Geology of the Jemez Area, Chapter 12: The Recent Past

The resurgence of the Valles caldera and the eruption of the ring domes was not the end of geological history in the Jemez. Geology is an ongoing process. In this chapter, we will look at geological processes in the last half million years — the very recent past, in geologic terms.

Potassium-argon dating becomes inaccurate for igneous rocks younger than about 100,000 years in age, because there isn't time for enough argon to accumulate to permit an accurate measurement. Carbon-14 dating is accurate only for ages of less than about 50,000 years. Other methods of dating are inherently uncertain. As a result, events occurring between 100,000 and 50,000 years ago were long difficult to date accurately. Fortunately, high-precision argon-argon dating has begun to close this gap.

An important source of information on life during this time period is caves, where animal and plant remains are preserved by cool temperatures and dry conditions away from sunlight. Such remains are not actually considered fossils, since there has not been time for significant mineralization to take place. Little of the original organic material has been replaced by minerals carried by groundwater.

Brrr!

Digital relief map of suspected rock glaciers in the
Jemez Mountains

Relief map of the Jemez with suspected rock glaciers highlighted in red and Shinarump lag field in yellow.

The geological record indicates that the Earth's climate is usually mild and stable. However, there have been several intervals of geological time when the climate became unstable, with large ice sheets developing near the poles and periodically advancing over much of the temperate zone. These intervals of geological time are known as ice ages or glaciations.

There have been at least five ice ages during the history of the Earth. Each occurred when the balance between heat arriving from the Sun and heat radiated away from the Earth was upset. Since astronomers have concluded that the Sun has steadily grown brighter over the last 4.5 billion years, decreases in the solar input are not a likely cause of ice ages (though there is evidence that shorter cold spells may be due to changes in solar activity.) However, the motion of the continents can affect solar heating, because continents reflect more of the Sun's radiation into space than oceans do. When continents are covered by epicontinental seas, this effect is reduced. It is also reduced when continents drift away from the equator, where most of the solar radiation falls. But the most likely cause of ice ages is reduced levels of greenhouse gases, such as carbon dioxide, in the atmosphere. Carbon dioxide is a greenhouse gas; it lets the relatively short-wavelength radiation from the Sun pass through to heat the Earth, but traps the longer-wavelength radiation re-emitted by the Earth. Carbon dioxide levels increase when volcanic activity is high or when vegetation dies off and decomposes. It is reduced when vegetation flourishes or when large areas of fresh silicate rock are exposed at the surface, allowing carbon dioxide to combine with sodium, calcium, and potassium in feldspar. Geologists estimate that weathering accounts for 80% of carbon dioxide that has been removed from the atmosphere, while coal and other carbon deposits from vegetation are a mere 20%.

The first known ice age occurred around 2.3 billion years ago, in the Paleoproterozoic, and likely was caused by reduced methane levels from the proliferation of cyanobacteria. Methane is an even more powerful greenhouse gas than carbon dioxide, but the methane in the early atmosphere was oxidized to carbon dioxide by oxygen generated by photosynthesis. The second ice age occurred in the Neoproterozoic, 770 to 590 million years ago, and was the worst ice age the Earth has experienced. It was likely caused by continuing decline in carbon dioxide levels and the movement of the continents towards the equators. This Varangian glaciation, as it is called, may have covered almost the entire earth in an ice sheet -- the "snowball Earth." When the ice finally retreated, an explosion of new forms of life took place, marking the beginning of the Phanerozoic Era. Two less spectacular ice ages occurred in the Ordovician-Silurian time interval, 480 to 420 million years ago, and in the Carboniferous-Permian time interval, 330 to 250 million years ago.

All these ice ages were characterized by periodic advances and retreats of the ice sheets. This is probably because the advance of the ice reduced vegetation cover and allowed carbon dioxide to build up and trap more heat. When the ice retreated, vegetation began proliferating again and removed the carbon dioxide from the atmosphere, allowing the climate to cool again. Each advance is called a glacial period or glaciation and the intervening warm spells are called interglacials.

We are living in an ice age now, likely triggered by the collision of the subcontinent of India with Asia and the exposure of vast amounts of fresh silicate rock in the Himalayan Mountains. Beginning about 2.6 million years ago, which marks the beginning of the current Quaternary period, ice began building up in Antarctica, which has remained mostly covered with an ice sheet ever since. Certain species of fossil algae called discoasters became extinct and, almost simultaneously, the first modern horses appeared, and this moment in time is formally defined as the start of the Pleistocene, the final epoch preceding our modern Holocene epoch. Greenland soon iced over as well, and there have been at least fourteen distinct glacial periods since then, each lasting about 100,000 years. These were separated by relatively brief interglacial periods of warmer weather in which the ice sheets retreated. We are living in such an interglacial period today, the Holocene, which began 11,700 years ago. Estimates of when the next glacial period will occur range from within centuries to over 50,000 years, depending on various assumptions about the causes of glacial periods (which are still uncertain) and the amount of carbon dioxide that will be released into the atmosphere by ourselves and our descendants.

Sea level today is somewhat higher than has been typical of the Cenozoic. All of written human history has taken place during an interglacial, a period of relative warmth between glaciations. The oceans today are high enough to flood the margins of the continents, forming continental shelves under shallow water in the North Sea, the Bering Strait, and elsewhere. Shallow seas covering continental crust are called epicontinental seas. The continents were almost entirely above water, with very little submerged continental shelf, during the peaks of the ice ages. But even the continental shelves of today are small compared with what they have been in the geological past: The Cenozoic has been a period of unusually low sea levels.

The Quaternary ice age is the culmination of a long period of global cooling that commenced in the Oligocene, 34 million years ago, when South America and Australia rifted away from Antarctica and turned that continent into an ice box. This has led some geologists to identify the Quaternary Ice Age as merely the culmination of a much longer Cenozoic Ice Age.

During the Quaternary glacial periods, ice sheets spread over much of North America and had a profound effect in shaping the topography of the northern and more mountainous parts of the continent. The Great Lakes were produced by glaciers, as was Long Island. However, ice sheets did not extend as far south as the Jemez Mountains, and there is little evidence that large glaciers ever formed there. Valleys in the Jemez generally have the V-shape characteristic of valleys cut by rivers rather than the distinctive U-shape of valleys cut by glaciers.

However, there are several boulder fields in the higher terrain of the Jemez that have been interpreted as inactive rock glaciers. Rock glaciers form when a boulder field is saturated with water which freezes to ice, and the ice does not completely melt during the summer months. The outer layer of boulders acts to insulate the ice. Such permanent ice is called permafrost. The ice accumulates within the boulder field to the point where the field becomes ductile, and the mixture of rock and ice can slowly flow downhill. Current conditions in the Jemez are much too warm to permit permafrost, even at the maximum elevations of around 3400 meters (11,000'), but temperatures were significantly lower during past glacial periods and may have been conducive to permafrost.

