Columbia River Flood Basalts

Elevation (m): 
State (Province, etc): 
Idaho, Washington, Oregon
Flood Basalt

General Overview


The Grand Ronde Basalt of the Columbia River Basalt Group.

Thick stacks of laterally extensive lava flows typify this flood basalt province. Photo by Thor Thordarson.



Area covered by Columbia River flood basalts shown in gray. Dashed lines are dike swarms. The outer limits of the Chief Joseph dike swarm are marked by CJ (vents for the flows in the Imhaha, Grande Ronde, and Wanapum Formations and Saddle Mountains Basalt). The Grande Ronde (GR) and Cornucopia (C) dike swarms are within the Chief Joseph dike swarm. The Monument Dike Swarm (M) was the vent for the Picture Gorge Basalt. The Paso Basin is near the confluence of the Columbia and Snake Rivers. Map based on Hooper (1997).

Almost everything about this volcanic province is impressive. The Columbia River Flood Basalt Province forms a plateau of 164,000 square kilometers between the Cascade Range and the Rocky Mountains. In all, more than 300 individual large (average volume 580 cubic km!) lava flows cover parts of the states of Idaho, Washington, and Oregon. At some locations, the lava is more than 3,500 m thick. The total volume of the volcanic province is 175,000 cubic km. Eruptions filled the Pasco Basin in the east and then sent flows westward into the Columbia River Gorge. About 85% of the province is made of the Grande Ronde Basalt with a volume of 149,000 cubic km (enough lava to bury all of the continental United States under 12 m of lava!) that erupted over a period of less than one million years. Flows eventually reached the Pacific Ocean, about 300 to 600 km from their fissure vents. The Pomona flow traveled from west-central Idaho to the Pacific (600 km), making it the longest known lava flow on Earth (the major- and trace-element compositions of the flow do not change over its entire length).


Feeder dikes form the vents for the flood basalts and they trend to the north-northwest to south-southeast across eastern Oregon and western Idaho (Swanson and others, 1975). Hundreds of vents have been recognized and mapped. Small vents, such as spatter cones, are associated with the feeder dikes. The vents systems are 50 to more than 200 km long and a few kilometers wide. Some vents are hidden under younger flows. Photo of dike in the Chief Joseph dike swarm cutting across Grande Ronde Basalt. Photograph courtesy of Stephen Reidel.

Most of the flows in the Columbia River Flood Basalt Province are tholeiitic basalt. Representative samples are given below. Data from Wright and others (in press) presented in Swanson and others (1989).


        1       2       3
SiO2    53.84   50.94   52.00
Al2O3   14.37   14.27   15.04
FeO*    11.37   13.50   10.45
MgO     5.25    4.57    7.19
CaO     8.97    8.56    10.39
Na2O    2.92    2.85    2.23
K2O     1.10    1.25    0.65
TiO2    1.75    3.12    1.62
P2O5    0.23    0.68    0.24
MnO     0.19    0.25    0.18
FeO* = total FeO.
1. High MgO Grande Ronde basalt.
2. Roza Member of the Wanapum Basalt.
3. Pomona Member of Saddle Mountains Basalt.

Volcanism began about 17.5 million years ago and ceased about 6 million years ago.

Most of the volume of the Columbia River Flood Basalt Province (85%) was erupted in only 1.5 million years from 17 to 15.5 million years ago. Volume of each formation, in cubic kilometers, is given in parentheses. Black dots separate formations. Data from Tolan and others (in press) presented in Swanson and others (1989).


Comparison of the Roza Member (~ 14.5 million years ago, volume=1300 km3, emplacement=5-15 years, eruption rate=2600-8100 m2/s) of the Columbia River Flood Basalt Province to lava flows from 1. Kupaianaha (1986-1992, ~0.5km3, 5.6 years, 2-5m2/s), 2. Mauna Loa (1859, 0.27m3, 10 months, 4 m3/s), and 3. Laki (1783-1784, 14.7 km3, 8 months, 1150-4250 m3/s). From Self and others (1997).

The tectonic origin of the flood basalts is not simple. Hooper (1997) identified three major factors:

1. the Yellowstone hot spot;
2. thinning of the continental lithosphere as a result of spreading behind the Cascade arc; and
3. the proximity of the fissure vents to the tectonic boundary between accreted terranes made of thinner, denser oceanic lithosphere and the more competent lithospheres of the old North American Plate.

Many flood basalt provinces are associated with known hot spots and the Yellowstone hot spot may have influenced magma generation for the Columbia River flood basalt but the vents were 300-400 km north of the hot spot track and the chemistry of the basalts suggest a source in the lithospheric mantle not the asthenosphere as expected for hot spot magmas.

