Saddle Mountains Basalt Research Articles

Igneous Rock Associations 27. Chalcophile and Platinum Group Elements in the Columbia River Basalt Group: A Model for Flood Basalt Lavas

The Columbia River Basalt Group is the youngest and best preserved continental Large Igneous Province on Earth. The 210,000 km3 of basaltic lavas were erupted between 16.6 and 5 Ma in the Pacific Northwest, USA. The peak of the eruptions occurred over a 700,000-year period when nearly 99% of the basalts consisting of the Steens, Imnaha, Picture Gorge, Grande Ronde and Wanapum Basalts were emplaced. In this study we examined the Platinum Group Elements (PGEs) Pt and Pd, and the chalcophile elements Cu and Zn in the Columbia River Basalt Group. The presence of Pt, Pd and Cu in the compositionally primitive Lower Steens, Imnaha and Picture Gorge Basalts suggests that the Columbia River Basalt Group magma was a fertile source for these elements. The PGEs are contained mainly in sulphides in the earliest formations based on their correlation with immiscible sulphides, sulphide minerals and chalcophile elements. Grande Ronde, Wanapum and Saddle Mountains Basalts are depleted in PGEs and chalcophile elements compared to earlier formations. Sulphur was saturated in many flows and much of it probably came from assimilation of cratonic rock from a thinned lithosphere. We propose a model where the presence or absence of PGEs and chalcophile elements results primarily from the interaction between an advancing plume head and the crust/lithosphere that it encountered. The early lavas erupted from a plume that had little interaction with the crust/lithosphere and were fertile. However, as the plume head advanced northward, it assimilated crustal/lithospheric material and PGE and chalcophile elements were depleted from the magma. What little PGE and chalcophile elements remained in the compositionally evolved and depleted Grande Ronde Basalt flows mainly were controlled by substitution in basalt minerals and not available for inclusion in sulphides.

Impacts of agricultural irrigation recharge on groundwater quality in a basalt aquifer system (Washington, USA): a multi-tracer approach

Irrigation in semi-arid agricultural regions can have profound effects on recharge rates and the quality of shallow groundwater. This study coupled stable isotopes (2Η, 18O), age-tracers (3H, CFCs, 14C), 87Sr/86Sr ratios, and elemental chemistry to determine the sources, residence times, and flowpaths of groundwater and agricultural contaminants (e.g. NO3–) in the Saddle Mountains Basalt Aquifer in central Washington, USA, where over 80% of the population depend on groundwater for domestic use. Results demonstrate the presence of two distinct types of water: contaminated irrigation water and pristine regional groundwater. Contaminated irrigation water has high NO3– concentrations (11–116 mg/l), 87Sr/86Sr ratios (0.70659–0.71078) within range of nitrogen-based fertilizers, detectable tritium (2.8–13.4 TU), CFC ages 20–40 years, high δ18O values (−16.9 to −13.5‰), and ∼100 percent modern 14C. Pristine regional groundwater has low NO3– concentrations (1–5 mg/l), no detectable tritium (≤0.8 TU), low δ18O values (−18.9 to −17.3‰) and 14C ages from ∼15 to 33 ky BP. Nitrogen and oxygen isotopes of NO3–, combined with high dissolved oxygen values, show that denitrification is not an important process in the organic-poor basalt aquifers resulting in transport of high NO3– irrigation water to depths greater than 40 m in less than 30 years.

Assessing subcontinental lithospheric mantle sources for basalts: Neogene volcanism in the Pacific Northwest, USA as a test case