This boulder field shows definite flow features in the satellite photograph. It is also quite striking at close range.

Particularly striking is the smooth slope of the hillside on which this glacier sits. There are no scattered boulders beyond a few feet of the edge of the flow.

The pictures don’t quite convey the bowl shape of the glacier surface. One gets the impression that the center was significantly deflated when the ice melted away, which is probably about right.

The rocks are covered with lichen. This suggests the flow has not been active in a very long time. There is actually such a thing as lichenometric dating, based on the diameters of individual lichen colonies, though it’s very uncertain. But one can make a guess that the flow has not moved in at least two thousand years.

I did notice one patch at the south edge that looked very fresh. This small area had either moved recently, or a forest fire burned off the lichen. Some of the boulders were spalled, as if they had been subjected to fire.

You can see a logging road crossing the flow in the upper photograph. Here’s a panorama looking west from the edge of that road.

Behind the trees at left is Rendondo Peak and Redondito. At center is the rugged terrain of debris flows, megabreccia, and low-volume Deer Canyon Member flows northeast of Redondo Peak. At far right are some of the other ring domes of the Valles caldera.

For comparison, here is an active rock glacier on the flank of Mt. Lincoln, near Alma, Colorado.

The rock glacier sits in a hanging valley, which is a result of conventional glacier activity. During the height of the last glaciation, a glacier ran down this valley and joined a larger glacier in the foreground valley. Now the valley is occupied by the rock glacier, which has the same characteristic steep front as the inactive rock glacier on Cerros del Abrigos.

With its steep front and flanks, its lack of any matrix between the boulders, and the indications of pressure ridges in the satellite photograph, the Cerros del Abrigos boulder field is almost certainly an inactive rock glacier.

The boulder fields of the Jemez likely first formed as debris flows or lag deposits. Lag deposits occur in many geological settings; for example, there are striking lag deposits in the Plaza Blanca area, where nearby river terrace gravels provide a source of large clasts.

By way of contrast, there are lag fields on Cerro Rubio and other locations that show no flow features and were probably never rock glaciers.

The period from 500,000 to 60,000 years ago marked a pause in volcanic activity in the Jemez. Sometime during this pause, but at least 200,000 years ago, the first bison arrived in North America. The appearance in the fossil record of Bison antiquus, the direct ancestor of the modern bison, defines the beginning of the Rancholabrean land mammal age of North America, which is named after the famous tar pits at Rancho La Brea in what is now Los Angeles. Bison fossil fragments were discovered just north of the village of White Rock in the 1990s.

Homo sapiens likely also arose during this time period, perhaps around a quarter of a million years ago. Homo neanderthalis appeared at about the same time, and the two closely related species would share the planet until at least 40,000 years ago. Or longer: There are indications that modern humans carry some Neanderthal genes.

A final burst of volcanism

The Jemez is not an extinct volcanic field, and will likely erupt again someday. There are, in fact, indications that a new cycle of volcanic activity has already commenced — though I hasten to add that there is little likelihood we will see an eruption in our lifetime.

The most recent eruptions in the Jemez are those that produced the El Cajete Pumice, the Battleship Rock Ignimbrite, and the Banco Bonito Flow, all of which have similar chemistry, seem to have originated from closely spaced vents, and are likely the result of a fresh batch of magma rising into the Valles magma chamber. Collectively they are known as the East Fork Member of the Valles Rhyolite.

The El Cajete Pumice

Around 74,000 years ago, a vent opened south of Redondo Peak at El Cajete. The date of this eruption was long very uncertain. This is because very young rocks are very difficult to date precisely by the usual radioisotope methods. There simply isn't time in a hundred thousand years for enough argon to accumulate for a reliable radiometric date based on the usual potassium-argon dating. Charcoal from burnt logs preserved in the volcanic beds from this eruption shows no measurable ¹⁴C, which means the eruption took place at least 42,000 years ago. Other dating methods, such as fission track, electron spin resonance, thermoluminescence, and optically stimulated thermoluminescence, attempt to measure accumulated radiation damage to a rock to determine the time since the rock was hot enough to erase such damage. ²¹Ne dating is slightly different in that it measures accumulation of ²¹Ne in rocks from cosmic ray bombardment rather than the accumulated damage from the rock's own radioactivity. These methods yield highly discordant results, varying from 140,000 years from fission track dating to as little as 26,000 years from ²¹Ne dating. Until quite recently, the accepted estimate, based mostly on electron spin resonance measurements, was that the eruption took place around 55,000 years ago.

New high-precision argon-argon dating has helped more precisely date the El Cajete flows. This technique relies on irradiating the rock with neutrons in order to convert a nonradioactive isotope of potassium, ³⁹K, to ³⁹Ar. When done under carefully controlled conditions and alongside with calibration samples, this allows the geologist to measure a very precise ratio of the ³⁹Ar so produced to ⁴⁰Ar produced from natural decay of ⁴⁰K. This in turn yields a very precise measurement of the ratio ⁴⁰Ar to total potassium and thus the age of the rock. High-precision argon-argon dating of the East Fork Member flows puts their ages very close to 74,000 years.

The eruption produced an eruption column that blanketed much of the southern Jemez with pumice beds. The map above shows only those areas where the beds are most substantial. El Cajete pumice occurs as isolated drifts and thin beds over much of the southern Jemez and covers many of the Bandelier Tuff mesas in western Bandelier National Monument.

The flat area is mostly alluvium, carried in by water after the eruptions from the vent ceased, but there is a patch towards the center that is an original deposit of El Cajete pumice. There are also some deep pits in the far side of the flats, used as barbecue pits for feeding guests when this was part of a private ranch.

At left in the photograph is a rim of pumice thrown up around the vent. At right is Redondo Peak. The low ridge at center is the east end of the Banco Bonito obsidian flow, which we'll have more to say about later on. The vent for the El Cajete pumice is though to be directly under El Cajete flats, but the vent for the Banco Bonito flow is further west, not far from the edge of the flow seen here.

The pumice rim is particularly well-preserved and remarkably uniform on the southwest side of the vent.

Although the El Cajete eruption produced widespread air fall pumice beds, the area immediately around the vent has coarser eruption products. These include large clasts of snowflake obsidian.

Snowflake obsidian is so named because of the white patches. These can vary in size from fairly small, as in these samples, to an inch across.

