Just a few years after Boyd's paper appeared, the U.S. Geological Survey mounted an extensive investigation of Yellowstone's geology, assigning some of its brightest young scientists to the task. Among them was Bob Christiansen, who studied the young ash flow tuffs in great detail. What follows is based on his research and that of his co-workers, including geologists, chemists, and geophysicists, some of whom continue their studies of Yellowstone today.

Christiansen and his team recognized that not one but two welded tuffs rimmed the plateaulava flows; one was 2.1 million years old (Huckleberry Ridge Tuff ) and the younger 0.65 million years (Lava Creek Tuff ). A third tuff, to the west in Idaho, was 1.3 million years old (Mesa Falls Tuff). Together they form the Yellowstone Group of tuffs.

"The Wonders of Yellowstone" - 98 Minutes - ~Telly Award Winner for Nature and Wildlife~ Two years in the making and just released, "The Wonders of Yellowstone" video has been highly requested, produced in DVD format and is now available. Take a complete tour of Yellowstone National Park as our Narrator Cathy Coan guides you to all the wonders of the park including all the geyser basins, wildlife, waterfalls and much more. We previously sold travel packets but these packets, maps and trail guides are all available at the park for free or minimal charge. More Info or Order Online

These tuffs demonstrated conclusively that the volcanic events forming Yellowstone were not the products of many million years of geologic change ending many millions of years ago. Rather, their time scale was compressed into only the last two million years. A long geologic history would have allowed a more leisurely progression of events-a lava flow here, then a million years later another flow there. A longer geologic history would also have called for intermittent periods of magma (molten rock) formation separated by periods of volcanic quiescence. Instead, this short time scale compressed the sequence of explosions and flows and required a heat source much larger and younger than ever before imagined.

Caldera's are large basin-shaped volcanic depressions more or less circular in form. Caldera eruptions on the Yellowstone scale have a world wide frequency of perhaps once every hundred thousand years. Somewhat smaller eruptions, on the scale of Crater Lake-Mount Mazama in Oregon, are more frequent, perhaps every 1,000 years or less. Such explosive eruptions were not isolated events. Rather, they were climactic stages of magmatic processes that extended over hundreds of thousands of years.

No one has ever seen a volcanic explosion on the scale of the Yellowstone eruptions, but smaller explosions have been observed and their activity described. Consider Mount Tambora, on the island of Sumbawa, Indonesia to grasp some idea of what's involved when a caldera forms during or just after an ash flow eruption. For about three years the volcano rumbled and fumed before a moderate eruption on April 5, 1815 produced thundering explosions heard 870 miles away. Next morning volcanic ash began to fall and continued to fall though the explosions became progressively weaker,

On the evening of April 10 the mountain went wild. Eye witnesses 20 miles away described three columns of flame rising from the crater and combining into one at a great height. The whole mountain seemed to be covered with flowing liquid fire. Soon these distant viewers were pelted with 8-inch pumice stones hurled from the volcano. Clouds of ash, borne by violent gaseous currents, blasted through nearby towns blowing away houses and uprooting trees. The village of Tambora was destroyed by rolling masses of incandescent, hot ash.

On April 16, booming explosions loud enough to be heard on Sumatra, 1600 miles to the west, continued into evening. Mount Tambora, still covered with clouds higher up, seemed to be flaming on its lower slopes. For a day or two, skies turned jet black and the air cold. When the eruption ended, the ash cloud drifted west and settled on all islands downwind. With the expulsion of so much magma, the mountain collapsed, unsupported from within, forming a great caldera. Lombok, 124 miles to the west, was covered by a blanket of ash two feet thick. Tidal waves crashed on islands hundreds of miles away. Waves and ashfalls killed more than 88,000 people.

Ash blasted into the stratosphere circled the earth several times causing unusually beautiful sunsets in London early that summer. In 1816, mean temperatures in the northern hemisphere dropped by half to more than 1° E Farmers in Europe and America called this the year without a summer.

Tambora's eruption was the largest and deadliest volcanic event in recorded history. How does it compare with the Yellowstone caldera eruptions? If we reduce all the ash from Tambora to dense rock equivalents and include all ash flow tuffs that formed at the same time, we come up with about 36 cubic miles of rock. Quite a bit compared with the destructive U.S. eruptions of Mount St. Helens in 1980 that produced about 1/4 cubic mile.

