexico and southern studies have shown that the continents are a and others (1981) showed that the Lu-Hf iso-
Arizona, has 6^(1.7 Ga) ~ +5 and TDM "" mosaic of relatively large segments, each with a topic system was yet another means of determin-
1.7-1.8 Ga. An additional region of province characteristic age or a limited range of ages, that ing crustal ages and that it had properties similar
1-type isotopic characteristics, herein named have been assembled by processes that are to those of the Sm-Nd system. Accompanying
"Mojavia," is found in eastern California and collectively referred to as "continental accre- improvements in the determination of radiomet-
western Nevada. Crust formation in each tion." The delineation of the boundaries of these ric ages of rocks (for example, Papanastassiou
province involved a large component of provinces can provide information about how and Wasserburg, 1969; Lugmair and others,
mantle-derived material plus a moderate the continents formed and how they have since 1975; Krogh, 1982a, 1982b) have led to refine-
amount (-20%) of pre-existing crust. As the been modified by breakup and continental drift. ments in the chronology and areal extent of Pre-
new crust was built outward from the Ar- The task is difficult, however, because, com- cambrian events (for example, Bickford and
chean nucleus, however, contributions of Ar- monly, multiple episodes of metamorphism and others, 1981; Hoffman and Bowring, 1984). The
chean material to the newly forming crust deformation have destroyed the original rela- advent of these new and improved methods

Additional material for this article (an appendix) may be obtained free of charge by requesting Supplementary Data 87-30 from the GSA
Documents Secretary.

Geological Society of America Bulletin, v. 99, p. 6 7 4 - 6 8 5 , 8 figs., 1 table, November 1987.

674

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 675

makes it desirable now to re-evaluate the age Sm-Nd SYSTEMATICS ferred to as the "crust-formation age" (TCF)-
and structure of the continents (compare with AND CRUSTAL AGES The subsequent isotopic evolution of this crust
Nelson and DePaolo, 1985). can be predicted from the empirically deter-
In this work, we have attempted to take the Model ages (TDM) are calculated based on a mined distribution of fsm/Nd values (defined in
methods of crustal isotopic mapping a step semi-empirical model illustrated in Figure 1. A Table 1) in the crust (for example, Haskin and
further by focusing on the nature and geometry reasonably well defined curve can be established others, 1966; Ben Othman and others, 1984)
of crustal age province boundaries as deter- for the time evolution of the e^d value (defined (inset of Fig. 1) and has been confirmed by var-
mined by Sm-Nd isotopic patterns and by inte- in Table 1; e Nd refers to the initial, chondrite- ious studies (for example, DePaolo, 1980; Nel-
grating data from a variety of generations of normalized l 4 3 Nd/ 1 4 4 Nd isotopic ratio; eNd(O) son and DePaolo, 1985; McCulloch and
igneous and metamorphic rocks in the same re- to the measured isotopic ratio) of the parts of the Wasserburg, 1978). The bulk of the continental
gion. The results of this work demonstrate how mantle that have been involved in the produc- crust has fsm/Nd values that are -40% lower
isotopic data can be used to determine crustal tion of continental crust (DePaolo, 1980,1983; than those of chondritic meteorites, and so the
ages and how the delineation of age province Nelson and DePaolo, 1984,1985; Jacobsen and «Nd values in the crust diverge from those of the
boundaries can contribute to the understanding Wasserburg, 1984). If "primitive original crust" mantle as the crust ages. Some parts of the crust
of crustal evolution in the Precambrian and to is defined as continental crust derived from the are more mafic and have fsm/Nd values that are
the detection of large-scale structures that may mantle with no admixture of older crust, then closer to those of chondritic meteorites, but on
be indiscernible by other means. the Nd isotopic evolution of that crust must the basis of the existing data, these are assumed
begin from a point along the mantle evolution to make up a relatively small proportion of the
curve corresponding to its age. This time is re- crust.

+ 10
PRIMITIVE
Man TL F
ORIGINAL C R U S T

co
o
o
i-
AVERAGE |
oO^'cP^* CRUSTAL EVOLUTION
•Nd

CRUSTAL ROCKS

-10

-0.6 -0.4 -0.2

fsm/Nd

I
Today T XCF
M2 M1

Age
Figure 1. Model for Nd isotope systematics of an average crustal segment. Primitive crust having low Sm/Nd is formed at TCp from a
depleted mantle source and evolves toward negative e^d values. Parallelograms represent the possible range of initial e^d for granitoids formed
at magmatic events TMI and TMJ. The shaded regions show the narrow range of values observed in strongly peraluminous granites. This
demonstrates how the isotopic evolution of a crustal segment may be determined by sampling basement rocks and crustally derived rocks of
different ages. Substantial additions (>30%) of new crustal material having higher Sm/Nd may alter the crustal evolution path, as indicated by
the dashed arrow at TMI. The inset shows the distribution of fsm/Nd values in rocks of the continental crust as compiled in Ben Othman and
others (1984).

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 676 BENNETT AND DEPAOLO

TABLE 1. NEODYMIUM ISOTOPIC DATA FOR CRUSTAL ROCKS O F THE WESTERN UNITED STATES

No, Sample Locality Rock type* [Ndjt ""Sm/'«Nd f § Age(Ga) 143Nd/144Nd(0)tt ±2o
'Sm/Nd 'Nd(T) TDM«*

Cretaceous and Tertiary gr.tnitoids

Arizona

1 GR-10 Gunnery Range PG ' 16.491 0.10894 -0.4461 0.06 -12.41 0.511201 15 -11.74 1.51
2 GR-11 Gunnery Range PG 12.400 0.10304 -0.4761 0.06 -13.25 0.511158 18 -12.53 1.49
3 GR-20 Gunnery Range PG 24.148 0.08673 -0.5591 0.06 -12.85 0.511179 19 -11.99 1.28
4 TC-05 Texas Canyon PG 17.028 0.10098 -0.4866 0.06 -10.49 0.511299 24 -9.76 1.27

California

5 WM-BC White Mtns. PG 33.382 0.09120 -0.5363 0.07 -19.36 0.510845 18 -18.42 1.74
6 W236 Old Woman Mtns. PG 16.786 0.13070 -0.3355 0.07 -16.61 0510986 25 -16.02 2.31

Middle Proterozoic granitoids

Arizona

7 HP-3 Hualapai Mtns. G 161.215 0.09056 -0.5396 1.37 -21.38 0510741 18 -2.78 1.86
8 ANT-1 Hualapai Mtns. G 7.221 0.11116 -0.4349 1.37 -16.68 0510982 22 -1.68 1.87

California

9 W-78-2(i Whipple Mtn. G 80.478 0.11868 -0.3966 1.40 -16.13 0.511010 19 -2.12 1.97
10 DM-1C Dead Mtns. G 37.581 0.10505 -0.4659 1.42 -9.87 0.511331 18 6.90 1.28
11 HM-4 Homer Mtns. G 53.768 0.07368 -0.6254 1.4 -26.06 0.510502 16 -3.97 1.89

Nevada

12 NY-13A Newberry Mtns. G 102.410 0.10579 -0.4622 1.45 -19.65 0.510831 22 -2.72 1.99
13 GB-2 Gold Butte G 75.031 0.09819 -0.5008 1.42 -19.15 0.510856 25 -1.13 1.83
14 NY-6B Newberry Mtns. G 70.586 0.12927 -0.3428 1.42 -14.57 0.511091 13 -2.23 2.08

Older Precambrian rocks

Arizona

15 43 1I3E Grand Canyon D 48.784 0.10812 -0.4503 1.67 -17.07 0.510963 26 1.96 1.85
16 Z.G. Grand Canyon GD 38.047 0.12277 -0.3758 1.67 -14.02 0.511118 26 1.85 1.88
17 E.C. Grand Canyon GN 8.508 0.16872 -0.1423 1.76 -3.38 0.511663 19 2.96 1.95
18 117-72 Grand Canyon GD 21.280 0.10109 -0.4861 1.67 -19.68 0.510829 15 0.84 1.91
19 AY-1 Government Can. GD 29.070 0.09750 -0.5043 1.74 -17.47 0.510942 22 4.73 1.70
20 AB-8 Bagdad GN 25.073 0.10848 -0.4485 1.74 -15.72 0.511031 27 4.02 1.75
21 AY-4 Yarber Wash QD 11.817 0.10350 -0.4738 1.74 -14.69 0.511084 23 4.00 1.60
22 AY-2 Yarber Wash QD 11.062 0.11124 -0.4345 1.73 -14.62 0.511088 26 4.40 1.71
23 AB-9 Bagdad AM 15.090 0.14958 -0.2396 1.78 -7.31 0.511462 26 3.49 1.85
24 MHM-4 Mohave Mtns. GN 20.424 0.10484 -0.4670 1.7 -19.44 0.510841 25 0.63 1.96