The white patches are typically cristobalite, a form of silica that is stable at high temperature. This poses a bit of a puzzle, as the temperatures measured in modern rhyolite magmas are not high enough to be in the stability range for cristobalite. The explanation is given by Ostwald's step rule, which says that supercooled liquids most easily crystallize to their most disordered solid form. Cristobalite is less ordered than quartz, and so a relatively cool rhyolite magma has an easier time crystallizing to cristobalite than quartz, even though quartz is the most stable product from very slow cooling.

These clasts are likely remnants of an obsidian dome that formed on the east side of the vent. Geologists have found evidence for at least three such glassy domes, each of which marked a pause in the explosive eruptions of pumice and pyroclastic flows.

Vesicular rhyolite is not bubbly enough to be classified as pumice, nor massive enough to be just plain rhyolite. The chunks I hefted felt like tuff. This chunk is larger than a basketball and is an impressive distance from the vent; given that it’s in a sheetwash deposit, and given the lay of the land, it probably originally landed even further from the vent.

Thick beds of El Cajete air fall pumice are exposed in road cuts along State Road 4 near the source vent.

El Cajete pumice
along State Road 4 near the source vent

South Mountain Rhyolite, El Cajete Pumice, and Banco Bonito obsidian in road cut. Near 35 49.675N 106 35.413W

Here the pumice lies on top of South Mountain Rhyolite and in turn is buried under Banco Bonito obsidian. Here's a sample of the pumice.

There is an impressive bed of El Cajete pumice in a driveway cut as one begins the ascent out of the Vallecitos de los Indios onto the south caldera rim along Forest Road 10.

El Cajete Pumice beds in the southern Jemez are relatively easily worked with simple agricultural implements and tend to hold moisture, and archaeologists have concluded that they played an important role in Native American settlements in the area. More recently, they have been extensively mined for use as abrasives, as an additive in concrete, for water filtration, and as the stone used to manufacture stone-washed jeans.

Many of the more distant beds are not primary beds (air fall in place) but have been reworked by streams. Some reworked deposits are found in Paliza Canyon.

There have been many pumice eruptions in the southern Jemez in the last eight million years, but El Cajete is a good first guess whenever loose pumice is found that has no obvious overlying beds. This can be confirmed by closer examination of the pumice, which reveals the distinctive mica phenocrysts.

The Battleship Rock Ignimbrite

Digital relief map of Battleship Rock Flow exposures in
the Jemez Mountains

Relief map of the Jemez with Battleship Rock Flow exposures highlighted in red.

The eruption of the El Cajete Pumice was followed by a series of pyroclastic flows that generally followed existing river valleys. These are most prominently exposed at Battleship Rock and along Cañon de San Diego to the north, where the Jemez River has cut back through the tuff to reestablish its drainage.

Battleship Rock itself marks the confluence of the Jemez and East Fork Jemez rivers, which have cut through the tuff deposited in their valleys to leave this dramatic erosional remnant. The rock is a welded tuff showing classical fiamme:

The rock is full of lithic fragments (bits of older rock caught up in the pyroclastic flow). The elongated brown patches are bits of soft rock, perhaps pumice or tuff (though the color is rather dark for that), that were softened and flattened as the very hot pyroclastic flow settled onto the surface and welded together.

The tuff is quite glassy north of Battleship Rock, so that the canyon walls glisten under the right lighting conditions.

Battleship Rock Ignimbrite at a different time of day. Looking northeast from 35 50.645N 106 38.185W

Further north, the flow becomes less densely welded and resembles Bandelier Tuff.

The Banco Bonito Flow

The most recent eruption in the Jemez produced the Banco Bonito Flow, which fills much of the southwest moat of the Valles caldera. This flow shows up clearly on satellite images and relief maps, because it is young enough to have experienced little erosion.

Perhaps the best view of the Banco Bonito Flow is had from the south rim of Valle de los Indios.

At center is Redondo Border and to its right is Redondo Peak. The knob right of Redondo Peak is South Mountain. Rhyolite of the South Mountain flows is exposed in the canyon bottom here, beneath the younger flows of the East Fork Member. The knob to the right of South Mountain, which is actually beyond Redondo Peak, is Cerro del Abrigo on the northeast side of the caldera. The two peaks dominating the right side of the panorama are Los Griegos and Cerro Pelado on the caldera south rim.

The plateau extending across much of the panorama is the Banco Bonito flow. This is an obsidian flow, now mostly devitrified. The rugged cliffs across the canyon are devitrified Banco Bonito obsidian resting on South Mountain Member flows, with only occasional small outcrops of the Battleship Rock flow exposed between the two.

Here's a somewhat closer view, from a stretch of highway where a recent forest fire exposed the bench.

The forested ridge in the middle distance, extending from and disappearing behind the trees at right, is Banco Bonito. Redondo Peak is right of center. El Cajete is located behind the ridge just to the right of Redondo Peak.

The source vent was likely located just west of El Cajete. Here the obsidian seems to have formed a dome before flowing into the lower terrain to the west.

The slopes here are covered with clasts of partially divitrified, flow-banded obsidian.

The long hill is likely a pressure ridge, which is a kind of ripple or wave produced on the semisolid surface of a flow by movement of still-liquid rock within. There are clearer examples further west along the Banco Bonito.

Like the El Cajete Pumice, the Banco Bonito Flow is well exposed on State Road 4 near its source vent.

In this area. the road cut has exposed the interior of the flow, which is only partially devitrified. Much obsidian, though not of tool quality, is exposed here.

The relief map shows that the flow has a number of large explosion pits. One of these is located on Forest Service land just east of State Road 4. Here's a view from the rim of the pit.

This kind of explosion pit probably does not represent a source vent, since these pits are distributed more or less randomly across the surface of the flow. Similar features are seen at Newberry Volcano, Oregon, and at other young obsidian flows. The likely explanation is that they are phreatomagmatic explosion craters, formed when groundwater accumulates within the upper part of the flow while its lower portion is still very hot. If drought lowers the water table, the decrease in pressure can allow the remaining groundwater to flash into steam and blow out very large craters like this one.

Support for this theory comes from the presence of broken blocks of devitrified obsidian in and around the crater.

Some very young explosion craters of this kind are found at Yellowstone, and the formation of one such crater in 1989 was actually witnessed by geologists nearby.

El Cajete Lake

The fourth known lake in the Valles caldera, following the early lake and the San Antonio and South Mountain lakes, was the El Cajete lake. This was formed when the El Cajete and other East Fork Member eruptions filled the southwest moat and dammed the East Fork Jemez River. Lake deposits from this most recent lake are extensive in the Valle Grande, and sometimes contain El Cajete pumice.

El Cajete Lake also produced lake bars and terraces of loose gravel. One of these is well exposed in the road cut on the south flank of Cerro Pinon.

The terrace is composed of loose gravel with considerable sand and clay, containing occasional bits of El Cajete pumice.

