Ages of the Steens and Columbia River Flood Basalts and Their

Home , Central Oregon, Steens Mountain

Ages of the Steens and Columbia River ¯ood basalts and their relationship to extension-related calc-alkalic volcanism in eastern Oregon

P.R. Hooper* G.B. Binger Department of Geology, Washington State University, Pullman, Washington 99164, USA K.R. Lees Department of Earth Sciences, Open University, Milton Keynes MK7 6AA, UK

ABSTRACT INTRODUCTION But the causal relationship between extension and the Columbia River ¯ood-basalt eruption Stratigraphic and chemical correlations The unusually large volumes of tholeiitic is more dif®cult to establish. The problem is of Tertiary volcanic units in eastern Oregon basalt, erupted within short time spans, that complicated by uncertainty about the relation- con®rm that the Steens Basalt represents characterize continental ¯ood-basalt provinces ships between the eruption of Steens Basalt the earliest eruptions of the Columbia Riv- are typically associated with lithospheric ex- (Carlson and Hart, 1987) over the impinging er ¯ood-basalt province. Field correlations tension (White and McKenzie, 1989). In some Yellowstone hotspot in southeast Oregon and are supported by major and trace element large-scale ¯ood-basalt provinces, extension is the larger eruption of the Columbia River Ba- analyses and con®rmed by 40Ar/39Ar dates. accompanied by lithospheric thinning and an- salt Group almost 300 km farther north on the Within the basalt of Malheur Gorge, situ- ticipates the separation of major tectonic Columbia Plateau. Similarity in age and composition has led many authors to assume that ated between Steens Mountain and the plates. More controversial is whether exten- the Steens Basalt was the earliest eruption of southernmost extent of the previously sion was the cause or consequence of the mas- the Columbia River ¯ood-basalt province mapped Columbia River Basalt Group, the sive ¯ood-basalt eruptions (White and Mc- (Brandon and Goles, 1988; Geist and Rich- lowest unit correlates with the Steens Ba- Kenzie, 1989; Richards et al., 1989; Hooper, ards, 1993; Camp, 1995; Hooper, 1997; Ta- salt, and the conformably overlying middle 1990). kahashi et al., 1998, among others), but the and upper units correlate with the Imnaha The prevalent view of the origin of the Co- stratigraphic relationships between the two and Grande Ronde Basalt Formations of lumbia River and other continental ¯ood ba- tholeiitic eruptions have never been estab- the Columbia River Basalt Group. New salts is that the large magma volumes that dis- lished. Furthermore, the radiometric ages sug- dates indicate that Imnaha and Grande tinguish ¯ood basalts require a hotspot in the gested that the Imnaha Basalt, the lowest for- Ronde Basalt Formations on the Columbia upper mantle. Many workers have ascribed mation of the Columbia River Basalt Group, Plateau (Ͼ90% of the Columbia River Ba- this hotspot to a plume arising from deep was older (17.2 Ma Ϯ 0.5 Ma; McKee et al., salt Group) erupted between 16.1 and 15.0 within the mantle (Campbell and Grif®ths, 1990), with or without the additional process- 1981) than more recent and precise 40Ar/39Ar Ma. These were immediately preceded by dates of the Steens Basalt (16.60 Ma Ϯ 0.02 the Steens Basalt, a plagioclase-phyric tho- es of signi®cant lithospheric extension and thinning (White and McKenzie, 1989; Rich- Ma; Swisher et al., 1990). leiite that erupted above the calculated po- ards et al., 1989; Hooper, 1990). In contrast, Subsequent mapping of the Tertiary volca- sition of the Yellowstone hotspot at 16.6 Carlson and Hart (1987) have ascribed the Co- nic rocks in part of east-central Oregon (Ferns Ma. In eastern Oregon, the ¯ood-basalt lumbia River basalt eruption to extension as- et al., 1993a, 1993b; Cummings et al., 2000), tholeiites of Steens Mountain and Malheur sociated with backarc spreading. Anderson together with analyses for major and trace el- Gorge form a voluminous but brief inter- 40 39 1 (1994) and his coauthors have argued that ements and new Ar/ Ar dates, has clari®ed lude (16.6±15.3 Ma) superimposed on the continental ¯ood basalts tend to occur along the stratigraphic relationship between the low-volume, calc-alkalic to mildly alkalic, boundaries between lithospheric plates of con- Steens Basalt and the Columbia River Basalt volcanism associated with continuing Eo- trasting thickness, which would allow creation Group, allowing them both to be linked to im- cene to present east-west extension. of convection systems adequate to account for pingement of the Yellowstone hotspot beneath the large magma volumes. Keywords: calc-alkalic volcanism, Colum- East-west extensional stress was prevalent 1 GSA Data Repository item 2002014, major and bia River Basalt, extension, hotspot, Steens during the eruption of the Columbia River Ba- trace element analyses, details of 40Ar/39Ar dates, Basalt. and petrographic descriptions of the volcanic units salt Group, as graphically illustrated by the of east-central Oregon, is available on the Web at consistent north-northwest±east-southeast ori- http://www.geosociety.org/pubs/ft2002.htm. Re- *E-mail: [email protected]. entation of the feeder dikes (Hooper, 1997). quests may also be sent to [email protected].

GSA Bulletin; January 2002; v. 114; no. 1; p. 43±50; 5 ®gures; 1 table; Data Repository item 2002014.

For permission to copy, contact [email protected] ᭧ 2002 Geological Society of America 43

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020 HOOPER et al.

TABLE 1. STRATIGRAPHY AND 40Ar/39Ar DATES FOR VOLCANIC UNITS eastern Oregon at 16.6 Ma (Engebretson et al., OF EAST-CENTRAL OREGON 1985; Pierce and Morgan, 1992). This study provides improved dates for the initial eruptions of the whole ¯ood-basalt province and emphasizes the distinction between the calc- alkalic to mildly alkalic magmatism associated with east-west lithospheric extension and the tholeiitic ¯ood basalts associated with the Yellowstone hotspotÐa distinction with important implications for interpreting the origin of ¯ood basalts.