The area and volume of the Columbia River Flood Basalt Province are impressive but the volume is one-tenth the volume of other large igneous provinces such as Deccan, Parana, Karoo, and the Siberian Traps.




Columbia River Basalts: Climate


Sulfur concentrations in parts per million (ppm) at several locations in a glassy sample of the Columbia River flood basalt.

Because of the great amounts of gas released by such large volume eruptions, flood basalts have the potential to impact the global climate. Self and others (1997) estimated plume heights of 3-6 km above fire fountains and 8-11 km above fissures. Plumes during some periods of the eruption were higher. Self and others suggested that sulfur dioxide and hydrogen sulfide released during these eruptions formed sulfate aerosols in the upper troposphere and lower stratosphere.

Degassing budget of the Roza eruption.

The amount of sulfur (S) in the magma and released during various stages of volcanic activity are given. From Self and others (1997).

To better constrain the effects of such an eruption Thordarson and Self (1996) determined the sulfur, chlorine, and fluorine concentrations in glassy samples and glass inclusions from samples from the Rosa flow. They found that most of the gas (66%) was released during the eruption (about 9,000 Mt of SO2) and that significant amounts were released as the lava flowed and crystallized (about 2,800 Mt of SO2). Based on their study, the following amounts of gas were released:

9,000 Mt of SO2,
1300 Mt of HF, and
400 Mt of HCl.

These amounts are enormous relative to well studied historic eruptions. For example, the 1991 eruption of Mount Pinatubo produced 20 Mt of SO2, large by historic standards but small compared to flood basalt eruptions. Assuming the Rosa flow was erupted over a 10 year period, Thordarson and Self (1996) estimated that this flood basalt eruption was equivalent to four times the atmospheric impact of the Pinatubo eruption for every month for a decade. Other lava flows in the Columbia River Flood Basalt Province have even greater volumes (and other large igneous provinces have volumes 10 times greater than the Columbia River Flood Basalt Province). The impact of these eruptions are difficult to model but thought to be severe, possibly causing mass extinctions. For example, the amount of sunlight reaching the surface would be reduced to the equivalent of the full moonlight. However, 5,000-10,000 years may have passed between eruptions, allowing the environment to recover from such large eruptions. Self and others (1997) concluded "that a continental flood basalt eruption probably could not cause mass extinctions, but a series of them during the growth of a CFB [continental flood basalt] province would have been able to stress the environment to such an extent that any major perturbation would have had a more extreme effect."



Columbia River Basalts: Features of a Typical Flow

Features of a Typical Flow

Typical joint features in the Roza Member of the Columbia River Flood Basalt based on the exposure at Banks Lake, Washington. From Self and others (1997).

Perhaps the most characteristic feature of the Columbia River Flood Basalt Province is the similarity of individual lava flows. Most flows consist of colonnade (base), entablature (middle), and a vesicular and scoracious top. Colonnade is caused by slow cooling of ponded lava. Entablature is probably the result of cooling caused by fresh lava being covered by water. The flood basalts probably damned rivers. When the rivers returned, the water seeped down the cracks in the cooling lava and caused rapid cooling from the surface downward (Long and Wood, 1986). The division of colonnade and entablature is the result of slow cooling from the base upward and rapid cooling from the top downward. Columnar jointing is also found in other igneous rocks.


Columbia River Basalts: Lava Deltas

Lava Deltas

Pillow lava forms when eruptions are underwater or when lava flows enter a body of water. The abundance of pillows and deltas made of pillows and hyaloclastite indicates that rivers and lakes were common features during the formation of the Columbia River Flood Basalt Province. Pillows tend can be up to 5 m long and about 0.5 to 1 m thick. They are elongate in the direction of flow and dip 20-30 degrees. Dips decrease higher up in the sequence. Photo by Thor Thordarson.

A delta in the Grande Ronde basalt made of pillow lava and hyaloclastite. Photo by Thor Thordarson.

Hyaloclastite is angular pieces of volcanic glass formed as lava fragments and shatters as it enters the water. The glass is altered or weathered to palagonite. Photo by Thor Thordarson.

Numerous lava deltas formed during the current eruption of Kilauea in Hawaii are very similar to those in the Columbia River Flood Basalt Province.


Columbia River Basalts: Photo Tour

Photo Tour of the Columbia River Flood Basalt Province

Aerial view of dikes in Joseph Canyon near the Washington-Oregon border. The dike cuts across 3000 feet of Grand Ronde Basalt. All photographs by and courtesy of Stephen Reidel.

View along the Imnahu River Canyon. Imnahu Basalt is overlain by Grande Ronde Basalt.

Imnahu River Canyon. Laterally extensive thick sheets of Imnahu Basalt overlain by Grande Ronde Basalt.