Both of these shrink to insignificance when compared with Yellowstone. The volume of volcanic rock produced by the first Yellowstone caldera eruption was about 600 cubic miles-about 17 times more than Tambora, and 2,400 times as much as Mount St. Helen's, an almost incomprehensible figure. One more statistic: Ash from Tambora drifted downwind more than 800 miles; Yellowstone ash is found in Ventura, California to the west and the Iowa to the east. It is likely the earth has seldom in its long history experienced caldera explosions on the scale of those that created Yellowstone.

Three gigantic caldera eruptions rocked the Greater Yellowstone Ecosystem. The first and largest, Huckleberry Ridge caldera, blew up about 2.1 million years ago. Its center was in western Yellowstone National Park, but it extended into Island Park, Idaho. Welded tuff from this cycle is called the Huckleberry Ridge Tuff. The yellow rocks along the road in Golden Gate between Mammoth Hot Springs and Swan Lake Flats are Huckleberry Ridge Tuff. So are the tuffs that hold up much of Signal Mountain in Grand Teton National Park, and that crop out along the west side of the Teton Range, in Idaho.

The second great explosion formed the Island Park caldera 1.3 million years ago. This caldera, the smallest of the three, lies just west of Yellowstone in Idaho, within the western part of the Huckleberry Ridge caldera.

The youngest caldera, Lava Creek, erupted the Lava Creek Tuff, 0.65 million years old. It overlaps the Huckleberry Ridge caldera, but its eastern margin is about 10 miles farther east. Because it is the youngest, its tuffs and associated lava flows are best exposed and its history best known. Its eruption may have destroyed the south part of the Washburn Range.

Although the Lava Creek Tuff is 0.65 million years old, its caldera began to evolve about 1.2 million years ago when rhyolite lavas flowed intermittently onto the surface of the Yellowstone Plateau from slowly forming, crescentic fractures. Over a period of 600,000 years these ring fractures grew and coalesced to form a system of fractures enclosing the part of the plateau that later collapsed into the Lava Creek caldera.

The ring fractures were a surface expression of a huge body of magma or molten rock, forming in upper levels of the earth's crust. As the magma chamber grew in volume, it stretched and bulged the crust above it. The upper crust was rigid and brittle; it fractured more easily than it bent; thus fractures, or faults, developed around the bulge. As the bulge rose higher, the ring fractures propagated downward toward the magma chamber.

In the magma chamber itself, the molten material was evolving chemically. Less dense materials were concentrating in the upper part of the chamber, including the more silica-rich magma, various gases, and water. Then, with maximum segregation in the magma chamber, volatiles concentrating in its upper part, and ring fractures propagating downward, the gun was loaded and cocked. What actually triggered the caldera-forming explosions is hard to say, but the pressures in the magma chamber must have exceeded the gravitational pressures of the overlying rocks.

Imagine a bottle of carbonated water lying in the sun. Pick it up, shake it vigorously, maybe tap the cap...boom, it blows off. Instantly the pressure in the bottle drops, the dissolved carbon dioxide exsolves into bubbles and an expanding mass of bubbles and water jets into the sky. In a few seconds, the event is over. Wipe off your face and check the bottle; some of the water remains, but most of the gas is gone. This simple scenario is a scaled-down analogy of what happened 600,000 years ago in Yellowstone when the volatile-rich upper part of the magma chamber vented and erupted the Lava Creek Tuff. The exolv-ing gas expanded in the magma, making a much larger volume of frothy fluid. This expanding, low-density hot gas and magma mixture rose rapidly. It vented at the surface as a sustained explosion of white-hot froth. A scene from the depths of Dante's Inferno.

Driven by hot vapors, giant fountains of incandescent ash at temperatures near 1,800° F burst from the ring fractures. Plumes of ash jetted into the stratosphere where planetary winds carried it around the world blanketing tens of thousands of square miles with thin coverlets of volcanic dust. Nearer the vents, fiery clouds of dense ash, fluidized by the expanding gas, boiled over crater rims and rushed across the countryside at speeds over one hundred miles per hour, vaporizing forests, animals, birds, and streams into varicolored puffs of steam. Gaping ring fractures extended downward into the magma chamber providing conduits for continuing foaming ash flows.

More and more vapor-driven ash poured from the ring fractures, creating a crescendo of fury. As the magma chamber emptied, large sections of the foundering magma chamber roof collapsed along the ring fractures, triggering a chain reaction that produced a caldera 45 miles long and 28 miles wide.