VOLCANIC STRATIGRAPHY IN EAST- CENTRAL OREGON

Many of the volcanic units in east-central Oregon have been given formal names by earlier workers (e.g., Kittleman et al., 1965, 1967). Lees (1994) grouped these and other units into informal sequences of similar age (Table 1), a convenient nomenclature that is followed here.

Basalt of Malheur Gorge

The oldest Tertiary volcanic sequence in the area studied, between Farewell Bend, Vale, and Juntura (Fig. 1), is the basalt of Malheur Gorge (Evans, 1990a, 1990b). It is divided into three distinct stratigraphic and compositional units (Table 1; Lees, 1994; Binger, 1997). At the base, the Lower Pole Creek unit is composed of dark, coarsely plagioclase- phyric basalt. Both chemically and petro- graphically, the Lower Pole Creek ¯ows are similar to ¯ows with reversed magnetic polarity in the lower part of the Steens Basalt section at Steens Mountain, 100 km to the south- west (Fig. 2; Binger, 1997; Johnson et al., 1998a). Analogous, coarsely plagioclase-phyric basalts with reversed magnetic polarity form the basal ¯ow unit of the Columbia Riv- ¯ows are overlain by the aphyric basaltic an- Birch Creek unit have a mean of 15.7 Ma Ϯ er Basalt Group at Squaw Butte, a few kilo- desites of the Birch Creek unit, which are pet- 0.1 Ma (Table 1). Two ¯ows from original meters northwest of Boise, Idaho (Figs. 1 and rographic and chemical equivalents of the sections of Imnaha basalt on the Columbia 2; Fitzgerald, 1984; Martin, 1984), and also Grande Ronde Basalt Formation of the Co- Plateau (Hooper et al., 1984) provide new form the magnetically reversed ¯ows at the lumbia River Basalt Group (Fig. 2; Lees, ages of 15.4 Ma Ϯ 0.2 Ma and 15.5 Ma Ϯ base of the otherwise magnetically normal Im- 1994). Only a few Birch Creek ¯ows are pre- 0.3 Ma, determined by using the same 40Ar/ naha Basalt along the southern margin of the sent at Malheur Gorge, and they have not been 39Ar techniques in the laboratory used to date Wallowa Mountains in northeast Oregon (e.g., recognized farther south. The number of the Grande Ronde Basalt on the Columbia sample C1 of Carlson, 1984; Hooper, 1997). Grande Ronde type ¯ows and their total thick- Plateau at between 16 and 15 Ma (Long and The Lower Pole Creek ¯ows are succeeded ness increase progressively north to the Co- Duncan, 1982; R.A. Duncan, Oregon State upward, without apparent break, by the mod- lumbia Plateau (Lees, 1994; P.R. Hooper, un- University, 1998, personal commun.). Dates erately plagioclase-phyric Upper Pole Creek published mapping, 1981). of the Imnaha and Grande Ronde Basalt ¯ows tholeiitic ¯ows with higher concentrations of Flows of Steens Basalt on Steens Mountain show no discernible correlation with strati-

SiO2, Al2O3, K2O, Ba, Rb, and Zr than are are dated at 16.60 Ma Ϯ 0.02 Ma (base) and graphic position (Long and Duncan, 1982). found in the Lower Pole Creek unit (Fig. 2). 16.59 Ma Ϯ 0.02 Ma (top) (Swisher et al., More recent 40Ar/39Ar dates for Grande Ronde The compositional range of the Upper Pole 1990). Two new dates for the Lower Pole Basalt by Baksi (1989) have a similar range Creek ¯ows is similar to the main magneti- Creek ¯ows (basalt of Malheur Gorge) have a (15.5 Ma to 16.1 Ma). The often quoted older cally normal section of the Imnaha Basalt of weighted mean of 16.9 Ϯ 0.8 Ma. Four new K/Ar dates for ¯ows of the Imnaha Basalt For- the Columbia River Basalt Group (Fig. 2; dates for Upper Pole Creek ¯ows have a mean mation fall outside this range at 17.2 Ma Ϯ Hooper et al., 1984). The Upper Pole Creek of 16.5 Ma Ϯ 0.3 Ma, and three from the 0.2 Ma (McKee et al., 1981). We suggest that

44 Geological Society of America Bulletin, January 2002

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020 AGES OF THE STEENS AND COLUMBIA RIVER FLOOD BASALTS

Figure 1. (A) Regional sketch map showing the distribution of Miocene ¯ood basalts in the Paci®c Northwest. These include the Steens Basalt, the basalt of Malheur Gorge (BMG), and the Columbia River Basalt Group (CRBG). Note that the Steens Basalt may have been displaced to the west by posteruptive east-west extension that formed the Oregon-Idaho graben (OIG) and by oblique right-lateral movement along the western Snake River Plain (WSRP) (Hooper et al., 2002). BÐBoise, CÐCornucopia dike swarm, DÐJohn Day basin ®lled with the Picture Gorge Basalt (Columbia River Basalt Group) fed by the Monument dike swarm, FÐFarewell Bend, GRÐ Grande Ronde dike swarm, JÐJuntura and Malheur Gorge, cut by the west-northwest±trending Malheur Gorge fault, NNRÐNorth Nevada rift, OWLÐOlympic-Wallowa lineament, PÐPortland, SÐSeattle, SBÐSquaw Butte, SMÐSteens Mountain, VÐVale fault zone, VIÐVancouver Island, WÐWallowa Mountains (horst). ``Suture'' refers to the Idaho suture zone, a lithospheric boundary in Idaho and Washington between the older, thicker, North American craton to the east and north and the Blue Mountains accreted oceanic terranes to the west (Armstrong et al., 1977; Fleck and Criss, 1985). (B) Cartoon of stratigraphic relationships among the Steens Basalt, the basalt of Malheur Gorge (including LPCÐLower Pole Creek, UPCÐUpper Pole Creek, and Birch Creek) and the formations of the Columbia River Basalt Group (CRBG) between Steens Mountain and the Columbia Plateau. Not to scale.