Califom'a

25 C-720H ML Pinos AM 16.569 0.18349 -0.0672 1.71 -1.97 C.511735 20 0.93 2.59
26 C-9020; Mt. Pinos PS 56.252 0.10410 -0.4708 1.71 -21.6 C.510731 24 -1.25 2.10
27 C-7201! Mt. Pinos AG 81.291 0.10714 -0.4553 1.67 -19.65 C.510830 21 -0.44 2.02
28 HM-1 Hexie Mtns. AG 113.784 0.11164 -0.4325 1.7 -17.93 C.510918 19 0.65 1.98
29 CW-2 Chuckwalla Mtns. PS 52.981 0.10595 -0.4613 1.7 -22.71 C.510674 24 -2.89 2.22
30 W-78-411 Whipple Mtns. AG 32.915 0.1131S -0.4248 1.75 -20.28 C.510798 29 -1.50 2.19
31 OW-5 Piute Mtns. AG 75.549 0.10916 -0.4450 1.75 -19.70 0.510828 24 -0.01 2.06
32 HC-5 Happy Canyon AG 55.494 0.10250 -0.4789 1.8 -23.73 0.510621 21 -1.95 2.23
33 OW-4 Old Woman Mtns. PS 53.269 0.12652 -0.3568 1.76 -15.82 0.511026 25 0.05 2.13
34 TUR-5 Turtle Mtns. GN 48.566 0.12944 -0.3419 1.7 -14.42 0.511098 19 0.28 2.07

Utah

35 MM-111 Mineral Mtns. GN 23.947 0.11493 -0.4157 1.72 -16.63 0.510985 19 1.45 1.94

Wyoming

36 MB-3 Baggott Rocks GD 38.533 0.12041 -0.3879 2.4 -31.90 0.510202 22 -8.39 3.40
37 MB-5 Medicine Bow GD 45.065 0.11679 -0.4063 1.8 -17.35 0.510950 22 1.18 2.04

British Columbia

38 83-BC-S.HW Shuswap GN 22.966 0.10066 -0.4883 2.1 -23.04 0.510657 20 2.93 2.14

Idaho

39 CHES HE GN Priest River GN 20.501 0.11848 -0.3977 1.7 -13.24 0.511158 16 3.86 1.73

Note: the values used for CHUR (chondritic reservoir) are 143
N d / 1 4 4 N d = 0.511836, 1 4 7 Sm/ I 4 4 Nd = 0.1967 (Jacobsen and Wasserburg, 1980); XSm = 6.54 x 10~ 12 yr~'. SSample localities, descriptions, and age control are in the appendix.
*Rock designations are D = diorite, GD = granodiorite, GN - gneiss, QD = quartz diorite, AM = amphibolite, AG = augen gneiss, PS = pelitic schist, PG - peraluminous granite, G s granite.
^Concentrations are in parts per million; precision is ±0.02%.
§ l47
fsm/Nd - [ ( ' ^ - " / ' " N c U p , » - S m / l 4 4 N < t C H U R ) - 1],
" e N d ( T ) = 10 4 [( H 3 N<l/ 1 4 4 Nd(T) s a m p | e + 143
N d / 1 4 4 N d ( T ) C H U R ) -1], where (T) refers lo the ciystallization age of the rock; i N d ( 0 ) is the measured isotopic value.
l8
^Corrected for oxygen composition, using 0 / 1 6 0 = 0.00210 and ,7
0 / l 6 0 = 0.000391; corrected for mass discrimination by normalizing to l46
N d / 1 4 2 N d = 0.63613.
§$The depleted-mantle model age is calculated using the equation of DePaolo (1981a), t N d ( T ) = 0.25T 2 - 3T + 8.5.

If the crust that formed at Tcf (Fig- 1) is the granitic magmas are formed mainly by melt- e^d values of most of the granitoids do cluster
affected by a subsequent episode of granitic ing of typical crustal rocks, the initial c^d values near the average crustal ffjd values and that
magmatism at time T M ,, the potential source will cluster at the value of the average crustal values close to those of the mantle are relatively
materials for the magmas span the range from rock. In areas underlain by Precambrian crystal- rare (Farmer and DePaolo, 1983, 1984). If a
the average crustal to the mantle e ^ values. If line basement rocks, it is observed that the initial significant amount of mantle-derived magma is

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 677

injected into the crust at time T M 1 , its effect is to the efjd values of the granites are reasonably For all samples analyzed, >2 kg of rock was
increase both the average e^d value and the av- representative of the post-magmatism lower crushed, split, and powdered. In some cases,
erage fsm/Nd value of the crust. This will cause a crust in this area, which would imply that the splits were taken of powdered samples obtained
discontinuity in the crustal evolution path, as isotopic composition of the crust was homoge- from others (see appendix). Approximately 300
shown in Figure 1. If no significant amount of nized during the magmatic episode and that the mg of each sample was dissolved using HF and
mantle-derived magma enters the crust, the crus- wide diversity of ¿Nd values observed in the HCIO4, and isolation of Nd, Sm, Rb, and Sr was
tal evolution will remain undisturbed. Empiri- upper crust does not persist into the lower crust. accomplished using the ion-exchange proce-
cally, it is found that additions of mantle magma This area is an example of crust that is conceived dures detailed in DePaolo (1978). Care was
to cratons subsequent to their initial stabilization with an e ^ value different from that of the man- taken to ensure that the samples were complete-
are minor (Farmer and DePaolo, 1983, 1984), tle and suggests a mechanism whereby this new ly dissolved. If any undissolved material re-
even in areas experiencing thermal events of crust could have a relatively limited range of mained after the initial dissolution, the sample
substantial magnitude. The exception to this rule initial values even though its sources were iso- was centrifuged, and the residue was placed ei-
is continental edges that are in close proximity to topically diverse. ther in a Teflon bomb or in a sealed Teflon
a subduction zone. If the main magmatic arc is beaker with an HF-HNO3 mixture for dissolu-
superimposed directly on the cratonic continen- SAMPLING AND tion. When dissolved, the residue was added
tal edge, there is sufficient mantle magma enter- ANALYTICAL PROCEDURES to the rest of the sample, and the completely
ing the crust that the crust becomes noticeably dissolved sample was spiked for Rb, Sr, Sm,
modified (Farmer and DePaolo, 1984). Granitoids of three different ages were inves- and Nd.
The results of a second episode of granitic tigated in this study. These include (1) lower Nd was measured as NdO + with the pressure
magmatism at time Tm2 are similar to those Proterozoic granitoids and gneisses that have in the ion source maintained at ~8 x 10~7 torr to
described for T M i , except that the range of iso- crystallization ages in the range from 1.65 to insure that enough O2 was available to support
topic values for the potential source regions of >1.8 Ga, which are the oldest rocks in each the formation of NdO + ions. Typical beam in-
the magmas is proportionally larger. This situa- region; (2) crustally derived "anorogenic" gran- tensity was - 1 . 2 x 10~ n amp. Detailed discus-
tion has also been investigated (Farmer and De- itoids, having Middle Proterozoic ages of ca. 1.4 sion of the mass spectrometric procedures at the
Paolo, 1983, 1984), and it has been found that Ga (Anderson, 1987); and (3) strongly peralumi- University of California, Los Angeles, labora-
the granitoids generally have e Nd values that nous granitoids of Mesozoic-Tertiary age. Mea- tory is given by DePaolo (1981b). Blanks aver-
cluster near the average crustal values. In partic- surement of these suites of crustal rocks allows aged 0.1 ng for Nd and 20 pg for Sm and were
ular, large bodies of strongly peraluminous gran- the Nd isotopic history of this portion of the negligible for all samples.
ite almost invariably have e ^ values that are continental crust to be determined.
identical to the average crustal values. The in- In defining crust-formation provinces, the RESULTS
ferred fields for strongly peraluminous granites sampling strategy was to select samples from a
are indicated by the shaded parallelograms in broad geographic area, to have those samples be The Sm-Nd isotopic data are presented in
Figure 1. representative of the crust of that region, and to Table 1. For each sample, an initial t N d value
In some cases, continental crust is apparently have independent age control for each sample. and a Tdm model age are given. The initial val-
generated not purely from the mantle but by a This was accomplished by obtaining representa- ues were calculated according to the best age
mixture of mantle and pre-existing crustal mate- tive samples from all areas that have extensive determination for each sample. In some cases,
rial. An example is the crust of northwestern exposures of Proterozoic crystalline rocks. For the ages were estimated by correlation with
Nevada (Farmer and DePaolo, 1983). Follow- the lower Proterozoic suite, granitic or ortho- dated units. The rationale of each assignment is
ing the establishment of a passive continental gneissic units having wide areal extent were se- explained in the appendix.2 The sample localities
margin in the late Precambrian, a large accumu- lected. These igneous rocks provide samples for are given in Figure 2. (The initial ratios and the
lation of continentally derived clastic sedimen- which the isotopic compositions are averages inferred crustal ages, on the basis of the model
tary rocks formed, and this was accreted to the over the substantial volumes of their source shown in Figure 1, are displayed in Figure 4.)
western edge of the North American craton, rocks. In three cases (CW-2, OW-4, and C- The figures contain data from previous work
along with some amount of ocean-floor sub- 90202), metasedimentary rocks were analyzed (DePaolo, 1981a; Farmer and DePaolo, 1983,
strate, during the Paleozoic and Mesozoic. This in an attempt to identify the presence of older 1984; Nelson and DePaolo, 1985) as well as
region was then involved in Mesozoic granitic (>2.0 Ga) crustal components. In some regions data from this study. The samples have been
magmatism and is now part of the North Amer- with sparse geochronologic data, particularly grouped according to their Nd isotopic charac-
ican craton. Because the sedimentary and oce- southern Nevada and southeastern California, teristics so that each group represents crust of a
anic volcanic rocks that accumulated to form middle Proterozoic anorogenic (ca. 1.4 Ga age) given model age (or range of ages). As shown in
the original accreted terranes were derived from granites were analyzed, as they were the best Figure 3, the samples allow subdivision of the
a variety of isotopically distinct sources ranging documented samples available. In areas lacking Proterozoic crust into three provinces, referred
from Archean crust to Paleozoic mantle, the exposures of Precambrian crystalline rocks, not- to herein as provinces 1, 2, and 3 in order of
original range of e ^ values in this terrane was ably northeastern California and southernmost decreasing model age.
very large (Farmer and DePaolo, 1983; Farmer, Arizona, strongly peraluminous granites of The cutoff ages distinguishing the three data
1985). This wide range of «^d values is still evi- Mesozoic-Tertiary age were used as the only groups are based on limits imposed by certain
dent in the rocks exposed at the surface. The available probes of the crust. The locality, de- well-defined geological/chronological bounda-
granitic rocks exposed at the surface, however, scription, and age control for all samples are ries within the study area. As shown in Table 1,
have a much smaller range of e Nd values cluster- given in the appendix.1 the crystallization ages of the oldest rocks from
ing near t N d = 0. These values also show a gra- the study area are in the range from 1.67 to 1.8
dient normal to the trend of the paleocontinental Ga. Because we attach significance to model
edge, the most mantle-like values being farthest 'The appendix may be obtained free of charge by
requesting Supplementary Data 87-30 from the GSA
from the pre-existing continent. It is possible that Documents Secretary. 2
See footnote 1.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 678 BENNETT AND DEPAOLO