Hot ash flows are fascinating. Driven by expanding gas, they are really clouds of hot glass shards and pumice plus expanding gas whose turbulence keeps everything flowing like water. But as the gas escapes, the viscosity increases, motion ceases, and the ash settles into a layer more than one hundred feet thick. This deposit is still extremely hot, and as it compresses under its own weight, the sticky glass shards fuse into a welded tuff. The upper part of the ash cools too rapidly to weld and is either unconsolidated or weakly cemented by vapors of escaping gas.

The engine of destruction didn't take long to run down, just a few hours or, at most, a few days. Hours? Days? Yes, incredible as it may seem. Evidence for the astonishing rapidity of this eruption is found in detailed study of the tuff. Eruptions that are separated by any significant period of time have discernible boundary effects that clearly separate one tuff from another. Runoff water, for example, would erode small channels in the surface of a flow or the chilled tops of separate flows would mark the emplacements of separate cooling units. No evidence exists to suggest such a cooling history in Yellowstone. Rather, the caldera venting appears to have developed in two separate parts of the magma chamber simultaneously and been continuous over a very short time. In a period of time reasonably inferred to be hours, more than 240 cubic miles of Lava Creek Tuff was emplaced around the caldera rim and within the caldera itself.

The explosions died away. A complex ecosystem was snuffed out and replaced by a sterile, steaming moonscape where hardly a living thing survived. The Yellowstone Plateau, the Teton Range, and thousands of surrounding square miles of Wyoming, Montana, and Idaho were barren and nearly lifeless for the third time in two million years.

The caldera-forming magma chamber, however, like our fizzed-out soda bottle, was far from empty. In fact, it may have contained 90 percent of its original magma volume. No sooner did the magma chamber roof collapse, than it began to rise again owing to pressures from underlying magma. Two resurgent domes soon began to form near the center of the elliptical caldera, one near Le Hardy Rapids on the Yellowstone River, and another east of Old Faithful.

The rejuvenated magma chamber also sent rhyolite to the surface where its eruption formed lava flows that buried part of the western resurgent dome and completely buried the caldera's western rim. Three such eruptive pulses about 150,000, 110,000, and 70,000 years ago produced about 240 cubic miles of rhyolite.

Because rhyolite lavas are rich in silica and poor in water, they tend to be quite viscous. Instead of flowing easily and rapidly as does Hawaiian basalt, rhyolite lava form piles of taffy-like incandescent rock whose margins will advance so slowly that observers will have to watch closely to see them moving.

Young rhyolite flows provide much of central Yellowstone's beauty; its lakes, waterfalls, and stream courses. For example, Yellowstone Lake fills a basin in the southeast part of the 600,000 year-old caldera between the east rim of the caldera and rhyolite flows on the west. Shoshone and Lewis lakes fill basins formed between adjacent flows. The Upper and Lower falls of the Yellowstone River tumble over resistant layers in caldera-filling flows. Nez Perce Creek, from its headwaters to its junction with the Firehole River, flows along a seam between lava flows. So does the Firehole River itself to its junction with the Madison River. The Gibbon River is pinched between younger flows and the Lava Creek Tuff through much of its course.

Driving west from Canyon Village you climb the steep eastern front of the Solfatara flow, drive miles across its rolling top, then descend its western slope to Gibbon River. Similarly, the drive from West Thumb to Old Faithful crosses several young rhyolite flows.

Silica, the primary constituent of rhyolite, provides a relatively sterile soil environment that is unfriendly to most living things. But not lodgepole pine. These hardy trees, pine grass, and fire-weed love such inhospitable sites. Their adaptability is why you see so many miles of boring lodgepole forest along Yellowstone roads.

In summary, three caldera eruptions and associated lava flows produced about 1,600 cubic miles of rhyolite in the last 2.1 million years. This staggering figure requires rates of magma production comparable to the most active volcanic regions on earth, such as Iceland and Hawaii. As we shall see, the processes that produced this enormous amount of magma also uplifted significant portions of northwestern Wyoming, southwestern Montana, and southern Idaho.

The Caldera Today

Is Yellowstone's history of volcanic activity at an end? Has time tamed its explosive violence, leaving only a heritage of aging geysers and eroding lava flows? Has the magma chamber beneath Yellowstone exhausted its supply of molten rock? Is it now incapable of producing more lava flows or explosions? Well, let's consider these questions; questions that have intrigued scientists ever since Yellowstone was discovered.