Geological Society of America Bulletin, January 2002 45

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020 HOOPER et al.

trend. These units were derived from distinct, probably crustal, sources. Seven units of the Hog Creek sequence have been dated (Table 1). Their ages overlap and generally lie within the range of a single unit (Table 1). Of these, the older dates ob- tained for the Little®eld Rhyolite are discard- ed as unreliable because this rhyolite is de- monstrably the youngest Hog Creek unit in the stratigraphic sequence. The weighted mean of the remaining dates is 15.3 Ma Ϯ 0.1 Ma. In summary, the ®nal products of the tholeiitic eruptions (the Hunter Creek Basalt) are inter®ngered with the silicic units of the Hog Figure 2. Ba vs. SiO2 plot of the Hunter Creek Basalt, the three units of the basalt of Malheur Gorge (BMG), and the magnetically reversed ¯ows at the base of the Squaw Creek sequence, and both are contemporane- Butte section (in Idaho; see Fig. 1 for location) of the Columbia River Basalt Group ous with the initial opening of the Oregon- (Martin, 1984, and this study). Fields for comparison are Grande Ronde Basalt (GRB) Idaho graben (Table 1). The date 15.3 Ma Ϯ and Imnaha Basalt of the Columbia River Basalt Group (Hooper et al., 1984; Hooper, 0.1 Ma represents both the end of the ¯ood- 2000) and the lower 43 magnetically reversed ¯ows of Steens Basalt (SB) at Steens Moun- basalt eruption in this part of east-central tain (Binger, 1997; Johnson et al., 1998a). Note the similarity in chemical composition (1) Oregon and the of the formation of the between the lowest of the three units (Lower Pole Creek) of the basalt of Malheur Gorge, Oregon-Idaho graben. the magnetically reversed ¯ows at the base of the basalt section at Squaw Butte, and the ®eld of reversely magnetized lower ¯ows of the Steens Basalt; and (2) between the upper Tims Peak Basalt two units (Upper Pole Creek and Birch Creek) of the basalt of Malheur Gorge and the ®elds of the Imnaha and Grande Ronde Basalts of the Columbia River Basalt Group on The Hog Creek sequence is unconformably the Columbia Plateau. Note also that the Hunter Creek Basalt is stratigraphically part of overlain by a series of distinct primitive, dik- the Hog Creek sequence but is interpreted as the ®nal eruption of the tholeiitic magma tytaxitic, nonporphyritic, olivine-bearing ba- (basalt of Malheur Gorge). The results of 700 major and trace element whole-rock X-ray salt ¯ows that erupted from a number of dif- ¯uorescence analyses from east-central Oregon used in this study are available (see foot- ferent centers along the northwest margin of note 1) from [email protected]. the Oregon-Idaho graben (Fig. 1; Table 1). These ¯ows were formally named the Tims Peak Basalt by Kittleman et al. (1965, 1967). these older dates can now be discounted, and Dinner Creek Ash-Flow Tuff (Evans, 1990a), The Tims Peak Basalt has a chemical trend we concur with R.A. Duncan (1999, personal whereas the rhyolite of Cottonwood Mountain distinct from that of the tholeiitic basalt of commun.) that all Imnaha and Grande Ronde occurs both below and, more typically, above Malheur Gorge, which coincides with the Basalt ¯ows (Ͼ90% of the Columbia River the Hunter Creek Basalt (Binger, 1997), sug- primitive end of the younger Keeney calc-al- Basalt Group on the Columbia Plateau) were gesting that these two eruptions were contem- kalic rock types (Fig. 4). The geographically erupted between 16.1 Ma and 15.0 Ma. poraneous. As both the rhyolite of Cotton- separate Tims Peak Basalt eruptive centers wood Mountain and the Little®eld Rhyolite each produced magmas with subtle but dis- Hog Creek Sequence are aligned along the faults that form the west- tinctly different compositions (Fig 3B; Binger, ern margin of the Oregon-Idaho grabenÐcut- 1997). New age determinations of the Tims In the Malheur Gorge area on the northwest ting across some of the graben boundary faults Peak Basalt have a weighted mean of 13.5 Ma shoulder of the Oregon-Idaho graben (Fig. 1; and cut by othersÐboth rhyolites are broadly Ϯ 0.1 Ma (Table 1). Cummings et al., 2000), the Birch Creek ¯ows contemporaneous with the initial faulting that of the basalt of Malheur Gorge are overlain Younger Eruptions Across the Oregon- formed the graben and with each other as ob- conformably by a conspicuous suite of vol- Idaho Graben served by Ferns et al. (1993a) and Cummings canic units of similar age that is assigned to et al. (2000). the Hog Creek sequence (Lees, 1994). These Younger volcanic activity occurs as small The Hunter Creek Basalt lies on the same include the Dinner Creek Ash-Flow Tuff local eruptions spread across the Oregon-Ida- chemical trend as the basalt of Malheur (DCT; Table 1; Haddock, 1967), the rhyolite ho graben. All are calc-alkalic or mildly al- of Cottonwood Mountain, the Hunter Creek Gorge. Thus, although stratigraphically a kalic, separated by thick sequences of terres- Basalt, and the Little®eld Rhyolite (Kittleman member of the Hog Creek sequence, the Hunt- trial sediment that accumulated in the et al., 1965, 1967; Ferns et al., 1993a, 1993b; er Creek Basalt is interpreted genetically as developing subsidiary grabens (Cummings et Binger, 1997; Cummings et al., 2000). the ®nal eruption of the tholeiitic magma (Fig. al., 2000). The Keeney sequence (13.4 Ma to These units are all conformable on the ba- 2). In contrast, the various silicic units display 10.1 Ma; Table 1; Lees, 1994) includes nu- salt of Malheur Gorge, and ®eld evidence im- scattered trace element concentrations (Fig. merous eruptive calc-alkalic centers. Despite plies that all units of the Hog Creek sequence 3A) and lack a positive geochemical trend to- their overall chemical similarities, each Kee- erupted over a very short period of time. In- ward higher incompatible element abundanc- ney eruptive center is distinguished by subtle tercalated bands and rock fragments of Hunter es, so they cannot have been derived from a but distinctive chemical differences (Fig. 3B). Creek Basalt composition are present in the basaltic parent along a single fractionation At the northern end of the Oregon-Idaho