cant observations with regard to the nature of

crust production in Early Proterozoic time. It is
also clear from Figure 4 and from the province
map (Fig. 3) that each of our geographic group-
ings shows only a relatively small part of the
total range of e ^ values, even though the data
from all of the provinces together fill the whole
range with no significant gaps. The samples
from province 1 are easily distinguished from
those of province 3, but the data for the province
2 samples overlap those from the other prov-
inces slightly.
A large range of e^d is also inherent in the
data for the granitic rocks of ca. 1.4 Ga (Fig. 4).
The pattern shown is similar to that of the ca.
1.7 Ga rocks. The difference between the use of
the TDM age as calculated from the measured
values and the grouping on the basis of the initial
ratios, (Fig. 3) is illustrated by these samples. For
example, samples HM-4 and HP-3 from the Col-
orado River region have low initial «Nd values
characteristic of province 1, but the T D M ages
are <2.0 Ga. In these cases, we infer that the e ^
value gives the best indication of the magma
source characteristics and that substantial Sm-
Nd fractionation has occurred during magma
genesis at 1.4 Ga, as is indicated by the ex-
tremely low f values (-0.625 and -0.540). This
Figure 2. Sample localities for the western United States. Numbered localities are from this has caused the model age to be lowered signifi-
study (Table 1 and the appendix in the GSA Data Repository). Also included are localities from cantly. Two samples (ANT-1 and GB-2) from
DePaolo (19811a), Nelson and DePaolo (1985), Farmer and DePaolo (1983,1984), and Mus- near the geographic boundary of provinces 1
selwhite (1984). Circles indicate lower Proterozoic rocks; squares, middle Proterozoic anoro- and 2 have isotopic characteristics that are tran-
genic granitoids; and triangles, Mesozoic-Tertiary crustally derived granitoids. sitional between the two provinces. One sample
(DM-lc) from eastern California is clearly
anomalous. It has a very high initial e ^ value
calculated assuming an age of 1.4 Ga. Although
this could be an indication that the magma was
ages that are greater than the crystallization ages, pleted mantle curve, differences between sam-
derived from a mantle source, we are skeptical
we choose 1.8 Ga as a natural cutoff. Rocks ples, and therefore, distinctions between the
of this interpretation because there is a possibil-
with TD M ages <1.8 Ga we interpret as being provinces, still exist.
ity that the age assignment is incorrect. This
derived totally by differentiation of mantle mate- The rocks of each of the three broad
sample has not been directly dated; it is assumed
rials with no admixture of older continental ma- crystallization-age groups show large overall var-
to be 1.4 b.y. old on the basis of a petrologic
terial. The next higher cutoff age was chosen as iations in initial e ^ values (Fig. 4). It is the
correlation with a dated unit (J. L. Anderson,
2.0 Ga. This is based on the upper limit of the magnitude of these variations in comparison to
1986, personal commun.). The Nd isotopic data
T"DM ages found for basement rocks of the Colo- analytical precision and in comparison to innate
would be more consistent with a Mesozoic age
rado Front Range (DePaolo, 1981a). This prov- geological variability, and the fact that geograph-
for this sample. Strontium isotopic data (not re-
ince encompasses most of the Colorado ically coherent subgroupings can be found that
ported herein) also strongly suggest that this
basement and borders the Archean province of have less variability, that allows us to use the
sample has a Mesozoic age.
Wyoming along a sharply defined boundary ex- data to define crustal age provinces. For exam-
posed in southern Wyoming (for example, ple, the initial «Nd values of the ca. 1.7-Ga rocks The distinction between provinces 1 and 3 is
Houston and o :hers, 1979). Within the Archean range from +4.73 to -2.89. This range of 7.6 shown dramatically by the Tertiary and Meso-
province, T D M ages are generally >2.7 Ga, so units is large in comparison to the analytical zoic peraluminous granitic rocks (Fig. 4). Prov-
that there is a major model-age discontinuity uncertainty of ~0.5 units. It is also large in com- ince 1 samples (WM-BC and W236) have eNd
between the Proterozoic rocks and the Archean parison to the total range of e^d values in the values lower than -16.0, and province 3 samples
craton. The TEIM ages that fall between 2.0 and available magmatic source materials at the time. have values between -10 and -12. This rela-
2.7 Ga are confined to the range 2.0 to 2.3 Ga, At 1.7 b.y. ago, the likely magma source mate- tionship has previously been demonstrated by
which we use as the limits of our province 1. rials varied in «Nd values from about +5 for the Farmer and DePaolo (1984). The data clearly
The details of the Nd isotopic variations that mantle to about -11 for rocks of the typical Late indicate a fundamental difference in the nature
lead to our interpretations are discussed below. Archean cratons that were in existence at the of the crust within the different provinces. Al-
As more data become available, the model-age time. The observed range of initial eNd values though the eNd values of the granites reflect and
curve may require refinement, in which case the covers about half the possible range. The fact define their different parental crustal sources, the
model ages presented herein would change that the range is large and the fact that it is much model ages are in general younger than those
slightly. Regardless of any changes in the de- smaller than the possible range are both signifi- indicated by the Proterozoic rocks. This results