Anyone who has seen a geyser or hot spring immediately thinks of heat. Early geologists speculated that the heat in geyser waters came from the cooling of young lava flows beneath the geyser basins. They speculated that rain and snow meltwater percolated into gravels and sands of the basins and into the young lava flows where it was heated before rising to the surface via geysers and hot springs. The lava flows were thought to be young, but even the most daring geologist tucked them well back in time. As we learned in Chapter 4, however, U. S. Geological Survey studies that dated the lava flows found some of them to be rather young, indeed.

Given that the youngest lava flows are only 70,000 years old, yesterday in geologic time, might not there still be molten magma beneath Yellowstone today? Direct methods, such as deep drilling, have not been employed to test this possibility, but other methods suggest magma exists beneath Yellowstone.

The earth's interior is warmer than its surface causing heat flow outward to the surface. The flow of heat in geyser basins is hundreds of times greater than normal heat flows. If the total conductive heat flow of major hydrothermal basins is averaged over the 965 sq. miles of the Yellowstone Caldera, we find flow levels that are 60 times greater than mean global rates.

Geophysical studies monitor the caldera and its magma body indirectly. From seismic studies we learn that shock waves from earthquakes and man-caused explosions traveling through the earth's crust are slowed significantly as they pass beneath the caldera. Material with a seismic velocity that is slower than normal underlies the caldera at depths as shallow as I mile. This may indicate local zones of molten magma in the upper crust. Near the northeast part of the caldera, seismic velocities are even lower to within about 2 miles of the surface; this may indicate a more continuous magma body that extends from the northeastern part of the caldera to about 10 miles beyond it. Down below the crust and in the mantle at depths of 100 miles, lower than normal local seismic velocities may indicate thin rising columns of magma.

Earthquake data also suggest that soft or molten rock is close to the surface of Yellowstone. Minor earthquakes jiggle Yellowstone hundreds of times each year, but above the caldera the foci of these quakes are extremely shallow, less than three miles below the surface. These clues suggest that the material underlying Yellowstone is still very hot and ductile, as would be expected if a magma chamber still exists.

Gravity studies back up conclusions drawn from seismic data. We know that gravity values across the Yellowstone Plateau are much lower than normal, and low gravity values are associated with low rock densities. In Yellowstone the low densities imply molten, thermally expanded material. As you might expect, the lowest gravity anomalies are found in the same place where seismic velocities are slowest-under the northeast caldera rim and beyond.

Local uplift and subsidence within Yellowstone are fast enough to be measured by surveying techniques. Benchmarks, points of precisely measured altitude, were established along the road systems of Yellowstone in 1923. One center of uplift on these surveys is at Le Hardys Rapids in the central part of the Yellowstone caldera and 3 miles down the Yellowstone River from its outlet from Yellowstone Lake. Until 1985, these surveys showed uplift at a rate of about V^ inch a year centered on Le Hardys Rapids, with total uplift since 1923 of about 3 feet. The profile of the Yellowstone River on both sides of Le Hardys Rapids suggests this uplift has been going on for a much longer time. Upstream from Le Hardys Rapids, the Yellowstone River is remarkably tranquil with a low gradient, whereas downstream it is many times steeper. Carbon dating of muds in the drowned channel of the Yellowstone River upstream from Le Hardys Rapids shows this overall uplift cycle had started by 3,000 years ago. Surveys in 1986 and later show this pattern of uplift has changed to subsidence, also at a rate of about 1/2 inch a year. We do not know if the change after 1985 represents the start of a major interval of subsidence or a minor reversal in a longer interval of uplift, but surveys as recent as 1993 show subsidence.

These various investigations of hydrothermal features, heat flow, seis-micity, earthquakes, gravity, and historic altitude change give us an interesting picture of what underlies the Yellowstone Plateau. These conditions are consistent with a large, partly molten magma body at shallow depth that extends northeast of the caldera rim. Although rocks underlying the rest of the caldera have low densities and low seismic velocities, the variations are less extreme, so the rocks there may be very hot but not necessarily contain much molten magma.

Thus we see that Yellowstone's fires are only banked, not out. Geologists don't expect another caldera explosion any time soon, but sometime new lava flows quite likely will once again consume lodgepole forests, and a new generation of geysers will burst forth, perhaps in Hot Springs Basin.