Hydrothermal Activity

The Jemez volcanic field is far from dead. It has experienced volcanism for at least the last fourteen million years, and, as we've seen, the most recent volcanic activity may have occurred just 74,000 years ago.

There is evidence that there is still a magma chamber beneath the Valles caldera. One clue is that seismic waves are seen to slow down as they pass beneath the caldera. This low-velocity zone is interpreted as a magma chamber that is only partially crystallized. Seismic waves passing through the low-velocity zone also experience higher attenuation than in normal rock.

Another clue is the continuing high rate of heat flow from deep underground, manifest both in a high temperature gradient in holes drilled in and around the caldera, and in hydrothermal activity in the same general area. Hydrothermal activity in the modern Jemez mostly takes the form of hot springs, though there are a few fumaroles in the Sulfur Springs area.

The most accessible hydrothermal feature of the Jemez is Soda Dam. This is a large natural dam of travertine, calcite deposited by hot water from nearby hot springs.

There are numerous small hot springs in the area, but the largest are just across State Road 4 to the west. These originally drained into a natural plumbing system within the dam. The calcite-rich water then emerged from a long fracture running the length of the dam, and the calcite was deposited on the dam surface. Highway engineers demolished the west end of the dam to improve State Road 4 in 1960 and inadvertently destroyed its plumbing system; the dam is now slowly eroding away.

The green around the spring flow is partly algae, partly sulfur bacteria feeding on the sulfur compounds in the spring water. The smell of sulfur is noticeable throughout this area.

Here's the exposed calcite interior of the dam where it was partially demolished.

Soda Dam is geologically young, probably less than 5000 years old. Calcite from the spring waters has cemented together beds of eroded gneiss from nearby outcrops, which have now been cut by the road.

There are a number of older travertine deposits in this area. Those near river level on the opposite side of the Jemez River are probably 60 to 110 thousand years old, while those well above the level of Soda Dam in the nearby canyon walls are around 0.48 to 1.0 million years old. These show that the hydrothermal system here has been long-lived, and also provide evidence that Cañon de San Diego was already partially incised within 250,000 years of the Valles event.

The location of the springs is significant. This is the area where the Jemez Fault crosses Canyon de San Diego, and the fault zone provides a natural pathway for ground water.

Old travertine is also found along the San Jose Fault, which runs along the west side of Borrego Mesa.

This is one of several stacks of broken rock cemented together with travertine that are found in this area.

The identification of this rock as travertine is confirmed both by its flow features and by the acid test for calcite, which rules out a flow-banded rhyolite (the only other real possibility.)

The age of the travertine deposits along the face of Borrego Mesa is unknown. However, stacks like those shown earlier seem to postdate the landslides, making them less than 70,000 years old.

Not all travertine in the Jemez region is directly connected with Jemez volcanism. Travertine beds are found near a number of major faults throughout the area. For example, there are extensive travertine beds along the Pajarito Fault in the southwest Jemez and in the Tierra Amarilla anticline to its south. (This is not the Pajarito Fault Zone near Los Alamos; the duplicate names are unfortunate.)

The hill in the middle of the valley is capped with travertine deposited by fluids emerging from a nearby major fault.

Springs in the Jemez area fall into three categories. The most numerous are cold springs, which are recharged by precipitation that does not penetrate deeply enough to be affected by the deep magma chamber. These are similar to cold springs found in all but the most arid climates throughout the world. Their chief significance in the Jemez is that they tend to follow boundaries between volcanic flows, where the rock is more permeable, and thus can serve as clues when mapping flows.

The hottest springs in the Jemez area are found within the Valles caldera itself, such as at Sulphur Springs.

Sulfur Springs is one of the most geothermally active areas of the Jemez, and was once mined for sulfur and later developed as a hot springs resort. Accounts vary on its fate: Some report that the resort failed during the Great Depression, while others report that it burned down in the 1960s. Now all that is left are the surviving buildings of the resort, abandoned trailers, and other detroitus.

Here the groundwater has been superheated by the very hot rock beneath the caldera floor, and it is full of sulfur compounds that give it a very low pH. These acid-sulfate hot springs tend to rapidly alter rock in their vicinity to clays, iron oxides, and sinter (amorphous silica.) The area immediately around the springs is underlain by Deer Canyon Member tuffs, but these have been altered to light clay minerals by the hydrothermal activity in the area. The smell of sulfur is unmistakable.

Another area of warm springs is found at the mouth of Alamo Canyon. This is a rocky area with bubbles emerging from several places. In some cases, there was what appeared to be sulfur deposits around the springs.

It is hard to imagine that this is anything but sulfur deposited by the spring. However, at the time I visited the area, the water was icy cold. I tried several vents; only one was lukewarm rather than cold. I also found that the water had no noticeable taste. My visit was in May, and the hot springs may have been diluted by heavy spring runoff. However, one road log for the area reports that it is fairly normal for the spring water here to be cold.

Outside the caldera, one finds warm springs that are dominated by chloride rather than sulfur compounds and have a neutral to slightly alkaline pH. The springs in the Jemez Springs area fall into this category. The water from these springs is meteoric water (water from rain and snow that has not been long underground) heated by the hot rock under the surface.

The distinction between chloride and acid-sulfate springs is not a sharp one, and many springs show both chloride and acid chemistry. The springs around Soda Dam are moderately acid-sulfate, but neither as hot nor as low in pH as the springs at Sulfur Springs.

Sulfur and volcanoes

Sulfur is a moderately common element in the earth's crust, and it is particularly associated with volcanic activity.

Sulfur has a complex chemistry compared with elements like silicon, oxygen, or even iron with its two oxidation states. Sulfur has many oxidation states, ranging from donating six electrons in sulfates to accepting two electrons in some sulfides. Most magmas contain some sulfur, but because sulfur has a strong affinity for iron, the amount of sulfur in a magma is closely tied to its iron content. However, sulfur is highly incompatible with silicates in all but the sulfate state, where it forms rare sulphosilicate minerals. Because of this incompatibility, the sulfur in a magma will often separate into a distinct sulfide phase rich in iron and other chalcophile (sulfur-compatible) elements, and this phase can produce rich ore deposits.

One of the surest signs that fresh magma has risen beneath a volcano is increased emission of sulfur dioxide, SO₂. This has a characteristic sharp smell and dissolves in water to produce an acid solution of sulfurous acid. When further oxidized to SO₃, as is likely to occur at a slow rate when exposed to air, it dissolves in water to produce powerful sulfuric acid. Other gaseous sulfur compounds emitted by volcanoes include hydrogen sulfide and sulfur vapor. Hydrogen sulfide has a strong odor of rotten eggs, while sulfur vapor is typically deposited as solid sulfur crystals near fumaroles.