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 679

from fractionation of Sm-Nd during magma

genesis; small differences in fsm/Nd f r o m 'he av-
erage source composition, projected over the
>1.5-b.y. time span between crust formation
and derivation of the granites lead to large dis-
crepancies in TDM.
The consistency of results is exemplified by
Nd data for granitoids from the Old Woman-
Piute Mountains of southeastern California. In
this one locality, a lower Proterozoic gneiss (the
Fenner Gneiss), a lower Proterozoic pelitic
schist, and a Cretaceous granite have similar
model ages and unequivocal province 1 char-
acteristics. This can also be demonstrated in the
Colorado Front Range (DePaolo, 1981a) and in
the Bagdad region of Arizona (compare data
from this study with Nelson and DePaolo,
1985).
Overall, the e ^ values for rocks of the three
age groups define distinct and subparallel iso-
topic evolution paths with average fsm/Nd cos-
tal values (Fig. 4). These data indicate that the
crustal segments had separate origins and have
remained distinct with little modification from
new mantle additions throughout their exist-
ence. The three provinces are aligned in order of Figure 3. Crust-formation provinces of the western United States. Provinces are distin-
decreasing model age with increasing distance guished on the basis of regional differences in the Nd isotopic evolution paths as determined by
from the Archean craton. The exception to this measurements of crustally derived samples of different ages (Fig. 4). The localities of Figure 2
generalization is the segment of province 1 lo- are recorded to indicate the initial e Nd and T d m values of individual samples. Diamonds =
cated mainly in southern California. The charac- Archean province (Tdm >2.7 Ga), squares = province 1 (T D M = 2.0-2.3 Ga), circles = province
teristics and distribution of province 1 rocks in 2 (T D M = 1.8-2.0 Ga), triangles = province 3 (T D M = 1.7-1.8 Ga), inverted triangles = Grenville-
this region are of particular interest owing to the age crust (Tdm <1-4 Ga). Boundaries are drawn to group together samples of similar isotopic
disparity between the model ages and the crystal- characteristics. SAF is the San Andreas fault; M.R. is the Mineral Range; G.C. is the Grand
lization ages. In the following discussion, we re- Canyon. Outlined areas are regions where Precambrian basement is exposed.
fer to this portion of province 1 as "Mojavia,"
named for exposures in the Mojave Desert of
southern California. It is characterized isotopi-
cally by the e Nd evolution curve shown in Figure
4, from which model ages in the range of 2.0 to availability of appropriate rock types and the The small differences in the oldest crystallization
2.3 Ga are derived. obscuring of the original geologic relationships ages, the Nd model ages, and initial e^d between
by the effects of subsequent faulting and these two regions are indicative of a gradational
DISCUSSION deformation. change in the crustal composition. The isotopic
The crystallization ages and initial e\<i values data are consistent with a model for the forma-
Crust Formation Provinces of the rocks from provinces 1 and 3 are shown tion of this segment of continental crust by con-
in Figure 4. For province 3, a regular pattern is tinuous incorporation of island-arc material in
A fundamental goal of this research is to un- seen that corresponds to that diagrammed in Fig- the time interval between 1.9 and 1.65 b.y. ago.
derstand the tectonic processes which formed ure 1 for the case in which no significant amount The Archean/province 2 transition is well
the North American continent in the Precam- of mantle-derived material is added to the crust documented stratigraphically and geochronolog-
brian. A large part of the interpretation rests on after its original formation from the mantle, that ically. The best continuous exposures of a Pre-
the nature of the transition zones between the is. t D M = ^crystallization- The evolution path for cambrian transect, with minimal Phanerozoic
provinces. A distinct, sharp boundary implies province 1 is similar, except that it clearly did deformation, are across this boundary in the
juxtaposition of two age terranes along faults or not form purely from mantle-derived material, basement uplifts in the Laramie, Sierra Madre,
sutures. Alternatively, a gradational change in but from a mixture involving a substantial and Medicine Bow Mountains of southeastern
the isotopic characteristics is interpreted as result- amount of older crust. The Nd data from these Wyoming. In this region, we have sampled
ing from continuous formation of the crust by an two provinces show no overlap. This fact, and across the Cheyenne belt (Houston and others,
accretionary process, for example, continental their geographic juxtaposition along the 1978), a shear zone separating Proterozoic and
growth by continuous accumulation of island California-Arizona border, suggests that the Archean basement. The T D m ages from two
arcs. In order to differentiate between these two chosen cutoff ages for the definition of the prov- samples (MB-5 and MB-3), collected 20 km
possibilities, it would be necessary to sample a inces are meaningful. The transition zone be- apart across the Cheyenne belt, show the clear
continuously exposed Precambrian transect tween province 2 and province 3 passes through distinction between the Archean character of the
across the boundaries. A detailed study of these a region with limited exposures of Precambrian basement to the north (TDM > 2.7 Ga) and the
zones is hampered, however, by the limited and Tertiary-Mesozoic plutonic rocks (Fig. 3). Proterozoic basement south of the shear zone

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 680 BENNETT AND DEPAOLO

( T d m ^ 2.0 Ga), with minimal involvement of would explain the extremely sharp contrast in the Colorado River. The Homer Mountains
older crustal ma terial. All Precambrian rocks to Nd isotopic characteristics across this zone. sample is included in the Mojavia terrane on the
the south, as sampled throughout the Colorado The boundary between Mojavia and provinc- basis of the 6^(1.4) = -3.97, the lowest of any
Front Range by DePaolo (1981a), have T D M in es 2 and 3 is best exposed in the Colorado of the middle Proterozoic granitoids in this
the range 1.8-2.0 Ga. As shown in Figure 4, the River region of southern Nevada, California, study. The two samples from the Hualapai
boundary mapped in southeastern Wyoming and Arizona (Fig. 3). The detailed spatial and Mountains have eN(j(1.37) values of -2.78 and
(Houston and others, 1978), and described as a statigraphic relationships between the lower -1.68, which overlap the range of initial e ^ j
suture between the Archean craton and Prot- Proterozoic rocks have been obscured by multi- values for Mojavia (-2.0 to -4.0). The Huala-
erozoic island arcs (Houston and others, 1979; ple intrusive and metamorphic episodes culmi- pai data also overlap the initial eNd values (+0.3
Karlstrom and Houston, 1984), must extend nating with a large amount of crustal extension to -2.5) measured by DePaolo (1981a) for mid-
westward through northern Nevada on the basis and detachment faulting in the Tertiary (for ex- dle Proterozoic granites from the Colorado
of Nd isotopic characteristics although it cannot ample, Stewart, 1980a). Figure 5 shows in Front Range, the extension of province 2 east-
be precisely located. The Nd data indicate that greater detail the sampling density in this region. ward. It is unclear whether the northern Huala-
province 1 pinches out west of the Front Range The maximum spatial resolution of the two prov- pai Mountains belong in province 1 or 2. We
near the Colorado-Wyoming border. This may inces that our sampling allows is - 5 0 km. The prefer to place them tentatively in province 1 on
result from Proterozoic overthrusting (Houston eastern boundary of province 3 passes through the basis of the rather strong contrast in isotopic
and others, 1978) or later strike-slip faulting Lake Mead west of the Gold Butte pluton composition with the nearby rocks in the Bag-
(Karlstrom and Houston, 1984; Duebendorfer [e Nd (1.4 Ga) = -1.13], southward through the dad area of Arizona.
and Houston, 1.986). Either type of faulting Hualapai Mountains, and continues south along Ex])osures of Precambrian supracrustal rocks
within Mojavia are extremely limited, but exist-
ing evidence suggests that the ca. 1.7-Ga
lithologies differ from those of province 3. De-
tailed studies of the supracrustal assemblages
from this region are lacking, but the general de-
scriptions indicate that the section is represented
predominantly by quartzofeldspathic gneisses,
with some pelitic schist and amphibolite. DeWitt
(1980) described the lower Proterozoic supra-
crustal rocks from the Halloran Hills, Nevada, as
95% psammites and siliceous metachert. He
noted the absence of a mafic or felsic volcanic
suite and that this absence is typical of the
southeastern Mojave region. The protoliths are
interpreted as impure quartz arenites and gray-
wackes deposited in a stable tectonic environ-
ment. Miller and others (1982) described a
lower Proterozoic supracrustal sequence from
the Old Woman-Piute Mountains as a thick pile
of metaclastic material with psammitic units and
pelites; amphibolites are widespread, although
not abundant. In contrast, the lower Proterozoic
supracrustal rocks of province 3, as represented
by the Yavapai Series in central Arizona, con-
tain thick sequences of mafic and felsic volcano-
genic rocks. These rocks have been interpreted
(Anderson and Silver, 1976; Condie, 1982) as a
greenstone belt associated with an island-arc
origin. The high initial e ^ values of these rocks
are clearly compatible with an island-arc setting.
Associated with these metavolcanic rocks are
volcanogenic massive sulfide deposits. Province
1, in contrast with provinces 2 and 3, lacks
significant sulfide deposits, with the exception of
the Boriana Canyon deposit in the Hualapai
Mountains. This area has already been noted as
being transitional with respect to Nd isotopic
Crystallization Age (Ga)
composition. Additionally, the supracrustal as-
semblage of amphibolite-grade quartz-muscovite
Figure 4. Initial «Nd versus crystallization age of western U.S. crustally derived granitoids
and quartz-biotite schists and gneisses and
defining separate evolution paths. Solid symbols are data from this study. Open circles are data
amphibolites (Stensrud and More, 1980) has the
from Farmer an d DePaolo (1983,1984); open squares are data from DePaolo (1981a). Ranges
characteristics of both provinces.
of initial values of Penokean-age crust are also indicated.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 681

2.0-2.3 Ga
PiuteA

Marble \ S Old
Mts. / JWoraan
\ / Mts

T u r t l e Mts

Figure 5. Enlarged region of Figure 3 (dashed square), showing details of the sampling distribution at the juncture of three isotopic provinces
in the southern Nevada, eastern California, and western Arizona region. Underlined numbers are initial e^d values for the lower Proterozoic
rocks (T ~ 1.7 Ga), plain numbers are initial e^d values for the middle Proterozoic granitoids (T ~ 1.4 Ga).