Many objections have been raised as to the ability of subcontinental lithospheric mantle to produce voluminous amounts of basalt, because this upper part of the mantle is thought to be refractory, and the geotherm is rarely above the peridotite solidus at these depths under continents. However, in the Pacific Northwest of the USA during the Neogene, the subcontinental lithospehric mantle has been proposed as a key source for basalts erupted within the northern Basin and Range, and for the Columbia River flood basalts erupted on the Columbia Plateau. An alternative explanation to melting in the subcontinental lithospheric mantle, which equally well explains the chemical compositions thought to originate there, is that these magmas were contaminated by crust of varying ages. Calc-alkaline lavas, which occupy the Blue Mountains in the center of this region, hold clues to the latter process. Their elevated trace element ratios (e.g., Ba/Zr, K2O/P2O5), coupled with differentiation indicators such as Mg? [molar Mg/(Mg + Fe)], and Sr, Nd, and Pb isotopic compositions, can most reasonably be explained by crustal contamination. Appraisal of continental peridotite xenolith data indicates that high trace element ratios such as Ba/Zr in continental basalts cannot result from melting in the subcontinental lithospheric mantle. Instead, as with the calc-alkaline lavas, these high ratios in the tholeiites most likely indicate crustal contamination. Furthermore, the peridotite xenoliths do not have a relative depletion in Nb and Ta that is observed in most of the lavas within the region. Relatively minor volumes of tholeiites erupted in late Neogene times in the northern Basin and Range (Hi-Mg olivine tholeiites) and Columbia Plateau (Saddle Mountains basalts), are the only lavas which have trace element and isotopic compositions consistent with being derived from, or largely interacting with a subcontinental lithospheric mantle in the Pacific Northwest. In contrast to the prior studies, we suggest that the mantle sources for most of the basalts in this region were ultimately beneath the lithospheric mantle.

Ferro-andesites in the Grande Ronde Basalt: their composition and significance in studies of the origin of the Columbia River Basalt Group

Flows from the Grande Ronde Basalt of the Columbia River Basalt Group include several with the chemical composition of ferro-andesites. These flows have SiO2 > 55%, MgO < 4%, and also have higher Fe/Mg than the average value for Grande Ronde Basalt. They are also distinctly richer in Cs, Rb, K, Ba, La, Ce, Th, and U than the remainder of the Grande Ronde Basalt flows, and possess small negative Eu anomalies. Their Pb isotopic compositions define a mixing line with a negative slope on a 206Pb/204Pb versus 207Pb/204Pb plot. Their Nd isotopic compositions lie between 143Nd/144Nd = 0.51252 and 0.51264, and their Sr isotopic compositions lie between 87Sr/86Sr = 0.7055 and 0.7060. These values define the enriched end of the Columbia River Basalt Group spectrum on a Sr–Nd epsilon diagram (excluding the Saddle Mountains Basalt). These ferro-andesite flows also form a compositional end member of the Columbia River Basalt Group and are sufficiently distinctive to warrant special consideration. We compare them with lavas from other tholeiitic provinces. Petrogenetically, they can be related to the Grande Ronde Basalt low-Mg basalts by plagioclase fractionation, or by clinopyroxene fractionation in partially melted eclogites. However, the situation may not be that simple, as their unique isotopic compositions are closely related to those of parental mantle materials, perhaps thus requiring separate reservoirs and (or) sources. Extending this argument to other parts of the Columbia River Basalt Group suggests that the origin of these basalts may be in a system of comparatively small magma chambers in the uppermost mantle, frequently replenished from a variety of sources.