The most sulfur-rich magmas are alkaline basalt magmas, which can contain up to 5000 parts per million of sulfur. A value of 1000 parts per million is typical of more common basalts.

Hydrothermal alteration

We saw earlier that hydrothermal alteration is common in the lower beds of the Paliza Canyon Formation exposed in canyon bottoms and the base of the Valles topographic rim. Hydrothermal alteration is not easy to date, but this episode of alteration probably immediately followed the eruption of the Bearhead Rhyolite around 7 million years ago. A second period of hydrothermal alteration began with the Toledo and Valles eruptions and continues to the present day.

Here is a rather spectacularly colored rock in an area of the Valle Jaramillo/Redondo graben saddle mapped as Deer Canyon Member.

The vivid green color is due to epidote. The rock is still very solid, which rules out significant alteration to clay. There are visible quartz crystals, as well as transparent lath-like crystals whose identity I’m not mineralogist enough to identify. Hydrothermal alteration is pervasive in the Deer Canyon and Redondo Creek Formations.

Erosional landforms

The landforms we see in the Jemez today are usually eroded, sometimes heavily. Northern New Mexico has experienced regional uplift over the last several million years, with the result that erosion has accelerated and is now destroying rock beds that, in some cases, are nearly 1.8 billion years old.

Weathering and soil

Erosion begins with the weathering of solid rock, which in some cases produces a layer of soil over the bedrock. Soils show great variety, even in as small a region as the Jemez Mountains, and I can only touch on a few aspects of soil formation.

There is now fairly general agreement among geologists that chemical weathering is the most important process for destroying solid rock, with mechanical weathering playing a secondary role at best. Mechanical weathering includes such processes as cracking from repeated thermal expansion and contraction by day and night and wedging from water freezing in existing pores and cracks. Chemical weathering involves slow dissolution of certain mineral consitutents by rainwater, which is slightly acid, and groundwater, which varies considerably in pH. Rainwater typically dissolves sodium, potassium, and calcium out of minerals containing these elements, such as feldspars and micas. This converts these minerals to soft clay minerals. For example, the weathering of potassium feldspar can be represented by the reaction:

3 KAlSi₃O₈ + 2 H₂CO₃ + 12 H₂O ⇌ KAl
3Si₃O
10(OH)
2 + 6 H₄SiO₄ + 2 K⁺ + 2 HCO₃⁻

That is, potassium feldspar exposed to dilute carbonic acid in rainwater disintegrates into illite, a clay mineral; silica (written here as silicic acid, H₄SiO₄); and dissolved potassium bicarbonate. Potassium ion is highly soluble in water, and almost all the dissolved potassium eventually reaches the ocean, to be recycled in subduction zones. Silica is much less soluble, with a maximum concentration of about 0.02 parts per million in water, and in dry climates most of it will remain in the soil as quartz. Only in climates much wetter than New Mexico is the silica leached away to form nearly pure beds of clay, which in the wettest climates can undergo further weathering to yield almost pure aluminum hydroxide. This is the origin of bauxite, the principal aluminum ore.

Calcium-rich feldspar, common in basalt and andesite, weathers to smectite by a similar process to produce dissolved calcium ions and silica. This is the origin of caliche in dry climates, where the calcium is redeposited as a hard later of calcium carbonate.

Iron can also be oxidized from ferrous to ferric iron, causing the ferrous minerals to disintegrate. An example we mentioned earlier was iddingsite, produced from oxidation of olivine in basalt lava flows.

Desert soils

The lower elevations around the Jemez are semiarid and produce characteristic desert soils. These are poor in organic matter and are alkaline, since the scanty rainfall is not efficient at leaching weathering products such as sodium, potassium, or calcium. Calcium, in particular, tends to accumulate in the subsurface as caliche. We saw examples of this in Chapter 6, where the older basalts of the Lobato Formation are often encrusted with caliche. Geologists estimate that caliche takes at least a century to form, which is useful for estimating the age of young surfaces and for estimating time spans of paleosols in older sedimentary beds.

Another characteristic of desert soil is desert pavement. This is well developed at some locations on the rim of White Rock Canyon.

The numerous exposed basalt clasts are an example of a lag deposit. The remaining soil around the larger clasts is often cemented, sometimes by hardy cyanobacteria that colonize the soil. This is fairly delicate terrain, and I'm sorry to say that the popularity of the area for hiking pretty much guarantees this desert pavement will eventually be pounded into oblivion. But I also can't bring myself to tell people to stay away; this is beauty I can't resist sharing. I'm proud of my home.

This is Orthoporus ornatus, the desert millipede. Though uncommon, tarantulas are also occasionally encountered in the White Rock area, along with many smaller species of spider.

This reminds me of an experience I had while out walking at twilight to the local grocery store. It was a pleasant warm evening, with the light fading fast. I had a head lamp lit so I could see my way.

I am accustomed to the occasional glint of light from the ground, from some crystal face oriented just right. But as I walked along, I saw a pair of greenish glints that rapidly flashed at me as I walked past. Curious, I moved closer, and saw that it was two spiders. Their compound eyes had given multiple green reflections as I moved by.

As I continued my walk, I spotted many more. The spiders seemed to be mostly of a single species, not large (the biggest had a leg span of perhaps an inch), brown striped, and clearly some kind of hunting spider. Of course, I was also a crepuscular hunter in search of prey, though my prey was an unwary gallon of milk with which to make my breakfast. Still.

The spiders were beautiful. It was the kind of magic moment I wish my wife or children had been along for.

Beyond the shadow of the ship,
I watched the water-snakes:
They moved in tracks of shining white
And when they reared, the elfish light
Fell off in hoary flakes.

Within the shadow of the ship
I watched their rich attire:
Blue, glossy green, and velvet black,
Then coiled and swam; and every track
Was a flash of golden fire.

O happy living things! no tongue
Their beauty might declare:
A spring of love gushed from my heart,
And I blessed them unaware:
Sure my kind saint took pity on me,
And I blessed them unaware.

The self-same moment I could pray;
And from my neck so free
The Albatross fell off, and sank
Like lead into the sea.

— Samuel Taylor Coleridge, Rime of the Ancient Mariner

Fire

A significant agent of erosion in the Southwest is wildfire.

Las Conchas fire of 2011. Looking northwest from near 35 49.342N 106 13.467W

Under natural conditions, the Jemez Mountains are subject to frequent, but usually low-intensity, wildfires. Geologists have taken borings from older trees and stumps in the Jemez and matched burn scars with tree rings to find the dates of major fires. It appears that, prior to about 1890, widespread fires took place every five to fifteen years. These were mostly surface fires, which burned off grass, shrubs, and tree seedlings but rarely destroyed more mature trees. The frequency of fires dropped precipitously in the final years of the 19th century, probably due to livestock grazing that removed grassy fuels.