Two other sets of observations document the (province 3) to potassic in the eastern California as defined by Kistler and Peterman (1973,
changing characteristics of the crust across the area (province 1), is observed directly in the ca. 1978) on the basis of initial Sr ratios in Meso-
Nd isotopic province boundary. Anderson 1.7-Ga granitoids, the parental rocks of the mid- zoic granitic rocks. Zartman (1974) interpreted
(1987) has observed a change in the mineralogy dle Proterozoic granitoid suite (Anderson, 1987). area la lead as being derived from Archean base-
of the 1.4-Ga anorogenic granite suite from bio- Our map of the crust-formation provinces ment and area lb lead as being derived from
titeihornblende to biotite±muscovite from west (Fig. 3) confirms, in some areas, the divisions Early Proterozoic basement. This Ia/Ib bound-
to east. The variation is mirrored by large suggested by previous workers on the basis of ary corresponds approximately with our prov-
changes in the oxygen fugacity of the magmas. Nd, Sr, and Pb isotopic data and, in other areas, ince 1/province 2 boundary in northern Nevada.
The anorogenic granites of the Marble Moun- indicates substantially different boundaries. It is Recent work by Wooden and others (1986, and
tains (province 1) crystallized under the most illustrative to compare the crustal divisions deter- 1986, personal commun.) has identified area la
reducing conditions. Measured f 0 2 values in- mined by different methods (Fig. 6). Zartman Pb compositions in Mojavia. Pb isotopes do not
crease to the east, including a marked rise record- (1974) divided the western United States into differentiate Nd isotopic provinces 2 and 3.
ed in samples from the Hualapai Mountains. four provinces on the basis of differences in the The Nd isotope crust-formation provinces par-
Within the resolution of Anderson's (1987) Pb isotopic compositions of ores and Mesozoic allel the general trends of the crystallization age
data, the transitions are coincident with the and Cenozoic igneous rocks (Pb types la, lb, II, divisions as defined by Condie (1981) on the
boundary defined herein on the basis of the iso- and III). Provinces II and III (Fig. 6) were inter- basis of a compilation of Rb-Sr whole-rock and
topic data. The characteristics of these crustally preted as reflecting the incorporation of miogeo- U-Pb zircon age determinations. At ~105°W
derivative granitic rocks result from systematic clinal and eugeoclinal sediments, respectively, longitude, for instance, incontrovertible Archean
variability in their source rocks. Additionally, a into the granitic magmas and ores. The bound- basement is present as far south as the Cheyenne
change in the character of the granitic basement ary between them and province I roughly corre- belt of southern Wyoming. Across the Archean/
rocks, from calc-alkaline in central Arizona sponds to the edge of the continental basement Proterozoic age break, the crust becomes pro-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 682 BENNETT AND DEPAOLO

gressively younger to the south. The differing

locations of the crystallization-age and Nd-
isotope province boundaries in Arizona may be
due to the limits chosen as significant for each
type of data. The major distinctions in bounda-
ries as defined by Pb isotopes, Sr isotopes, and
crystallization ages occur in areas identified as
province 1 on th e basis of the Nd isotope data.

Tectonic Implications

The regional Nd isotopic map of the western

United States can be used to test the compatibil-
ity of terranes of unknown affinity (so-called
"suspect terranes") to cratonic western North
America. Two major crystalline rock terranes
have been distinguished by Powell (1981) in the
eastern Transveise Ranges, the Joshua Tree and
the San Gabriel terranes. The San Gabriel ter-
rane contains Prucambrian supracrustal and plu-
tonic rocks, 1.66-1.72 b.y. old on the basis of
U-Pb ages (Silver, 1982). The Joshua Tree ter-
rane consists of lower Proterozoic rocks (1.65
Ga, in Powell, 1981) overlain by mature meta-
sedimentary rocks of unknown age. In this
study, one sample from each of these terranes
(both east of the San Andreas fault) was ana-
lyzed. CW-2 is a pelitic schist from the oldest Figure 6. Comparison of crustal boundaries defined on the basis of crystallization ages
unit in the San Gabriel terrane. The resultant (dashed lines) (Condie, 1981), Pb isotopes (dotted lines) (Zartman, 1974), and Nd isotopes
model age (2.22 Ga) and initial e Nd (-2.89) are (heavy solid lines). Edge of Precambrian basement is as defined on the basis of initial ^ S r / ^ S r
compatible with the terrane being part of the ratios in Tertiary and Mesozoic granitic rocks (Kistller and Peterman, 1978). Ages refer to
Mojavia province. This is support for the sugges- crystallization ages as determined by zircon U-Pb data and whole-rock Rb-Sr data. Roman
tion by Silver (1982) that this terrane extends numerals are lead isotopic provinces.
southeast to the Colorado River and south to
Sonora and furth ermore, that it is exotic to south-
eastern North A merica, at least by comparison DePaolo (1985) have also identified a region in
of the San Andreas fault, supporting the correla-
with the rocks exposed in Arizona. A sample of Wisconsin (the Penokean terrane) having prov-
tion of the two regions. The range of initial e ^
1.67-b.y.-old augen gneiss from the Joshua Tree ince 1 characteristics indicated by the Nd data
values measured in Proterozoic crystalline rocks
terrane (HM-1) yields a model age of 1.98 Ga from rocks having 1.8-Ga and ca. 1.4-Ga crys-
of the eastern and western Transverse Ranges
and an e ^ T ) of +0.32. This initial value is on tallization ages. Patchett and Bridgwater (1984)
and the Mojavia region is not observed in any
the upper end of the range for Mojavia samples measured similar values in the Ketilidian rocks
other rocks of the southwestern United States.
but is much lower than observed for either prov- of southernmost Greenland. As is shown in Fig-
This does not necessarily mean that these ter-
ince 2 or 3. The earliest history of the Joshua ure 7, the anomalous provinces are aligned
ranes were once continuous or directly attached
Tree terrane is thus similar to that of the San across North America. Greenland, restored to its
to one another. Although it is possible to con-
Gabriel terrane, and both terranes are related to pre-drift position, fits into this pattern, although
struct paleotectonic models whereby these ter-
the other basement rocks of southeastern there are some complications across southeast-
ranes may have been rafted onto North America
California. ern Canada. Parts of the north-trending Dakota
from long distances, our data provide no reason
The Precambrian rocks of the Frazier Moun- to suppose that these terranes are particularly Mobile belt (Van Schmus and others, 1987)
tain-Mount Pin DS-Alamo Mountain plutons in far-travelled, beyond the offset known to have may have similar isotopic characteristics. All of
the western Transverse Ranges (west of the San resulted from movement along the San Andreas these regions are bounded on one or both sides
Andreas fault) have been proposed (Frizzell and fault system in late Cenozoic time. by Archean rocks, with the exception of Moja-
Powell, 1982) on the basis of similar lithologic Consideration of the distribution of areas that via. This distribution of province 1-type rocks
units and sequencing relations to be correlative have province 1 isotopic characteristics, both in may represent the original continental configura-
with the rocks of the San Gabriel terrane in and outside of the western United States, is rele- tion, implying that Archean rocks were once
eastern California east of the San Andreas fault. vant to understanding the origin of these ter- present to the west of Mojavia. Alternatively,
Three basement rocks, an amphibolite, a pelitic ranes. Three other areas that have province 1 the province 1-type rocks may have formed as a
schist, and an augen gneiss, from this area were isotopic characteristics have been documented continuous belt with the Archean cratons to the
analyzed. The resultant model ages (for the sam- in North America. The region in northern Ne- north and Proterozoic to the south. This necessi-
ples having sufficiently low Sm/Nd ratios), 2.10 vada and Utah shown in Figure 3 is identified as tates the existence of a major left-lateral shear
Ga and 2.02 Ga, and calculated initial e Nd val- province 1 on the basis of measurements of low zone that transported Mojavia southward. In
ues, +0.92, -1.37, and -0.09, are similar to those initial CNd in Mesozoic and Tertiary granitic light of the apparent truncation of provinces 1
measured in the eastern Transverse Ranges east rocks (Farmer and DePaolo, 1984). Nelson and and 2 by Mojavia and of the profound changes

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 683

This method should be particularly useful for the

case in which the continental fragments have
been separated since the Precambrian and have
had very different subsequent histories.