Lead isotopes and the sources of the Columbia River Basalt Group

A detailed study of 208Pb/204Pb, 207Pb/204Pb, and 206Pb/204Pb ratios suggests that the number of Pb isotopic reservoirs required for the Columbia River Basalt Group (CRBG) (as currently defined) must be increased from the presently accepted four to at least six. The identities of the six reservoirs are two of mid‐ocean ridge basalt (MORB) slightly contaminated with sediment (Rl and R2, Rl being “Cascades” Pb); one a probable plume component (R3); two of unspecified mantle material but probably of Rl type, contaminated with local crustal Pb (R4 and R5); and the sixth a complex enriched reservoir of 2150 Ma age (R6). Rl is an end‐member for all CRBG, except the Saddle Mountains Basalts. R2 is a second end‐member for the Picture Gorge Basalt and appears to be unique to it. Both Rl and R2 are related to the Juan de Fuca subduction zone and the back arc spreading associated with it. R3 is the principal source for most of the Wanapum Basalt, as well as for most of the chemically evolved portions of the Grande Ronde Basalt. Volumetric and tectonic considerations require that one source of the CRBG be a plume or, at least, a nonlocal crustal or uppermost mantle source, and R3 is the ideal candidate for that role. R4 is the second source for the Imnaha Basalt, and R5 is the source for the 206Pb rich varieties of the Grande Ronde Basalt. R4 and R5 are identified with contamination by local Phanerozoic crust because of their similarity to local crustal Pb and because the rocks which contain them also have Cu contents correlated with their Pb isotopic compositions. R6 is the parent for Pb in all the Saddle Mountain Basalts. This last source appears to have been homogeneous 2150 m.y. ago but has since split into a number of separate, discrete pockets, each with its own characteristic Pb isotopic signature. Some of these appear in individual Saddle Mountain Basalt flows, while other sources have mixed with R4 to produce individual Saddle Mountain Basalt flows of the Ice Harbor set. R6 is similar in all its isotopic properties to the Precambrian basement that is exposed to the immediate north of the Columbia River Basalt Group. The overall model for Columbia River Basalt Group genesis is termed “plume reinforced back arc magmatism”.

The Huntzinger flow: Evidence of surface mixing of the Columbia River Basalt and its petrogenetic implications

Variation in chemical composition within flows of the Columbia River Basalt Group (CRBG) is typically small, but one flow, the Huntzinger flow, exhibits a large intraflow compositional range. The composition varies both vertically and laterally, but mass-balance calculations for both major/minor oxides and trace elements show that the variation can be explained by simple binary mixing between two end members. These end-member compositions are identical to two Saddle Mountains Basalt flows, the basalt of Lapwai (Wilbur Creek Member) and the Asotin Member. Primary intraflow features of the Huntzinger flow are highly variable, which is atypical for most CRBG flows. In addition to common features such as a scoriaceous flow top, an entablature, and a colonnade, the flow exhibits evidence of invasion into sediments, pillow-hyaloclastic zones, ophimottling, and veins. Ophimottling and veins can be directly related to compositional variations; veins occur in zones of mixing between the two end members, and ophimottling is confined to zones dominated by the Asotin composition. The Huntzinger flow is dominated by clinopyroxene and plagioclase, with lesser amounts of olivine and variable amounts of glass. The mineralogies of the two end-member compositions are similar to those of the Asotin and Lapwai flows from the Clear-water embayment. The intermediate compositions of the Huntzinger flow contain characteristics of both flows, suggesting mixing of the two. The compositional variation indicates that mixing increases from the eastern part of the Pasco Basin, where the Asotin composition occurs in the center of the flow and is surrounded by Lapwai, to the western side, where there is almost complete mixing of the Lapwai and Asotin. This pattern is analogous to that produced when two flows mix by free turbulent flow from a free jet. Based upon this analogy, a flow emplacement model is proposed in which the basalt of Lapwai is initially emplaced in the Pasco Basin and, prior to solidification, is invaded by and mixed with the Asotin lava. This model places several petrogenetic constraints on this interval of the Saddle Mountains Basalt. The close spatial and temporal association of the Asotin Member and basalt of Lapwai with yet a third flow, the Wahluke flow (Wilbur Creek Member), allows the rate of eruption to be estimated and constrains the rate of magma recharge to a storage chamber. The role of crustal contamination can also be evaluated from this model. The conditions required for surface mixing suggest that this may not have been a unique occurrence and that other CRBG flows may be composite flows resulting from mixing.