46 Geological Society of America Bulletin, January 2002

Figure 3. Plots illustrating the consistent differences in chemical composition between volcanic units in east-central Oregon. (A) The Sr vs. Ba plot distinguishes the most important silicic units. CWRÐrhyolite of Cottonwood Mountain, DCÐDevine Canyon Ash-Flow Tuff, DCTÐDinner Creek Ash-Flow Tuff, LRÐLittle®eld Rhyolite, WB-1 and WB-2Ðtwo ignimbrites in the Westfall Butte Complex, WCÐ

Wildcat Creek Ash-Flow Tuff. (B) The TiO2 vs. Zr plot distinguishes the more primitive Tims Peak Basalt from the various basaltic andesites and andesites of the Keeney sequence. The plot also distinguishes ¯ows from the geographically distinct eruptive centers of both the Tims Peak Basalt and the Keeney basaltic andesite centers. Tims Peak Basalt: GFÐGrasshopper Flat type, HTÐHat Point type, JGÐJuniper Gulch type, PSRÐPence Spring Reservoir sill. Keeney sequence: CCÐCherry Creek, FMÐFreezeout Mountain, GSRÐGray Stud Reservoir, NRÐNegro Rock, SCÐSpring Creek, SSÐSkull Spring, VHÐVines Hill, WCÐWire Corral, WSÐWillow Springs. GSR, SC, SS, and WS are included by Kittleman et al. (1965) in the basalt of Cedar Mountain, whereas WC, CC, and Sh-5 are eruptive cenber assigned by them to the ``Shumway Ranch Basalt.''

؍ Figure 4. Plots to illustrate the chemical distinction between the 16.4 Ma to 15.3 Ma tholeiitic ¯ood basalts (Steens Mountain Basalt lled squares) and the extension-related calc-alkalic Tims Peak Basalt (®lled triangles) and® ؍ plus signs; basalts of Malheur Gorge basaltic andesites of the Keeney sequence (open triangles), both of which postdate the initiation of the Oregon-Idaho graben at 15.3 Ma.

(A) FeO*/MgO vs. SiO2 plot demonstrates the iron enrichment that accompanies increasing SiO2 in the tholeiites and the lack of such

enrichment in the calc-alkalic sequences. (B) P2O5 vs. TiO2 plot illustrates the higher P2O5 /TiO2 ratio of the calc-alkalic over the tholeiitic trends.

graben, a thick basin sequence of volcanic and The youngest volcanic units in the study DISCUSSION sedimentary strata (the Bully Creek Forma- area lie within the western Snake River Plain tion; Ferns et al., 1993a) overlies the Keeney (Fig. 1) and are assigned to the Kivett se- Volcanic-Tectonic Relationships sequence and includes the Devine Canyon quence. They form small volcanic vents sur- Ash-Flow Tuff (Walker, 1990), which has a rounded by local ¯ows of andesite and alkali Two physically and chemically distinct mean age of 9.5 Ma (Table 1). The Bully basalt lying along or parallel to the northwest- types of Tertiary magmatism are observed in Creek Formation is locally capped by ¯ows of trending faults of the Vale fault zone (Fig. 1). eastern Oregon; tholeiitic and calc-alkalic. porphyritic olivine basalt assigned to the Sour- The andesites are calc-alkalic with major ele- The large-volume tholeiitic ¯ood-basalt type dough sequence (Lees, 1994). The olivine ba- ment contents similar to those of ¯ows of the includes typical Columbia River basalts lying salts are mildly alkalic, with lower silica con- Keeney sequence. The basalts are alkalic with conformably on the Steens Basalt. Clari®ca-

tents and higher TiO2/Zr ratios than ¯ows of calcic clinopyroxenes and low silica concen- tion of the stratigraphic relationship between either the Keeney sequence below or the Kiv- trations. The vent that forms Malheur Butte the Steens Basalt and Columbia River basalts, ett sequence above. Equivalent ¯ows of oliv- yields a date of 0.8 Ma (Ϯ 0.7 Ma), and two backed by more precise 39Ar/40Ar dates, per- ine basalt farther south have been dated at 7.8 samples from Brogan Summit give ages of 1.7 mits the Steens Basalt to be regarded with Ma Ϯ 0.5 Ma (Table 1; Fieblekorn et al., Ma (Ϯ 0.1 Ma) and 1.9 Ma (Ϯ 0.3 Ma) (Table some con®dence as the earliest manifestation 1982; Hart, 1982). 1). of the ¯ood-basalt eruption. The coincidence

Geological Society of America Bulletin, January 2002 47

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020 HOOPER et al.

Figure 5. Estimated volcanic eruption rates in east-central Oregon over the past 23 m.y. Pale shading is the Steens Basalt and the basalt of Malheur Gorge tholeiites (estimated from Carlson and Hart, 1987, and this study). In black are the small extension-related calc-alkalic eruptions discussed in the text and Table 1. Dashed line indi- cates estimated eruption rates with time for the Columbia River Basalt Group on the Columbia Plateau (see Tolan et al., 1989).