ORIGIN OF 2.0-2.3 Ga
MODEL AGES

For all of the lower Proterozoic rocks of Moja-

via and related terranes, the published crystalliza-
tion ages are between 1.65 and 1.8 Ga, and the
model ages are 2.0-2.3 Ga. Two possibilities
exist to account for this age difference. (1) The
rocks originally crystallized between 2.0 and 2.3
b.y. ago and were subsequently so severely modi-
fied that little evidence of their original crystalli-
zation age remains, or (2) the rocks sampled are
a mixture of Archean crustal material [eNd(1.7)
-11] and mantle derivatives [eNd(l-7) +5]
and therefore have intermediate initial e\a
values and corresponding older model ages.
If the first case is true, we would expect there
to be some remaining evidence for a 2.0 to 2.3
Ga orogenic episode. Although detailed U-Pb
zircon studies have provided some evidence for
crystallization ages in the range 1.8 to 2.0 Ga, no
zircon data from rocks within Mojavia indicate
an event between 2.0 and 2.3 Ga. Silver and
others (1961) interpreted U-Pb zircon data from
a gneiss in the Panamint Mountains, California,
as reflecting a minimum age of 1.8 Ga. More
recently, J. Wooden (1985, personal commun.)
has determined an upper intercept of 1.95 Ga
for a metavolcanic sample from the New York
Mountains, California. Hedge and others (1983)
determined a zircon age of 2.0 Ga from a drill
Figure 7. Map of North America, showing the distribution of Nd-model age provinces on the core sample from Antelope Island in northern
basis of data from this study, Nelson and DePaolo (1985), Fanner and DePaolo (1983), Utah, but Nd isotopes measured from related
Patchett and Arndt (1986), and Van Schmus and others (1987). M indicates the Mojavia samples give Archean model ages (>2.6 Ga)
terrane. The adjacent dashed curve is the inferred location of a Proterozoic shear. rather than model ages in the 2.0- to 2.3-Ga
range. If there was an orogenic episode some-
time between 2.0 and 2.3 Ga, then an event
in the crustally derived rocks across this bound- brian. Additionally, an area of granulite-grade capable of resetting all U-Pb zircon systematics,
ary, this possibility must be considered. The Proterozoic rocks in southern Nevada straddles with no exceptions, must be postulated, which
amount of motion along the postulated shear is the province 1/province 2 boundary shown in seems unlikely. Although U-Pb isotope systemat-
weakly constrained. Approximately 400 km of Figure 3, and therefore this metamorphism is ics in zircons can be altered by subsequent meta-
motion is required to restore alignment of the quite likely younger than any translational mo- morphic episodes, it would be difficult to
southernmost portion of Mojavia (the area of tion. The granulite metamorphism does not af- envision the wholesale resetting of all ages in a
the Whipple Mountains) with the Nevada re- fect the ca. 1.4-Ga Gold Butte pluton (Thomas heterogeneous suite of rocks over a large area.
gion of province 1. If the Proterozoic shape of and Carneglutti, 1986) and is therefore older Outside of the western United States, two ex-
the Wyoming craton was similar to that of the than Middle Proterozoic and implies that any amples of rocks that may be derived from -2.1-
present day, then ~ 1,000 km of displacement is movement is also older than 1.4 Ga. Radiomet- Ga crystallization-age rocks and that have
required to move Mojavia northward to its ric dating of the metamorphic event could ~2.1-Ga model ages are known to us. One is a
western margin. further constrain the age of the proposed offset. Mississippian orthogneiss from the Yukon Terri-
We can estimate the age of this proposed Mojavia and the correlated province 1-type tory which has T D M = 2.09 Ga (Dusel-Bacon
offset by considering the oldest continuous struc- terranes provide a distinctive marker for re- and Aleinikoff, 1985) and U-Pb upper intercept
tures that cross it. The lower Paleozoic miogeo- gional correlations. Province 1 crust is of limited ages of 2.0 Ga and 2.25 Ga (Aleinikoff and
clinal section is continuous through the Great spatial extent and time duration. Nd data from others, 1981). The other sample is an ortho-
Basin (for example, Stewart, 1980b) and into other continents are limited, but as more data gneiss from the Shuswap Complex in British Co-
southern California (Miller, 1981), without become available, it may be possible to use lumbia (Table 1) with a crystallization age of 2.1
major breaks. This then, constrains the age of crust-formation characteristics as a tool for re- Ga, on the basis of Rb-Sr whole-rock data (R.
latest movement to be older than late Precam- construction of paleocontinental configurations. Armstrong, 1986, personal commun.). These

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 684 BENNETT AND DEPAOLO

data constitute evidence for at least a limited 12

amount of new crust being formed in North
America in the 2.0- to 2.3-Ga age interval. The
relationship, if indeed any exists, between these
8
distant rocks and the Mojavia crystalline base-
ment is totally unknown. 4
Origin of province 2 isotopic characteristics
by mixing of older crustal material with newly
0
forming Proterozoic arc material requires a
mechanism to generate the gap in model ages
between the Archean craton and province 1 and -4
the more gradual transitions between provinces •o
2
1 and 2 and between provinces 2 and 3. The W
model herein proposed is that province 1 formed
by accumulation of arc material onto, or in the - 8
proximity of, the Archean cratonic margin be-
ginning at ca. 1.9 Ga. The newly forming crust
incorporated Archean sedimentary material to - 1 2

yield an older Nd isotopic signature. As the con-

tinent grew outward, the accumulating arc - 1 6
material created an increasingly effective screen
to contributions of material from the Archean
crust. Figure 8 diagrams the progressive Nd iso- - 2 0
topic evolution resulting from this process.
Crystallization Age (Ga)
At Tj, the range of possible e Nd values for
granitoids is from th e mantle value of +4 to the Figure 8. Model of progressive Nd isotopic evolution for the western United States. Paral-
Archean crustal value of-11. The «Nd values for lelograms indicate the possible range of values for each new segment of crust, with the shaded
province 1 could have resulted from a mixture portion indicating the observed initial values. Province 1 isotopic characteristics result from an
of mantle and cruiital components, the latter admixture of Archean crust and material derived from depleted mantle at ~1.9 Ga. Each
being 10% to 25% by mass depending on the Nd succeeding continental addition forms from the same percentage of pre-existing crust and new
concentrations chosen for the end-member mantle material, but the Archean component is screened and diluted until the Nd isotopic
compositions. This newly formed material then composition of the crust is similar to the depleted-mantle Nd isotopic composition. In province
evolves along its own evolution path. At time 3, the crystallization ages of the oldest rocks are thus equal to the model ages.
T 2 , the province 2 granitoids incorporate ap-
proximately the same percentage of pre-existing
crustal material, bvit that pre-existing crust is cordant zircons, the patterns of discordance growth applies to other time periods of Earth
mainly from province 1, which is mostly mantle indicating inheritance of zircons as old as Ar- history is presently unknown; it may be applica-
derived and thus is not very old at the time of chean. This indicates that an Archean compo- ble only to periods of growth following long
formation of province 2. The difference between nent is present in the province 1 crust. intervals of magmatic quiescence such as the pe-
the depleted mantle e Nd value and the e\d value Furthermore, Stacey and Zartman (1978) inter- riod from 2.6 to 1.9 Ga, when apparently very
of the crustal end-member is smaller, so that the preted high 2 0 7 Pb/ 2 0 4 Pb in ores from the Gold little new crust was produced.
Nd isotopic characteristics are less effectively Hill district, Utah (north of the province 1 /prov-
shifted from the mantle values. ince 2 boundary in western Utah), as resulting CONCLUSIONS
At time T 3 , province 3 forms, again by the from recycled Archean crust.
same mixing process, but the isotopic signature Continental growth in this area during the We have determined Nd initial isotopic com-
of the contaminating material (from province 2) time period from ~ 1.9 to 1.65 b.y. ago is seen in positions of Precambrian rocks over a wide areal
is so similar to the mantle value that the effect is our model as a continuous process, with variable (>1.5 x 106 km 2 ) extent in the western United
negligible. Subsequent plutonism and deforma- rates rather than discrete steps. The limited reso- States. This has allowed us to define three Early
tion could further homogenize the isotopic sig- lution of model age information (±0.1 Ga) and Proterozoic crustal age provinces. We have used
nature of both the sedimentary rocks and the arc eNd values (±0.5 units) precludes our discerning the patterns of crust formation characteristics to
rocks in each province. Measurements of high the detailed timing of the perturbations, but address the problem of crustal genesis and re-
model ages (T D M = 2.0-2.3 Ga) in lower Prot- these may yield to detailed chronological study lated tectonic problems. East of the lower Colo-
erozoic rocks have been made by Wilson and using the U-Pb zircon method. The typical arc- rado River region, the Nd initial ratios (and
others (1985) in Sweden and Patchett and forming process may always involve significant model ages) change in a manner parallel to the
Bridgwater (1984) in Greenland. These studies amounts of pre-existing crustal material. The crystallization age trends, that is, becoming
similarly concluded that the data indicate mix- case we observe in the southwest Untied States, younger southward away from the Archean of
ing of Archean and Proterozoic material, al- where crust building went on at a rapid rate for Wyoming. This is interpreted as reflecting con-
though different mechanisms are proposed. at least 200 m.y., is one in which the continent tinuous continental growth without major tec-
Additional evidence for crustal mixing is from grew outward until the material being added tonic disruption. West of the California-Arizona
the U-Pb data of Chen and Moore (1979) from was of similar age and isotopic characteristics to border, the Nd isotopic trends crosscut crystalli-
the Independence dike swarm of eastern Cali- the existing adjacent crust. zation age boundaries. Discontinuities in the
fornia. Within these Jurassic dike rocks are dis- The extent to which this model of crustal metamorphic grade and in the nature of the Prot-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user
 NEODYMIUM ISOTOPIC MAPPING, WESTERN UNITED STATES 685