Genetic interpretation of lead-isotopic data from the Columbia River Basalt Group, Oregon, Washington, and Idaho

Lead-isotopic signatures of basalts from the Columbia River Basalt Group of central Oregon, Washington, and western Idaho, and of the high-alumina olivine basalts from southeastern Oregon plot mostly in the field of Pb-isotopic data from volcanic rocks of the Cascade Range. In contrast to the mid-ocean ridge basalts (MORB), the Pb-isotopic data for several of the formations form small, tight clusters, and the Nd- and Sr-isotopic data show discrete variation between these basalt groups. Pseudoisochrons are not observed. A single model will account for the observed isotopic and trace-element data from samples from most of the Columbia River Basalt Group. This model calls for partial melting of the convecting oceanic-type mantle and mixing with, or contamination by, fluids derived from continental-derived sediments. It requires that crustal components are not only subducted along the trench, but that they are also transported in the low-velocity zone of the mantle at least 400 km behind the active volcanic arc into the back-arc environment represented by the Columbia Plateaus province. Lead isotopes in the Saddle Mountains Basalt, the youngest formation, suggest that these magmas were formed by partial melting of a 2.6-b.y.-old subcontinental lithosphere that extends as far west as south-central Washington near Ice Harbor Dam. The source material for these flows is characterized by high Th/U, Th/Pb, Rb/Sr, and Nd/Sm ratios and light REE enrichment. An additional component, presumably granitic crust, is necessary to explain the Pb-isotopic data of the Umatilla and perhaps the Pomona Members of the Saddle Mountains Basalt. This contamination model is supported by published δ18O data and petrological interpretation. Published studies of basalts from the Deccan Traps, the Tasmanian dolerites, the Snake River Group, and the volcanics of the Yellowstone Plateau province also suggest that melting of old subcontinental lithosphere beneath the continental crust may be an important process in basalt genesis on a global scale.

Intracanyon flows of the Columbia River Basalt Group in the lower Columbia River Gorge and their relationship to the Troutdale Formation

The Priest Rapids Member (Wanapum Basalt) and the Pomona Member (Saddle Mountains Basalt) of the Columbia River Basalt Group crossed the Miocene Cascade Range into western Oregon and Washington as intracanyon flows about 14 m.y. and 12 m.y. ago, respectively. A thick, allogenic, bedded palagonite complex underlying the Priest Rapids intracanyon flow originated when the first Priest Rapids flow (Rosalia chemical type) interacted with a shallow lake on the Columbia Plateau, displacing water that flushed hyaloclastic debris into an incipient ancestral Columbia River channel. The Priest Rapids flow then overfilled the canyon, forcing the river northward, where it established the Bridal Veil channel. The Bridal Veil channel, part of a river system that extended across the Columbia Plateau, was only partly filled by the Pomona flow, allowing the river to remain in this channel through the Cascade Range. The Troutdale Formation is made up of (1) older alluvial gravels deposited below and above the Pomona intracanyon flow while the ancestral Columbia River occupied the Bridal Veil channel and (2) younger, more varied alluvial deposits characterized by clastic and hyaloclastic debris from Boring and High Cascade volcanism, spread over a much wider area, when aggradation overfilled the Bridal Veil channel less than 6 m.y. ago. The shift of the lower Columbia River to its present course took place during this time of maximum alluviation, aided by scattered eruptions of younger volcanoes of the Boring and High Cascade Lavas. The present Columbia River Gorge was formed in post-Troutdale time by entrenchment of the Columbia River during rapid uplift of the Cascade Range of northern Oregon during the past 1 to 2 m.y.

Implications of oxygen-isotope data and trace-element modeling for a large-scale mixing model for the Columbia River Basalt

Current debate regarding the origin of the Columbia River Basalt Group centers on whether the considerable geochemical diversity in these basalts is the result of derivation from a heterogeneous mantle or of extensive interaction of the magmas with crustal rocks. Compositionally diverse rocks such as the Pomona and Elephant Mountain Members of the Saddle Mountains Formation have identical ( 87 Sr/ 86 Sr ratios (0.7078–0.7080). These radiogenic rocks have nearly identical δ 18 O values (6.24–6.36) to those of Picture Gorge (6.24), the least radiogenic of the Columbia River Basalts ( 87 Sr/ 86 Sr = 0.7038). This suggests that the mantle beneath the Columbia Plateau is heterogeneous at least with respect to Sr and probably other components. Basalts with low δ 18 O values (6.19–6.36) can be found in all formations of the Columbia River Basalt, an observation not compatible with models requiring extensive assimilation of upper-crustal rocks to explain isotopic differences between formations. Evidence for crustal contamination for some flow rocks in the Grande Ronde, Wanapum, and Saddle Mountains Basalts does exist. Contamination, however, seems to have affected isolated members of formations rather than formations as a whole. This conclusion is consistent with the results of trace-element modeling, which suggests that the Grande Ronde and Saddle Mountains Basalts were produced as numerous independent magmas from a heterogeneous mantle. The magmas ascended following independent paths, some undergoing crustal contamination, others not.