Human efforts since 1900 to reduce the occurrence of fires have likely had the perverse effect of making fires less frequent, but more destructive. Without the frequent burning out of ground debris and younger trees, fuel builds up until devastating crown fires destroy the forest ecosystem. The trend towards a warmer and drier climate accentuates the hazard of wildfire. Recent destructive fires near the Pajarito Plateau include the La Mesa fire of 1977 (15,444 acres), the 1996 Dome fire (16,516 acres), the 1998 Oso Complex fire (5,185 acres), the 2000 Cerro Grande fire (48,000 acres), and the 2011 Las Conchas fire (150,000 acres).

But even before the arrival of humans in New Mexico, wildfires contributed significantly to erosion. The more intense fires are capable of speeding the weathering of some kinds of rock.

Fire-scorched boulder on Las Conchas. 35 47.838N 106 31.687W

This boulder is in an area that was burned over by the 2011 Las Conchas fire, a particularly hot and destructive fire, and the rock show a thin layer of soot. There is also some indication of exfoliation, the shedding of an outer layer of rock, from the heat of the fire. It seems likely that this is a significant cause of weathering over geological time scales in an area where natural wildfires are a regular occurrence.

Fire also increases erosion by stripping the ground of vegetative cover. This allows raindrops to hit bare soil and rainfall to run downhill without the moderating influence of vegetation. This in turn leads to flash floods, in which a summer monsoon thunderstorm dropping heavy rain in a brief period can produce sudden and violent flooding of channels further down the watershed. These can cause spectacular erosion in geologically insignificant intervals of time.

Colluvium

Weathering is followed by erosion, in which weathering products are moved to new locations.

Digital relief map of colluvium in the Jemez
Mountains
Relief map of the Jemez with colluvium deposits highlighted in red

The earliest stage of transport of eroded rock in mountainous areas is the formation of colluvium at the base of cliffs. In the Jemez area, this is particularly obvious around mesas of the Tsherige Member, Bandelier Tuff, which typically are surrounded by talus slopes cut into the underlying Otowi Member that are covered with clasts of Tsherige Member.

Colluvium at the base of a mesa of Tsherige Member, Bandelier Tuff. 35.801N 106.224W

Scree at base of Tsankawi Mesa. 35 48.055N 106 13.433W

Colluvium is unconsolidated fragments of rock and soil that accumulate at the base of a slope from relatively gentle processes, such as rainwater flowing over the surface. Colluvium does not include landslide deposits, which are not a gentle process, nor alluvial fans produced by the concentrated action of water. Colluvium is sometimes distinguished from scree, which is rock shards that have accumulated at the base of a cliff from which they obviously broken off, but most geologic maps of the Jemez area have generally mapped it all as colluvium.

The Otowi Member is often difficult to distinguish from a colluvium-mantled talus slope at a distance, since it is usually present only under cliffs of more durable Tshirege Member and it tends to weather to a gentle slope. For example, as one hikes down the trail on the south side of Pueblo Canyon onto the talus slope, one sees that it is indeed deeply mantled with soil.

Colluvium on south wall of Pueblo Canyon. 35 52.890N 106 15.945W

Further complicating the picture is the fact that, because colluvium is composed of weathered fragments of nearby rocks, the two can sometimes be hard to tell apart.

Colluvium formed from Otowi Member in San Antonio Canyon. 35 56.572N 106 38.791W

The colluvium here looks like a tuff, but it’s very poorly consolidated. The geologic map maps this area as colluvium, but shows a bank of Otowi Member just up the hill. So this is reworked tuff, formed from sediments eroded off the original tuff bed. It is the historical use of tuff to refer both to primary pyroclastic deposits and to reworked volcanic ash that led to adoption of the term ignimbrite to describe primary pyroclastic deposits. However, the International Union of Geological Sciences discourages the continued use of this term, preferring the term ash flow tuff for primary pyroclastic deposts composed mostly of ash.

Some of the most ambiguous beds are found in the area north of Los Alamos, where beds mapped as Cerro Toledo sediments in the Guaje Mountain quadrangle are mapped as colluvium in the adjoining Puye quadrangle. This accounts for the "boundary fault" in the map of colluvium deposts east of Los Alamos. The New Mexico Bureau of Geology & Mineral Resources at Socorro is presently making a major effort to produce a set of quadrangle maps for the entire Jemez region that are consistent across map boundaries.

Landslides

Relief map of the Jemez with toreva blocks of White Rock Canyon highlighted in green and other landslides highlighted in red.

Not all rock debris found at the feet of high terrain got there by gentle processes. The geological picture in many parts of the Jemez Mountains is of very hard, dense lava flows lying on top of soft sediments. This is a perfect recipe for landslides.

Toreva blocks in White Rock Canyon

Some of the most spectacular landslides in the Jemez area are in White Rock Canyon, where Cerros del Rio basalt overlies poorly cemented sediments of the Tesuque and Chamita Formations of the Santa Fe Group. Because the basalt is so durable, the mass wasting in White Rock Canyon consists of what geologists call Toreva blocks. These are named after the location they were first recognized (Toreva, Arizona.) An entire block of the canyon rim comes loose as a unit and slides partway down the canyon wall, more or less intact, so that the original bedding is still present in the block. A particular characteristic of these blocks is that the block rotates slightly towards the surface from which it detaches.

The Toreva blocks in White Rock Canyon include some of the best examples known to geologists.

From the west canyon rim, south of Overlook Park, one has a beautiful view of one of the most impressive Toreva blocks to be found below the canyon rim. The Google satellite image doesn't do it justice; try switching to Map Terrain view, although this also doesn't quite do it justice. Fortunately, it's highly photogenic.

Toreva block on west side of White Rock Canyon Looking southwest from. 35 48.986N 106 11.422W

The broad meadow is a slice of basalt canyon rim that has broken loose and slumped partway down the canyon wall. You can see that the far end of the meadow is nearly at the level of the canyon rim, and the nearer end has slumped a greater distance, almost as if the block was hinged at the far end. You can see that there is a trail down the slump block that affords some nice views of cross sections of the basalt flows making up the canyon rim.

The detachment of the Toreva block has exposed as least two flow units here. These cliffs are a favorite location for local rappellers.

The Toreva block gives us a nice view from the canyon rim high above of hexagonal fracturing of the lava flow. We saw this picture earlier, in the chapter on the Cerros del Rio.

Not all coherent detachment produces blocks rotated towards the cliffs. Relatively shallow slumping of otherwise coherent blocks can leave the blocks rotated outwards rather than inwards. Some fine examples are seen just north of the Jemez region, at the north end of Black Mesa.