Hurley, P. M„ and Rand, J. R., 1969, Pre-drift continental nuclei: Science,

erozoic supracrustal sedimentary and intrusive Both authors thank J. Aleinikoff for samples v. 164, p. 1229-1242.
sequences across this isotopic boundary further from the Mineral Mountains, Utah; A. Arnold Jacobsen, S. B., and Wasserburg, G. J., 1980, Sm-Nd isotopic evolution of
chondrites: Earth and Planetary Science Letters, v. 50, p. 139-155.
indicate its significance. We interpret the model- for Gunnery Range samples; R. S. Babcock for Karlstrom, K. E„ and Houston, R. S., 1984, The Cheyenne Belt: Analysis of a
Proterozoic suture in southern Wyoming: Precambrian Research, v. 25,
age discontinuity as resulting from an offset of samples from the Grand Canyon; and J. L. p. 415-446.
>400 km of the Mojave region along a mid- Anderson for a suite of anorogenic granites. This Kistler, R. W., and Peterman, Z. E., 1973, Variations in Sr, Rb, K, Na, and
initial 8 7 Sr/ 8 6 Sr in Mesozoic granitic rocks and intruded wall rocks in
Proterozoic fault from a region that has similar work was funded by National Science Founda- central California: Geological Society of America Bulletin, v. 84,
p.3489-3512.
isotopic characteristics to the north. This demon- tion Grants EAR 82-12377 and EAR 84-15143. 1978, Reconstruction of crustal blocks of California on the basis of
strates the potential value of isotopic studies as a initial strontium isotopic compositions of Mesozoic granitic rocks: U.S.
Geological Survey Professional Paper 1071,17 p.
large-scale mapping method to discern structure Krogh, T. E., 1982a, Improved accuracy of U-Pb zircon ages by the creation of
more concordant fractions using a high gradient magnetic separation
in Precambrian terranes, even in areas where the REFERENCES CITED
technique: Geochimica et Cosmochimica Acta, v. 46, p, 631-636.
geologic relationships are obscured by many Aleinikoff, J. N., Dusel-Bacon, C., Foster, H. L., and Futa, K., 1981, Protero- 1982b, Improved accuracy of U-Pb zircon ages by the creation of more
zoic zircon from augen gneiss, Yukon-Tana na Upland, east-central concordant fractions using an air abrasion technique: Geochimica et
generations of later metamorphic and tectonic Alaska: Geology, v. 9, p. 469-473. Cosmochimica Acta, v. 46, p. 637-650.
Lugmair, N. B„ Scheinin, S., and Marti, K., 1975, Sm-Nd age of Apollo 17
events. The validity of this approach has been Anderson, C. A., and Silver, L. T., 1976, Yavapai Series—A greenstone belt:
basalt 75075: Evidence for early differentiation of the Lunar exterior:
Arizona Geological Society Digest 10, p. 13-21.
partially confirmed in correlating Precambrian Anderson, J. L., 1987, Proterozoic anorogenic granites of the southwestern Lunar Science Conference, 6th, Proceedings, p. 1419-1429.
U.S.: Arizona Geological Society Digest, IS, in press. McCulloch, M. T., and Wasserburg, G. J., 1978, Sm-Nd and Rl>Sr chronology
terranes displaced across the San Andreas fault Ben Othman, P., Polve, M., and Allegre, C. J., 1984, Nd-Sr isotopic composi- of continental crust formation: Science, v. 200, p. 1003-1011.
Miller, C. A., Howard, K. A., and Hoisch, T. D., 1982, Mesozoic metamor-
zone, regions previously correlated on the basis tion of granulites and constraints on the evolution of the lower continen-
phism, thrusting and plutonism, Old Woman-Piute Range, southeast-
tal crust: Nature, v. 307, p. 510-515.
of similar stratigraphy. Bickford, M. E., Harrower, K. L., Hoppe, W. J., Nelson, B. K., Nusbaum, R. L., ern California, in Frost, E. G., and Martin, D. L., Mesozoic-Cenozoic
and Thomas, J. J., 1981, Rb-Sr and U-Pb geochronology and the tectonic evolution of the Colorado River region, California, Arizona,
The western United States provides a unique distribution of rock types in the Precambrian basement of Missouri and and Nevada: San Diego, California, Cordilleran Publishers, p. 561-581.
Kansas: Geological Society of America Bulletin, Part 1, v. 92, Miller, E. L., 1981, Geology of the Victorville region, California: Geological
situation for the study of crust-forming proc- p. 323-341. Society of America Bulletin, Part I, v. 92, p. 160-163.
Musselwhite, D. S., 1984, Isotopic and trace element study of the Woods
esses, as it represents a strip of -1,300 km Chen, J. H., and Moore, J. G., 1979, Late Jurassic Independence dike swarm in
Mountain caldera complex [M.S. thesis]: Los Angeles, California, Uni-
eastern California: Geology, v. 7, p. 129-133.
width accreted to the Archean margin without Condie, K. C., 1981, Precambrian rocks of the southwestern United States and versity of California, 127 p.
adjacent areas of Mexico: New Mexico Bureau of Mines and Mineral Nelson, B. K., and DePaolo, D. J., 1984, 1,700 Myr greenstone volcanic
significant subsequent disruption. This enables Resources Resource Map 13, scale 1:1,000,000. successions in southwestern North America and isotopic evolution of
the Proterozoic mantle: Nature, v. 312, p. 143-146.
study of the process of crust formation over a 1982, Plate-tectonics model for Proterozoic continental accretion in the
1985, Rapid production of continental crust 1.7-1.9 b.y. ago: Nd iso-
southwestern United States: Geology, v. 10, p. 37-42.
300-m.y. time interval. Additionally, the margin DePaolo, D. J., 1978, Study of magma sources, mantle structure and the topic evidence from the basement of the North American midcontinent:
differentiation of the Earth using variations of 143 Nd/ 144 Nd in igneous Geological Society of America Bulletin, v. 96, p. 746-754.
was - 8 0 0 m.y. old at the time the process rocks [Ph.D. thesis]: Pasadena, California, California Institute of Tech- Papanastassiou, D. A., and Wasserburg, G. J., 1969, Initial strontium isotopic
nology, 360 p. abundances and the resolution of small time differences in the formation
started and therefore of very different isotopic of planetary objects: Earth and Planetary Science Letters, v. 5,
1980, Sources of continental crust: neodymium isotope evidence from
composition, thus allowing the Archean com- the Sierra Nevada and Peninsular Ranges: Science, v. 209, p. 684-687. p. 361-376.
1981a, Neodymium isotopes in the Colorado Front Range and crust- Patchett, P. J., and Arndt, N. T„ 1986, Nd isotopes and tectonics of 1.9-1.7 Ga
ponent of the new crust to be determined. The mantle evolution in the Proterozoic: Nature, v. 291, p. 193-196. crustal genesis: Earth and Planetary Science Letters, v. 78, p. 329-338.
Patchett, P. J., and Bridgwater, D., 1984, Origin of continental crust of 1.9-1.7
scenario for crustal development of the western 1981b, A neodymium and strontium isotopic study of the Mesozoic
Ga age defined by Nd isotopes in the Ketilidian terrain of South Green-
calc-alkaline granite batholiths of the Sierra Nevada, and Peninsular
United States proposed on the basis of Nd iso- Ranges, California: Journal of Geophysical Research, v. 86, land: Contributions to Mineralogy and Petrology, v. 87, p. 311-318.
p. 10470-10488. Patchett, P. J., Kuovo, O., Hedge, C. E., and Tatsumoto, M., 1981, Evolution