The Grande Ronde River cutting through the Grande Ronde Basalt. Note plateau at top right.

The Snake River in Hells Canyon. Flood lavas of the Columbia River Basalt Group overlie the rocks (light colored rocks in low areas) of the Wallowa accreted terrain.

The Roza flow with its pillow-plagonite complex near The Dalles, Oregon.

Tree cast is a flood basalt lava flow.

The Martindale flow is cut by a dike that feeds the overlying Goose Island flow. These flows are part of the Ice Harbor Member. Note the columnar jointing in the Martindale flow.

Dry Falls, a waterfall carved by glacial floods. The rim of the falls is four miles wide.

The Interpretive Center at Dry Falls describes the geologic history of the area.

Palouse Falls formed during the catastrophic glacial floods that sculpted the Channel Scablands during the last Ice Age. Palouse Falls is 198 feet tall and is most spectacular in the spring and early summer. Palouse Falls is six miles above the Snake River.

Landsat image of the Channeled Scablands of eastern Washington.

VolcanoWorld wishes to thank Stephen Reidel for generously sharing his photographs of the Columbia River Flood Basalts.

Columbia River Basalts: Rate of Emplacement

Rate of Emplacement

Any model for the emplacement of the Columbia River Flood Basalts must explain how lava that travels 600 km can still be hot enough to be chilled to glass as it entered the Pacific Ocean.

Shaw and Swanson (1970) proposed that high eruption rates were required. Great volumes of turbulent lava would reach the ocean in only a few days, thus staying hot. The fronts of these flows might be 50 m high and 100 km long. They would move at rates of 3 to 5 km per hour, down the gentle slope to the west. The eruption rate would need to be high, 1 cubic km/day/linear kilometer of fissure, about 1000-10,000 the eruption rates of Hawaiian and Iceland eruptions (mid-ocean ridges produce only 3 cubic km of lava each YEAR)(Swanson, 1975). The flows would pond in depressions, making lava lakes 30-40 meters thick and 200-400 kilometers in diameter. A few years to a few tens of years would be need for the lava to cool completely (Long and Wood, 1986).


Hon and others (1994) studied lava flows in Hawaii and drew analogies to the Columbia River Flood Basalts. In Hawaii, they documented the growth and inflation of lobes of lava by the internal injection of more lava. With each pulse of new lava, the flow would grow thicker. Flows advance by breakouts at the front of the flow. Applied to the flood basalts, this process was non-turbulent and much slower but could still insulate hot lava great distances from the vent. Self and others (1997) concluded that flows covering 700 to 2,000 square km formed over many years, not days or weeks as implied by the early model. Cross-section of emplacement and inflation of a pahoehoe sheet flow from Self and others (1997).

Development of the Roza compound flow field based on the study of Martin (1989; 1991) and Thordarson (1995). Total time of emplacement for all five major flows is estimated to be 5-15 years. Insulating crust protected the interior, molten cores of flows and allowed the lava to travel great distance from its vent(s). As more lava erupted it caused inflation of flows and break outs of new lava at the flow front. Arrows show direction of flow. Thick bars represent vents. From Self and others (1997).

Numerous features indicate that inflation of lava flows played a major role in the emplacement and growth of the Columbia River Flood Basalts. This photo and sketch show four lobes of lava. A mound of sediment separates lobes 1 and 2. Lobes 2 and 5 are separated by a suture zone that developed between the two lobes as they inflated. Lobe 3 overlies the younger lobes 1 and 2. Photo and sketch courtesy of Thor Thordarson.

This photo and sketch show a tumulus with an axial crack about 50 cm in width (flow 3). As flow 4 covered the tumulus small lobes of lava filled the crack prior to the main core of the flow buried the entire tumulus. Tumuli are common features on inflated lava flows. Photo and sketch courtesy of Thor Thordarson.


This photo and sketch show a tumulus (flow 1) with a large crack in its side. Flow 2 filled the low area adjacent to the tumulus and then sent two thin lobes (tongues) of lava into the crack. Photo and sketch courtesy of Thor Thordarson.


This photo shows the bottom of a pahoehoe flow in the Roza Member resting directly on a scoria bed.

The lava drapes over bombs and small clasts are embedded in the base of the flow. Photo courtesy of Thor Thordarson.

Columbia River Basalts: Further Information


The Cascades Volcano Observatory homepage has additional information on the Columbia River Basalt Group.


Fuller, R.E., 1931, The aqueous chilling of basaltic lava on the Columbia River Plateau: American Journal of Science, v. 21, p. 281-300.

Hooper, P.R., 1987, The Columbia River Flood Basalt Province: Current Status, in Mahoney, J.J., Coffin, M.F., eds., Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism: American Geophysical Union Monograph 100, p. 1-27.