These blocks are rotated slightly outwards rather than inwards, indicating coherent but shallow slumping that is not of Toreva type. Such slumping tends not to preserve the lowermost part of the block as well.

Because of their nature, landslides are difficult to date. Those in White Rock Canyon are less than 1.6 million years old, judging from the way that deposits of Guaje Pumice on the rim match deposits on the Toreva blocks. However, this is only a upper limit on the age, which is probably much younger.

Borrego Mesa

There are large landslides on the west side of Borrego Mesa. Here the hard cap is basalt of the Paliza Canyon Formation and the underlying poorly consolidated sediments are mostly Zia Formation of the Santa Fe Group. The landslides here do not take the form of toreva blocks, but more closely resemble a rock avalanche, with the bedding thoroughly destroyed.

Looking across the valley, one sees smaller but similar landslides of Bandelier Tuff on Chinle Group sediments.

Most of the landslides appear to be at least 74,000 years old, the age of the El Cajete Pumice. This is because beds of the pumice are found in many locations on top of the landslides.

Not all the landslides are older than the El Cajete Pumice; a few small slides in this area appear to lie on top of beds of the El Cajete Pumice.

This is a rockslide deposit below a prominent knob of andesite, of the same composition as the boulders here.

San Miguel Mountains

The northern flank of the St. Peter's Dome volcanic center collapsed sometime between its emplacement 8.7 million years ago and the eruption of the El Cajete Pumice 73,000 years ago, based on the presence of El Cajete deposits on top of the landslide. Here the landslide blocks were composed of andesitic lavas atop volcaniclastics that were more strongly cemented than the Santa Fe Group beds elsewhere. One block survived in more or less coherent form.

This is interpreted as a landslide block that was so highly rotated that the lava beds are left are now oriented vertically.

As with other landslides of the Jemez region, time control is poor. However, field relationships in Capulin Canyon, at the base of the landslide, hint that this landslide postdates the Valles event, and so took place between 1.25 million and 73,000 years ago.

Vallecitos de los Indios

The northern slopes of Vallecitos de los Indios are covered with landslide deposits formed from Banco Bonito obsidian flows atop mudstone of the Abo Formation. These are relatively accessible, being crossed by the hiking trail to McCauley Springs. The size of some of the clasts in the landslide is impressive.

Landslide deposits along the McCauley Springs trail. Somewhere near 35.824N 106.638W

The largest bounders here are devitrified obsidian boulders of the Banco Bonito flow. The reddish soil shows its origins in the mudstones of the Abo Formation.

Alluvial fans

Over time, colluvium is washed further downhill by streams and floods to produce alluvial fans. These are found along much of the inner rim of the Valles caldera, as well as other locations of high relief.

The ideal alluvial fan has a distinctive shallow cone shape, with its tip at a canyon emerging from an escarpment. Here periodic floods in the canyon emerge, spread out, and deposit their sediments on the surface of the fan. Alluvial fans can be found in almost any climate, but are most recognizable in arid climates, and there are some spectacular alluvial fans in Death Valley and other parts of the Basin and Range.

Surprisingly, tthe Jemez region lacks well-developed alluvial fans, even at lower and drier elevations along escarpments. This may be yet another effect of regional uplift, increasing erosion rates to the point where fans do not accumulate. More typical of this area are bajadas, which are coalesced alluvial fans along an escarpment. These lack the distinctive cone shape, resembling more of an apron of sediment around the high ground. They are distinct from pediment surfaces in that they are composed of deposits of sediments rather than being eroded out of bedrock. They differ from colluvium in that concentrated flow, not rainwash or gravity, is the main mechanism of transport.

One can examine an alluvial fan on the Valles Caldera Trail, part of the Valles National Preserve. The trail descends the inner rim of the caldera on its southeast to the caldera floor, where the alluvial fan includes local debris flow deposits.

Individual boulders are composed of the Tschicoma Formation dacite of the caldera rim.

Debris flows continue to be a part of the modern Jemez landscape. They are particularly active in areas that have suffered recent fire damage, such as the area south of Redondo Peak (which was burned during the 2011 Las Conchas fire).

A river runs through it

Herrick erred in thinking the Bandelier Tuff was older than the Cerros del Rio Basalt, but he was right in his observation that flows of the Cerros del Rio Basalt repeatedly blocked the Rio Grande in White Rock Canyon.

The presence of the Totavi Lentil shows that the Rio Grande already existed 5 million years ago, when the Puye Formation began to be laid down. The history of this river thus overlaps most of the recent history of the Jemez volcanic field itself. Rather than trying to "interbed" that story with the story of the Jemez Mountains, I'm going to step back to 5 million years ago and tell that story now as a coherent narrative of its own.

The location of the Lentil, well west of the modern Rio Grande and at a higher elevation, suggests that the Rio Grande originally ran further to the west and was slowly forced eastward by accumulating sediments of the Puye Formation. The river was at least 225m (738') above its current level when it first appears in the geologic record. The river was not yet fully integrated along its entire modern valley. The full integration of the Rio Chama and Rio Grande through the Espanola Basin is thought to have occurred sometime between 2.8 and 4.0 million years ago.

Once it had eroded through the basalt cap, the Rio Grande in White Rock Canyon rapidly cut deeply into the underlying soft sediments of the Santa Fe Group, The river reached its maximum depth 2.8 million years ago, at a level about 30 m (100') below the modern level. It left behind a series of terraces to the west, which are now deeply buried under the Pajarito Plateau and are known only from test wells. These divide Santa Fe sediments beneath from Puye Formation sediments above. The maximum depth is recorded by a flow of olivine basalt very close to the modern river level near Ancho Canyon, which has a radiometric age of 2.47 million years.

The relatively young age of this flow shows that it must have followed a deep canyon in the surrounding, much older, Santa Fe Group beds and early Cerros del Rio beds. This flow marked the beginning of a series of natural dams, formed by volcanic activity, that forced the river level back up in upper White Rock Canyon.

The highest of these dams was a flow of tholeiitic lava at the current location of Water Canyon that erupted around 2.46 million years ago. This drove the level of the river back up to 285m (935') above its current level. The resulting lake, Culebra Lake, covered much of the Espanola Basin. It is now believed that much of the preservation of the Puye Formation can be attributed to Culebra Lake, which raised the base level for erosion of the Puye sediments.

Digital relief map of approximate extent of Culebra
Lake

Relief map of the Jemez with approximate extent of Culebra Lake highlighted in blue.

Lacustrine deposits are highly vulnerable to erosion, and few traces remain in the Espanola Valley. One of the more dramatic surviving exposures of lacustrine sediments is the diatomite mine northwest of Espanola visible from Clara Peak.