Gorge fault (Johnson et al., 1998b). Younger volcanic activity (Keeney, Sourdough, and Kivett sequences) occurred as small-volume eruptions across the developing Oregon-Idaho graben. These units are calc-alkalic to mildly alkalic and accompanied the continued evolution of the graben as it ®lled with thick units of terrestrial sediment and pyroclastic hori- zons (Cummings et al., 2000).

Wider Tectonic Implications

The eruption of the Miocene ¯ood basalts of eastern Oregon, Washington, and Idaho can in space and time of the Steens Basalt eruption vada. They are consistently associated with be seen as a single brief but dramatic interlude with the calculated impingement of the Yel- east-west extension and widely attributed to (16.6±14.5 Ma) within the long-standing calc- lowstone hotspot beneath southeast Oregon decompressional melting of lithospheric alkalic to mildly alkalic volcanism associated (Engebretson et al., 1985; Pierce and Morgan, source components (Robyn, 1979; Thorkel- with the east-west extension that has domi- 1992) supports the model that connects the or- son, 1989; Fitton et al., 1991; Janecke, 1992; nated Eocene to the present tectonics of the igin of this ¯ood-basalt province with the ar- Hawkesworth et al., 1995; Hooper et al., inland Paci®c Northwest at the northern end rival of the Yellowstone hotspot. 1995; Morris and Hooper, 1997; Morris et al., of the Basin and Range province. In contrast to other classic ¯ood-basalt 2000). The association of these small-volume The ¯ood-basalt eruption began with the provinces (Ellam and Cox, 1991; Peng and calc-alkalic eruptions with east-west extension Steens Basalt, which covered an estimated Mahoney, 1995), these early ¯ows of the is particularly clear in east-central Oregon. In 65 000 km2 of southeastern Oregon (Carlson ¯ood-basalt province are normal phyric tho- the Malheur Gorge area on the northeast and Hart, 1987) at 16.6 Ma, a time when plate leiites that apparently lack alkalic or picritic shoulder of the Oregon-Idaho graben (Fig. 1), reconstruction (Engebretson et al., 1985) and members other than simple cumulates (Gunn the Hog Creek sequence includes the last the track of the Yellowstone hotspot (Pierce and Watkins, 1970; Carlson and Hart, 1987; ¯ows of the tholeiitic sequence (Hunter Creek and Morgan, 1992) both place the hotspot be- Johnson et al., 1998a; J.G. Evans, 1997, per- Basalt) that inter®nger with the silicic units neath southeast Oregon. Eruption of the sonal commun.). directly associated with the graben boundary Steens Basalt, therefore, correlates both in The ¯ood-basalt tholeiites were erupted faults. A short but signi®cant hiatus in vol- space and time with the arrival of the hotspot. during a short interval (16.6 Ma to 15.3 Ma) canic activity, from 14.5 Ma to 13.5 Ma, sep- The Steens Basalt was followed, without a de- between numerous, small-volume, calc-alkalic arates the Hog Creek sequence from the prim- tectable hiatus, by ¯ows typical of Imnaha and volcanic units that lie beneath the Steens Ba- itive calc-alkalic Tims Peak Basalt (Table 1). Grande Ronde basalt eruptions (Fig. 1B). We salt at Steens Mountain and above the tholei- During the volcanic hiatus, east-west exten- conclude this continuum implies a causal re- ites in the Malheur Gorge (Fig. 5). At Steens sion continued. The Oregon-Idaho graben lationship between the impingement of the Mountain, the pretholeiitic calc-alkalic units boundary faults continued to form, cutting the hotspot on the base of the lithosphere and the are dated, from the base upward, at 21.3 Ma, rhyolite of Cottonwood Mountain and the Lit- eruptions in the whole Columbia River ¯ood- 22.1 Ma, 19.5 Ma, and 17.8 Ma (Everden et tle®eld Rhyolite, and a combination of right- basalt province. al., 1964; Laursen and Hammond, 1974; Anita lateral and vertical movement displaced mem- The clear distinction between the two types L. Grunder, 1998, personal commun.; Johnson bers of the Hog Creek sequence along the of magmatism in east-central Oregon and their et al., 1998a). In the Malheur Gorge, the Hog west-northwest±trending Malheur Gorge fault, respective associations with a long-continuing Creek and younger (post±15.3 Ma) calc-alkal- which acted as a pull-apart structure to form extensional regime on the one hand and with ic to mildly alkalic volcanic units are directly a narrow graben along the eastern end of the the abrupt impingement of the hotspot on the associated with formation of the Oregon-Idaho gorge (Hooper et al., 2002). As a result of this other, are critical to interpreting the origin of graben. extensional deformation, the Tims Peak Basalt the ¯ood basalts. There is now little evidence Calc-alkalic Eocene to Holocene magmatic lies unconformably on units of the Hog Creek to support backarc spreading as the prime events occurred from British Columbia to Ne- sequence and is not displaced by the Malheur cause of the ¯ood-basalt eruptions (Carlson