Chemical and mineralogic variations within four dikes of the Columbia River Basalt Group, southeastern Columbia Plateau

Four dikes within the western portion of the Chief Joseph dike swarm of the Mio-Pliocene Columbia River Basalt Group show systematic major-element, mineralogic, and textural trends across their widths. Two of the dikes are of Wanapum Basalt (Dodge chemical type) and served as source vents for two flows of Dodge chemistry and petrography that are exposed in the study area. A third dike of Saddle Mountains Basalt served as a vent for the Wenaha flow (Elephant Mountain chemical type), also present in the area. Two additional dikes of probable Saddle Mountains Basalt cannot be correlated unambiguously with any flows exposed in the area, but one of them closely resembles the Buford flow chemically and petrographically. The chemical variations across all four dikes indicate that the chilled dike margins are slightly enriched in felsic components and incompatible elements (SiO 2 , P 2 O 5 , TiO 2 , and K 2 O) and depleted in MgO and FeO compared to dike interiors. Most of these chemical trends can be related to the marked enrichment in glass at chilled dike margins compared to their more crystalline interiors. Microprobe analyses of glasses from chilled dike margins, when compared to whole-rock analyses from dike interiors, agree with the above-described trends for SiO 2 , TiO 2 , MgO, and probably K 2 O. FeO is also concentrated in the three glasses analyzed, producing an opposing and stronger trend (increase) than shown by the whole-rock data. This probably reflects the greater concentration of opaque oxides within the marginal tachylite compared to the more crystalline dike interiors, which lack tachylite. This variation in glass content is, of course, controlled by the rapid quenching of magma against the relatively cold country rock and the progressively decreasing cooling rate toward the interior of a dike as intrusion continues. In addition to variations in crystallinity, fractionation and/or crustal contamination may also have contributed to or modified these major-element trends. Mineralogic trends inward from dike margins are from mafic-to-felsic, with plagioclase anorthite content, and perhaps the Ca content of augite, decreasing. Orthopyroxene (hypersthene-bronzite) is present only in the chilled margin of the Wenaha dike and exhibits well-developed reaction rims of augite. These mafic-to-felsic mineralogic trends were superposed on the felsic-to-mafic chemical trends.

Post 12 m.y. rotation of southwest Washington

Paleomagnetic field directions from the basalt of Pack Sack Lookout are compared to those from the Pomona Member of the Saddle Mountains Basalt of the Columbia River Basalt Group. The Pomona crops out over a wide region on the Columbia Plateau east of the Cascade Range, and the basalt of Pack Sack Lookout crops out well to the west of the Cascades about 30 to 60 km east of the Washington coast. Our paleomagnetic results support existing geologic and geochemical evidence that both these units are part of a single 12‐m.y.‐old flow that erupted in western Idaho and flowed to the Washington coast. The magnetic data further indicate that southwest Washington has undergone clockwise rotation with respect to the Columbia Plateau and stable North America. The data suggest that both a large‐scale regional rotation of ∼16° and locally complex small‐scale rotations exist, the two being present to different degrees in the eastern and western regions of southwest Washington. The Cascade Range appears to coincide with the tectonic boundary, separating rotated and unrotated regions of Washington state. Rotation of southwest Washington appears to have been associated with the rotation of large pieces of western Oregon and southern Washington, possibly as a result of either a ridge‐push force from the Basin and Range province or shear along the Pacific‐North America plate boundary.