Hooper, P.R. and Hawkesworth, C.J., 1993, Isotope and geochemical constraints on the origins and evolution of the Columbia River Basalt: Jour. Petrol., 34, 1203-1246.

Hooper, P.R., and Conrey, R.M., 1989, The tectonic setting of the Columbia River Basalt eruption: in S.P. Reidel and P.R. Hooper, Eds., Volcanism and Tectonism of the Columbia River Flood Basalt Province: Geol. Soc. Amer. Special Paper 239, p. 293-306.

Hooper, P.R., Gillespie, B.A., and Ross, M.E., 1995, The Eckler Mountain Basalts and associated flows, Columbia River Basalt group: Canadian Journal of Earth Science, v. 32, p. 410-423.

Long, P.E., and Wood, B.J., 1986, Structures, textures, and cooling histories of Columbia River basalt flows: Geol. Soc. America Bull., v. 97, p. 1144-1155.

Reidel, S.P., Tolan, T.L., and Beeson, M.H., 1994, Factors that influence the eruptive and emplacement histories of flood basalt flows: a field guide to selected vents and flows of the Columbia River Basalt Group, in Swanson, D.A., and Haugerud, R.A., eds., Geologic field trips in the pacific Northwest: 1994 geological Society of America Meeting, Chapter 1B.

Schmincke, H.U., 1967, Fused tuff and pepperites in south-central Washington: Geol. Soc. America Bull., v. 78, p. 319-330.

Self, S., Thordarson, T., and Keszthelyi, L., 1997, Emplacement of continental flood basalt lava flows, in Mahoney, J.J., Coffin, M.F., eds., Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism: American Geophysical Union Monograph 100, p. 381-410.

Swanson, D.A., and Wright, T.L., 1981, Guide to geologic field trip between Lewiston, Idaho, and Kimberly, Oregon, emphasizing the Columbia River Basalt group, in Guides to some Volcanic terranes in Washington, Idaho, Oregon, and Northern California, Circular 838, edited by D.A. Johnston and J. Nolan, p. 1-16.

Swanson, D.A., and Wright, T.L., 1978, Field guide to field trip between Pasco and Pullman, Washington, emphasizing stratigraphy, vent areas and intra-canyon flows of the Yakima Basalts, in Proceedings, Geological Society of America, Cordilleran Section meeting, Pullman.

Swanson, D.A., Wright, T.L., and Helz, R.T., 1975, Linear vent systems and estimated rates of magma production and eruption for the Yakima Basalt on the Columbia Plateau: American. J. Science, v. 275, p. 877-905.

Swanson, D.A., Wright, T.L., Hooper, P.R., and Bentley, R.D., 1979, Revisions in stratigraphic nomenclature of the Columbia River Basalt Group: U.S. Geological Survey Bulletin 1457-G, p. G1-G59.

Swanson, D.A., Anderson, J.L., Camp, V.E., Hooper, P.R., Taubeneck, W.H., and Wright, T.L., 1981, Reconnaissance geologic map of the Columbia River Basalt Group, northern Oregon and western Idaho: U.S. Geological Survey Open-File Report 81-797, scale 1:250,000.

Swanson, D.A., Cameron, K.A., Evarts, R.C., Pringle, P.T., and Vance, J.A., 1989, Cenozoic volcanism in the Cascade range and Columbia Plateau, southern Washington and northernmost Oregon: Field excursions to volcanic terranes in the western United States, Volume II: Cascades and Intermountain West, Chapin, C.E., and Zidek, J., eds., New Mexico Bureau of Mines & Mineral Resources Memoir 47, p. 1-43.

Taubeneck, W.H., 1970, Dikes in the Columbia River basalt in northeastern Oregon, western Idaho, and southeastern Washington; in proceedings of Second Columbia River basalt Symposium: Eastern Washington State College press, Cheney, p. 73-96.

Thordarson, Th., and Self, S., in review, The Roza Member, Columbia River Basalt group: A gigantic pahoehoe lava flow field formed by endogenous process?: Journal of Geophysical Research (volume on Long Lava Flows).

Tolan, T.L., Reidel, S.P., Beeson, M.H., Anderson, J.L., Fecht, K.R., and Swanson, D.A., 1989, Revisions to the estimates of the areal extent and volume of the Columbia River Basalt Group, in Reidel, S.P., and Hooper, P.R., eds., Volcanism and tectonism in the Columbia River flood-basalt province: Geological Society of America Special Paper 239, p. 1-20.

Walker, G.P.L., 1968, Compound and simple lava flows and flood basalts: Bull. Volcanology, v. 35, p. 579-590.

Waters, A.C., 1961, Stratigraphic and lithologic variations in the Columbia River basalt: American J. Science, v. 259, p. 583-611.