48 Geological Society of America Bulletin, January 2002

and Hart, 1987); the pervading extension was and to the increased volume of magmatism Armstrong, R.L., Taubeneck, W.H., and Hales, P.O., 1977, Rb/Sr K/Ar geochronometry of the Mesozoic granite indeed present, but its association was with along the North Nevada rift. rocks and their Sr isotope composition, Oregon, Wash- the small-scale calc-alkalic volcanism, not Eruption of the ¯ood basalts from linear ington and Idaho: Geological Society of America Bul- with the ¯ood basalts. vent systems parallel and adjacent to the edge letin, v. 88, p. 397±411. Baksi, A.K., 1989, Reevaluation of the timing and duration Formation of the 50-km-wide Oregon-Idaho of older, thicker cratonic crust resembles the of extrusion of the Imnaha, Picture Gorge, and Grande graben, immediately following the early erup- geometry advocated by Anderson (1994) for Ronde Basalts, Columbia River basalt group, in Rei- del, S.P., and Hooper, P.R., eds.: Geological Society of tions of ¯ood basalt, presumably displaced the the origin of these massive eruptions by America Special Paper 239, p. 105±112. Steens Mountain volcanic center westward means of convection induced by lithospheric Binger, G.B., 1997, The volcanic stratigraphy of the Juntura from an original position closer to the Idaho boundaries of different thickness. But the co- region, eastern Oregon [Ph.D. thesis]: Pullman, Wash- ington State University, 176 p. 87 86 lithospheric suture zone (the Sr/ Sr line of incidence of the ®rst eruptions occurring pre- Brandon, A.D., and Goles, G.G., 1988, A Miocene sub- Armstrong et al., 1977) (Fig. 1A). Restoration cisely where and when the Yellowstone hot- continental plume in the Paci®c Northwest: Geochem- of the Steens Mountain eruption up to 50 km spot is ®rst recognized appears to con®rm the ical evidence: Earth and Planetary Science Letters, v. 88, p. 273±283. to the east strengthens the concept of the ¯ood association of these eruptions with the arrival Camp, V.E., 1995, Mid-Miocene propagation of the Yel- basalts erupting along an approximately north- of a hotspot. The origin of the hotspot remains lowstone mantle plume head beneath the Columbia unclear, but given elevated upper-mantle tem- River basalt source region: Geology, v. 23, south linear zone of weakness or ``thin zone'' p. 435±438. (Thompson and Gibson, 1991; Zoback et al., peratures and the consequent production of Campbell, I.H., and Grif®ths, R.W., 1990, Implications of 1994; Camp, 1995) extending from Steens exceptionally large volumes of basaltic melt mantle plume structure for the evolution of ¯ood ba- along such a cratonic boundary, that melt salts: Earth and Planetary Science Letters, v. 99, Mountain north to the Washington State bor- p. 79±93. der, parallel to and just west of, the old lith- would inevitably migrate to the shallower Carlson, R.W., 1984, Isotopic constraints on Columbia Riv- ospheric suture. In all probability this was a crust-mantle boundary beneath the accreted er ¯ood basalt genesis and the nature of the subcon- terranes to the west of the lithospheric suture. tinental mantle: Geochimica et Cosmochimica Acta, relatively narrow zone of weakness, perhaps v. 48, p. 2357±2372. an older graben developed by earlier extension Carlson, R.W., and Hart, W.K., 1987, Crustal genesis on in the relatively thin accreted lithosphere to CONCLUSIONS the Oregon Plateau: Journal of Geophysical Research, v. 92, p. 6191±6206. the west of the suture. This ``thin zone'' prob- Cummings, M.L., Evans, J.G., Ferns, M.L., and Lees, K.R., ably did not include either the North Nevada Stratigraphic, petrographic, and chemical 2000, Stratigraphic and structural evolution of the rift (NNR, Fig. 1; Zoback et al., 1994) or the correlations demonstrate that the Columbia middle Miocene syn-volcanic Oregon-Idaho graben: River ¯ood-basalt province began with the Geological Society of America Bulletin, v. 112, west-northwest±trending western Snake River p. 668±682. Plain (Hooper et al., 2002), both of which are eruption of the Steens Basalt in southeast Ellam, R.M., and Cox, K.G., 1991, An interpretation of Karoo picrites in terms of interaction between as- more obviously associated with calc-alkalic to Oregon at 16.6 Ma, at the time and place of the independently calculated impingement of thenospheric magmas and the mantle lithosphere: mildly alkalic magmatism and/or east-west Earth and Planetary Science Letters, v. 105, the Yellowstone hotspot. The focus of the extension. p. 330±342. ¯ood-basalt eruption then moved north to the Engebretson, D.C., Cox, A., and Gordon, R.G., 1985, Rel- Although not the direct cause of the ¯ood- southeast corner of the Columbia Plateau, ative motion between oceanic and continental plates basalt magmatism, extension may have been in the Paci®c basin: Geological Society of America where 90% of the eruption of the Columbia in¯uenced by the proximity to the Yellow- Special Paper 206, 59 p. River Basalt Group (the Imnaha and Grande Evans, J.G., 1990a, Geology and mineral resources map of stone hotspot. Softening of the lithosphere by Ronde Basalt Formations) took place between the Jonesboro quadrangle, Malheur County, Oregon: the hotspot and accompanying massive tho- Oregon Department of Geology and Mineral Indus- 16.1 and 15.0 Ma. In east-central Oregon, the leiitic extrusions, may account for the rapid tries, Geological Map Series 66, scale 1:24 000. voluminous ¯ood-basalt eruptions were super- Evans, J.G., 1990b, Geology and mineral resources map of east-west extension that formed the 50 km the South Mountain quadrangle, Malheur County, imposed in a short interval (16.6±15.3 Ma) on wide Oregon-Idaho immediately following the Oregon: Oregon Department of Geology and Mineral the intermittent, small, and compositionally Industries, Geological Map Series 67, scale 1:24 000. ®rst tholeiitic eruptions. It is also possible that variable calc-alkalic to mildly alkalic volca- Everden, J.F., Savage, D.E., Curtis, G.H., and James, G.T., this enhanced extension dragged part of the nism that has accompanied east-west exten- 1964, K-Ar dates and Cenozoic mammalian chronol- cratonic lithospheric keel westward to inhibit ogy of North America: America Journal of Science, sion here, and in the whole inland Paci®c v. 262, p. 145±198. further eruptions from immediately above the Northwest, since the Eocene. Ferns, M.L., Brooks, H.C., Evans, J.G., and Cummings, hotspot and to channel the tholeiitic magmas M.L., 1993a, Geologic map of the Vale 30 ϫ 60 mi- northward along the lithospheric ``thin zone'' nute quadrangle, Malheur County, Oregon and Owy- ACKNOWLEDGMENTS hee County, Idaho: Oregon Department of Geology toward the Washington border. This model and Mineral Industries, Geological Map Series 77, provides a plausible explanation for the sud- Jim Evans, Howard Brooks, Mark Ferns, and scale 1:100 000. den cessation of ¯ood-basalt activity in east- Mike Cummings introduced us to the area, gave us Ferns, M.L., Evans, J.G., and Cummings M.L., 1993b, constant encouragement, and provided published Geologic map of the Mahogany Mountain 30 ϫ 60 central Oregon at 13.5 Ma and is a further and unpublished maps. Chris Hawkesworth and Jen- minute quadrangle, Malheur County, Oregon and possible explanation (see Geist and Richards, da Johnson played vital roles in the ®eld and re- Owyhee County, Idaho: Oregon Department of Ge- viewed an earlier manuscript. Vic Camp and Rick ology and Mineral Industries, Geological Map Series 1993, and Camp, 1995) for the northward mi- 78, scale 1:100 000. Conrey suggested many useful improvements to the gration of the ¯ood-basalt magma to the Co- Fieblekorn, R.B., Walker, G.W., MacLeod, N.S., McKee, ®nal paper, and Bob Duncan provided six critical E.H., and Smith, J.G., 1982, Index to K-Ar determi- lumbia Plateau, where it continued to erupt for 40 39 Ar/ Ar dates. We are grateful to all. nations for the State of Oregon: Isochron/West, v. 37, another 7 m.y. (Tolan et al., 1989). p. 3±60. It is also plausible, if speculative, that the Fitton, J.G., James, D., and Leeman, W.P., 1991, Basic mag- REFERENCES CITED matism associated with Cenozoic extension in the lateral spreading of the plume head of the hot- western United States: Compositional variations in spot contributed not only to the northward mi- Anderson, D.L., 1994, The sublithospheric mantle as the space and time: Journal of Geophysical Research, gration of the eruption, but also both to the source of continental ¯ood basalts; the case against v. 96, p. 13693±13711. the continental lithosphere and plume head reservoirs: Fitzgerald, J.F., 1984, Geology and basalt stratigraphy of east-to-west migration of silicic activity along Earth and Planetary Science Letters, v. 123, the Weiser embayment, west-central Idaho, in the previously weakened Brothers fault zone p. 269±280. Bonnichsen, Bill, and Breckenridge, R.M., eds., Ce-