topic data is consistent with the emerging tec- 1983, The mean life of continents: Estimates of continent recycling rates of continental crust and mantle heterogeneity: Evidence from Hf iso-
topes: Contributions to Mineralogy and Petrology, v. 78, p. 279-297.
tonic model for the western United States as from Nd and Hf isotopic data, and implications for mantle structure:
Powell, R. E., 1981, Geology of the crystalline basement complex, eastern
Geophysical Research Letters, v. 10, p. 705-708.
representing a series of arc sequences (compare DePaolo, D. J., and Wasserburg, G. J., 1976, Inferences about magma sources Transverse Ranges, southern California: Constraints on regional tec-
and mantle structures using variations of l43 Nd/ 144 Nd: Geophysical tonic interpretation [Ph.D. thesis]: Pasadena, California, California Insti-
with Karlstrom and Houston, 1984; Condie, Research Letters, v. 3, p. 743-746. tute of Technology, 441 p.
Silver, L. T., 1982, Evolution of the crystalline rock terranes of the Transverse
1982) accreted to an Archean cratonic margin. DeWitt, E., 1980, Geology and geochemistry of the Halloran Hills, southern
Ranges, southern California: Geological Society of America Abstracts
California, and implications concerning the Mesozoic tectonics of the
southeastern Cordillera [Ph.D. thesis]: State College, Pennsylvania, with Programs, v. 14, p. 234.
The model presented to account for the Nd Pennsylvania State University, 322 p. Silver, L. T., McKinney, C. R„ and Wright, L. A., 1961, Some Precambrian
isotopic signatures implies that all newly form- Duebendorfer, E. M., and Houston, R. S., 1986, Kinematic history of the ages in the Panamint Range, Death Valley, California [abs.]: Geological
Cheyenne belt, Medicine Bow Mountains, southeastern Wyoming: Society of America Special Paper 68, p. 55.
ing crust in the vicinity of a continental margin Geology, v. 14, p. 171-174. Stacey, J., and Zartman, R. E., 1978, Lead and strontium isotopic study of
Dusel-Bacon, C., and Aleinikoff, J. A., 1985, Petrology and tectonic signifi- igneous rocks and ores, Gold Hill district, Utah: Utah Geology, v. 5,
incorporates a significant amount (-20%) of the cance of augen gneiss from a belt of Mississippian granitoids in the p. 1-15.
pre-existing crust. This observation lays the Yukon-Tanana terrane, east-central Alaska: Geological Society of Stensrud, H. L., and More, S., 1980, Precambrian geology and massive sulphide
America Bulletin, v. 96, p. 411-425. environments of the west-central Hualapai Mountains, Mohave County,
groundwork for future studies to answer ques- Farmer, G. L., 1985, A Sm-Nd isotopic study of Precambrian/Cambrian sedi- Arizona—A preliminary report: Arizona Geological Society Digest 12,
mentary provenance in the Great Basin, and implications for the tec- p. 155-165.
tions concerning the mechanisms of crustal tonic evolution of the western U.S.: Geological Society of America Stewart, J. H., 1980a, Basin-Range structure in western North America: A
growth and concerning the differences in the Abstracts with Programs, v. 17, p. 578. review, in Smith, R. B,, and Eaton, G. P., eds., Cenozoic tectonics and
Farmer, G. L., and DePaolo, D. J., 1983, Origin of Mesozoic and Tertiary regional geophysics of the western Cordillera: Geological Society of
process of arc formation through Earth history. granites in the western U.S. and implications for pre-Mesozoic crustal America Memoir 152, p. 1-32.
structure, 1. Nd and Sr isotopic studies in the geocline of the northern 1980b, Geology of Nevada: Nevada Bureau of Mines and Geology Spe-
For example, in the Archean, were similar or Great Basin: Journal of Geophysical Research, v. 88, p. 3379-3401. cial Publication 4,136 p.
different amounts of pre-existing crustal material 1984, Origin of Mesozoic and Tertiary granites in the western U.S. and Thomas, W. M., and Cameglutti, D. H., 1986, Granulite fades metamorphism
implications for pre-Mesozoic crustal structure, 2. Nd and Sr isotopic in the Gold Butte area, Clark Co., Nevada: Geological Society of Amer-
incorporated into newly forming crust via sub- studies of unmineralized and Cu- and Mo-mineralized granites in the ica Abstracts with Programs, v. 18, p. 192.
Precambrian craton: Journal of Geophysical Research, v. 89, Van Schmus, W. R., Bickford. M. E., and Zietz, I., 1987, Early and Middle
duction? Does incorporation of older crustal p. 10141-10160. Proterozoic provinces in the central United States: American Geophysi-
material take place in the mantle or by assimila- Frizzell, V. A., Jr., and Powell, R. E., 1982, Crystalline rocks near Frazier and cal Union Geodynamic Series, v. 15, p. 43-68.
Alamo Mountains, western Transverse Ranges, California: A compara- Wilson, M. R., Hamilton, P. J., Fallick, A. E., Aftalion, M., and Michard, A.,
tion of material in the lower crust? Furthermore, tive study: Geological Society of America Abstracts with Programs, 1985, Granites and Early Proterozoic crustal evolution in Sweden: Evi-
v. 14, p. 164. dence from Sm-Nd, U-Pb, and O isotope systematics: Earth and Plane-
does the presence of a variable Archean compo- Haskin, L. A., Frey, F. A., Schmitt, R. A., and Smith, R. H., 1966, Meteorite, tary Science Letters, v. 72, p. 376-388.
nent in some crustal regions correlate with and solar, and terrestrial rare-earth distributions: Physics and Chemistry of Wooden, J., Stacey, J., Howard, K., and Miller, D., 1986, Crustal evolution in
the Earth, v. 7, p. 167-321. southeastern California: Constraints from Pb isotopic studies: Geologi-
explain chemical variations in the basement Hedge, C. E„ Stacey, J. S., and Bryant, B., 1983, Geochronology of the Far- cal Society of America Abstracts with Programs, v. 18, p. 200.
mington Complex, Wasatch Mountains, Utah: Geological Society of Zartman, R. E., 1974, Lead isotopic provinces in the Cordillera of the western
rocks? America Memoir 157, p. 37-44. United States and their geologic significance: Economic Geology, v. 69,
Hoffman, P. F„ and Bowring, S. A., 1984, Short lived 1.9 Ga continental p. 792-805.
margin and its destruction, Wopmay orogen, northwest Canada: Geol-
ACKNOWLEDGMENTS ogy, v. 12, p. 68-72.
Houston, R. S„ 1978, A regional study of rocks of Precambrian age in that pari
of the Medicine Bow Mountains lying in southeastern Wyoming: Geo-
logical Survey of Wyoming Memoir 1,167 p.
Bennett thanks C. Miller, K. A. Howard, Houston, R. S., Karlstrom, K. E., Hills, F. A., and Smithson, S. B., 1979, The
R. Powell, and D. Miller for generously sharing Cheyenne Belt: A major Precambrian crustal boundary in the western
United States: Geological Society of America Abstracts with Programs,
MANUSCRIPT RECEIVED BY THE SOCIETY JULY 3 , 1 9 8 6
REVISED MANUSCRIPT RECEIVED MARCH 3 0 , 1 9 8 7
their time and knowledge on collecting trips. v. 11, p. 446. MANUSCRIPT A C C E P T E D A P R I L 2 , 1987

Printed in U.S.A.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/99/5/674/3998409/i0016-7606-99-5-674.pdf

by Univ Alberta Library user