Geological Society of America Bulletin, January 2002 49

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020 HOOPER et al.

nozoic geology of Idaho: Idaho Bureau of Mines and perposed extensional systems, east central Idaho: Ev- Pierce, K.L., and Morgan, L.A., 1992, The track of the Yel- Geology Bulletin 26, p. 103±128. idence for an Eocene tectonic transition: Tectonics, lowstone hot spot: Volcanism, faulting, and uplift, in Fleck, R.J., and Criss, R.E., 1985, Strontium and oxygen v. 11, p. 1121±1138. Link, P.K., Kuntz, M.A., and Platt, L.B., eds., Regional isotope variations in Mesozoic and Tertiary plutons of Johnson, J.A., Hawkesworth, C.J., Hooper, P.R., and Binger, geology of eastern Idaho and western Wyoming: Geo- central Idaho: Contributions to Mineralogy and Pe- G.B., 1998a, Major and trace element analyses of logical Society of America Memoir 179, p. 1±53. trology, v. 90, p. 201±308. Steens Basalt, southeastern Oregon: U.S. Geological Richards, M.A., Duncan, R.A., and Courtillot, V.E., 1989, Geist, D., and Richards, M.A., 1993, Origin of the Colum- Survey Open-File Report 98±482, 26 p. Flood basalts and hot spot tracks: Plume heads and bia Plateau and the Snake River Plain: De¯ection of Johnson, J.A., Hooper, P.R., Hawkesworth, C.J., and Binger, tales: Science, v. 246, p. 103±107. the Yellowstone plume: Geology, v. 21, p. 789±792. G.B., 1998b, Geologic map of the Stember Ridge Robyn, T.L., 1979, Miocene volcanism in eastern Oregon: Gunn, B.M., and Watkins, N.D., 1970, Geochemistry of the quadrangle, Malheur County, southeastern Oregon: An example of calc-alkaline volcanism unrelated to Steens Mountains basalts, Oregon: Geological Society U.S. Geological Survey Open-File Report 98±105, subduction: Journal of Volcanological and Geothermal of America Bulletin, v. 81, p. 1496±1516. scale 1:24 000. Research, v. 5, p. 149±161. Haddock, G.H., 1967, The Dinner Creek Ash¯ow Welded Kittleman, L.R., Green, A.R., Haddock, G.H., Hagood, Swisher, C.C., Ach, J.A., and Hart, W.K., 1990, Laser fu- Tuff of the Malheur Gorge area, Malheur County, A.R., Johnson, A.M., McMurray, J.M., Russell, R.G., sion 40Ar/39Ar dating of the type Steens Mountain Ba- Oregon [Ph.D. thesis]: Eugene, University of Oregon, and Weeden, D.A., 1965, Cenozoic stratigraphy of the salt, south-eastern Oregon, and the age of the Steens 111 p. Owyhee region, south-eastern Oregon: University of geomagnetic polarity transition [abs.]: Eos (Transac- Hart, W.K., 1982, Chemical, chronological and isotopic sig- Oregon Museum of Natural History Bulletin 1, 45 p. tions, American Geophysical Union), v. 71, p. 1296. ni®cance of low potassium, high aluminum olivine Kittleman, L.R., Green, A.R., Haddock, G.H., Hagood, Takahashi, E., Nakajima, K., and Wright, T.L., 1998, Origin tholeiite in the north-western Great Basin [Ph.D. the- A.R., Johnson, A.M., McMurray, J.M., Russell, R.G., of the Columbia River basalts: Melting model of a sis]: Cleveland, Ohio, Case Western Reserve Univer- and Weeden, D.A., 1967, Geologic map of the Owy- heterogeneous plume head: Earth and Planetary Sci- sity, 410 p. hee region, Malheur County, Oregon: University of ence Letters, v. 162, p. 63±80. Hawkesworth, C.J., Turner, S., Galalgher, K., Hunter, A., Oregon Museum of Natural History Bulletin 8, scale Thompson, R.N., and Gibson, S.A., 1991, Subcontinental mantle plumes, hot spots and preexisting thin spots: and Bradshaw, T.K., 1995, Calc-alkaline magmatism, 1:125 000. Journal of Geophysical Research, v. 148, p. 973±977. lithospheric thinning and extension in the Basin and Laursen, J.M., and Hammond, P.E., 1974, Summary of ra- Thorkelson, D.J., 1989, Eocene sedimentation and volca- Range: Journal of Geophysical Research, v. 100, diometric ages of Oregon and Washington rocks, nism in the Fig Lake graben, southwestern British Co- p. 10271±10286. through 1972: Isochron/West, v. 9, 32 p. lumbia: Canadian Journal of Earth Science, v. 26, Lees, K.R., 1994, Magmatic and tectonic changes through Hooper, P.R., 1990, The timing of crustal extension and the p. 1368±1373. eruption of continental ¯ood basalts: Nature, v. 345, time in the Neogene volcanic rocks of the Vale area, Tolan, T.L., Reidel, S.P., Hooper, P.R., Beeson, M.H., Fecht, p. 246±249. Oregon, northwestern U.S.A. [Ph.D. thesis]: Milton K.R., Bentley, R.D., and Anderson, J.L., 1989, Revi- Hooper, P.R., 1997, The Columbia River ¯ood basalt prov- Keynes, UK, Open University, 283 p. sions to the estimates of the areal extent and volume 40 39 ince: Current status, in Mahoney, J.J., and Cof®n, Long, P.E., and Duncan, R.A., 1982, Ar/ Ar ages of Co- of the Columbia River basalt group, in Reidel, S.P., M.F., eds., Large igneous provinces: Continental, oce- lumbia River Basalt from deep boreholes in south- and Hooper, P.R., eds., eds., Volcanism and tectonism anic and planetary ¯ood volcanism: American Geo- central Washington: Richland, Washington, Rockwell in the Columbia River ¯ood-basalt province: Geolog- physical Union Geophysical Monograph 100, p. 1±28. Hanford Operations, Report RHO-BW-SA-233 P. ical Society of America Special Paper 239, p. 1±20. Hooper, P.R., 2000, Chemical discrimination of Columbia Martin, B.S., 1984, Paleomagnetism of basalts in north-east Walker, G.W., 1990, Miocene and younger rocks of the River basalt ¯ows: Geochemical and Geophysical Oregon and west central Idaho, [M.S. thesis]: Pull- Blue Mountains region, exclusive of the Columbia Geosystems, v. 1. man, Washington State University, 133 p. River Basalt Group and associated ma®c lava Hooper, P.R., Kleck, W.D., Knowles, C.R., Reidel, S.P., and McKee, E.H., Hooper, P.R., and Kleck, W.D., 1981, Age of ¯ows: U.S. Geological Survey Professional Paper Thiessen, R.L., 1984, Imnaha Basalt, Columbia River Imnaha BasaltÐOldest basalt ¯ows of the Columbia 1437, p. 101±118. Basalt Group: Journal of Petrology, v. 25, p. 473±500. River Basalt Group, northwest United States: Iso- White, R.S., and McKenzie, D.P., 1989, Magmatism at rift Hooper, P.R., Bailey, D.G., and McCarley Holder, G.A., chron/West, no. 31, p. 31±33. zones: The generation of volcanic continental margins 1995, Tertiary calc-alkaline magmatism associated Morris, G.A., and Hooper, P.R., 1997, Petrogenesis of the and ¯ood basalts: Journal of Geophysical Research, with lithospheric extension in the Paci®c Northwest: Colville batholith, N.E., Washington: Implications for v. 94, p. 7685±7729. Journal of Geophysical Research, v. 100, Eocene tectonics in the northern US Cordillera: Ge- Zoback, M.L., McKee, E.H., Blakeley, R.J., and Thomp- p. 10303±10319. ology, v. 25, p. 831±834. son, G.A., 1994, The northern Nevada rift: Regional Hooper, P.R., Johnson, J.A., and Hawkesworth, C.J., 2002, Morris, G.A., Larson, P.B., and Hooper, P.R., 2000, ``Sub- tectono-magnetic relations and middle Miocene stress A model for the origin of the western Snake River duction style'' magmatism in a nonsubducting setting: direction: Geological Society of America Bulletin, Plain as an extensional strike-slip duplex, Idaho and The Colville Igneous Complex, northeast Washington v. 106, p. 371±382. Oregon, in Bonnichsen, B., White, C.M., and Mc- State, U.S.A.: Journal of Petrology, v. 41, p. 43±67. Curry, M., eds., Tectonic and Magmatic Evolution of Peng, Z.X., and Mahoney, J.J., 1995, Drillhole lavas from MANUSCRIPT RECEIVED BY THE SOCIETY FEBRUARY 16, 2001 REVISED MANUSCRIPT RECEIVED JUNE 25, 2001 the Snake River Plain Volcanic Province: Idaho Geo- the northwestern Deccan Traps, and the evolution of MANUSCRIPT ACCEPTED AUGUST 29, 2001 logical Survey Bulletin 30. the Reunion mantle plume: Earth and Planetary Sci- Janecke, S.U., 1992, Kinematics and timing of three su- ence Letters, v. 134, p. 169±185. Printed in the USA

50 Geological Society of America Bulletin, January 2002

Downloaded from https://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/114/1/43/3387249/i0016-7606-114-1-43.pdf by University of Nevada Reno user on 02 March 2020