The mine is operated by J.H. Rhoades Pumice Company. Diatomite, also known as diatomaceous earth or as kieselguhr, is a sediment consisting mostly of fossilized remains of diatoms. Diatomite makes an excellent abrasive and is also used as a filtering medium.

Diatoms are single-celled algae that form silica shells. They first appeared in the fossil record around 200 million years ago, and may have evolved to fill ecological niches opened by the Permian-Triassic extinction. They subsequently became a major component of plankton and may account for 20% of the Earth's modern oxygen production. Diatoms flourish in both fresh and salt water, and the diatoms here are freshwater diatoms, possibly from Lake Culebra.

The Rio Grande had cut back down to within 130m (430') of its current level at the time of the Toledo event, 1.6 million years ago. Exposures of the Otowi Member are present on the east side of the river north of Cochiti, but the resulting dam must have been short-lived, and the river rapidly cut down to its former channel. The failure of this natural dam may account for deposits of Otowi Member-rich clasts far downriver, at Socorro and Las Cruces. Another short-lived dam was formed shortly afterwards by an eruption of basaltic andesite that overlies the Otowi Member. There is a shallow paleocanyon east of the river in this area that shows the river course was temporarily diverted.

The river was down to 120m (390') above its current level when the Valles event took place, 1.25 million years ago. At this time, the canyon was much narrower than it is today, with a maximum width of perhaps 600m (2000') and no evidence of the large slump blocks seen today.

The Tsherige Member completely filled White Rock Canyon between Chaquehi and Frijoles Canyon, and filled an eastern meander of the canyon further north.

Tsherige Member filling eastern meander in paleochannel of Rio Grande. Looking east from 35 47.367N 106 12.630W.

Here the beds of the Bandelier Tuff overlie a thin layer of river gravel, which in turn lies atop Cerros del Rio basalts. From here the paleocanyon turned sharply west, roughly along the modern Water Canyon.

Cerros del Rio basalts in the confluence of Water Canyon and Potrillo Canyon. 35 47.445N 106 12.712W

At far left and far right are the basalt cliffs that mark the sudden drop from the relatively shallow Potrillo Canyon into the much deeper Water Canyon. Water Canyon descends from the Pajarito Plateau at center right, and continues to its confluence with White Rock Canyon at left. Here part of the east rim of White Rock Canyon is visible, with Montoso Peak on the skyline.

The large outcropping of light pink Bandelier Tuff in the south wall of Water Canyon marks the westward meander of the paleocanyon. This outcropping has not slumped down the canyon wall; it was this thick when deposited 1.25 million years ago. Since older Cerros del Rio basalt and underlying Santa Fe Group sediments form the rest of the canyon wall, this shows that the Bandelier Tuff filled a deep westward meander of the river.

This meander also left an outcrop of Bandelier Tuff on the north wall of the canyon.

Banderlier Tuff filling meander in White Rock paleocanyon exposed on north wall of Water Canyon. 35 47.578N 106 13.840W

The paleocanyon continues south along an unnamed tributary to Ancho Canyon, where the modern canyon runs almost parallel to the paleocanyon and exposes a great length of thick Tsherige Member beds.

Banderlier Tuff filling meander in White Rock paleocanyon in unnamed tributary to Ancho Canyon. 35.783458N 106.225969W

Contact
between Cerros del Rio and Bandelier Tuff in lower Frijoles
Canyon

Contact between Cerros del Rio and Bandelier Formations in lower Frijoles Canyon. 35 45.888N 106 15.661W

The damming of the Rio Grande near Frijoles Canyon raised the river level to the highest since its beginnings in the Miocene, at 325m (1070') above its current level. The Rio Grande was forced 2 km (1.25 miles) east of its former course, and while it likely rapidly cut through the tuff, it then encountered 200m (660') of solid basalt that would have taken significantly longer to cut through. Thus a second great Culebra Lake was formed that extended perhaps 70 km (45 miles) to the north.This likely formed the lake bars on the White Rock Canyon rim near Overlook Park.

There are scattered beds of similar gravel for at least a mile further down the canyon rim, and there is also a considerable quantity of this gravel on a landslide block east of this point, halfway down the canyon rim. The gravels beds appear to overlie remnants of the Guaje Pumice and thus must be younger than 1.62 million years in age. The dam produced by the Tsherige Member is estimated to have been 100m (180') higher than the canyon rim where these cobbles are located.

The river was still probably around 150m (490') above its modern level 0.62 million years ago. This is based on the location of ash beds from the great Lava Creek eruption in Yellowstone at that date, which spread ash across the United States. The rate of erosion has been particular great in the last 100,000 years or so, at about 50 cm (20") per thousand years. White Rock Canyon is now about 250m (820') deep near the White Rock Overlook. The rate of incision probably increased significantly sometime between 0.7 and 0.3 million years ago, when the San Luis Basin of southern Colorado became integrated into the Rio Grande watershed, more that doubling the watershed area above White Rock Canyon. This was also about the time the upper and lower reaches of the Rio Grande finally became integrated in the El Paso area. (Prior to then the Rio Grande ended in a series of lakes in northern Mexico.) The final integration of the Rio Grande dramatically dropped the base level of the upper Rio Grande, which had continued to slowly deposit sediments in the northern Rio Grande Rift up to that time. Thereafter rapid erosion set in, producing most of the badlands seen along the northern Rio Grande today.

There have been several smaller lakes over the last 70,000 years, mostly created by a large slump near Water Canyon that has repeatedly slid into the river channel during periods of unusually wet climate.The last such lake is thought to have formed about 12,400 years ago, close to the start of the Holocene.

Zeolite ledges

There are a number of prominent ledges eroded into the Bandelier Tuff in Ancho Canyon.

The ledges do not correspond to any flow boundaries or other bedding features in the tuff, and they can be traced into other canyons along the Pajarito Plateau. They are believed to show lake levels in nearby White Rock Canyon. The lake water would have saturated the tuff up to the lake level, and the few inches just above the water level would have produced an ideal environment for zeolites to form in the tuff. These are hydrous silicate minerals that fill pore spaces and increase the durability of the tuff. Radiometric analysis of the zeolite crystals gives an age of about 1 million years for the ledges, but this age is rather uncertain.

The Sierra Ladrones Formation

Digital relief map of old river gravel exposures in the
Jemez Mountains

Relief map of the Jemez with Sierra Ladrones Formation outcroppings highlighted in red

Deposits interpreted as old river gravel are present throughout the Cochiti area. These river gravels cannot be dated directly, but some beds are overlain further north by formations known to be about 2.5 million years old, showing that parts of the old river gravel must be older than this. The older portions of these gravel beds were mapped as "old gravels" in the classic Smith and Bailey map of the Jemez Mountains, but they have since been assigned to the Sierra Ladrones Formation of the upper Santa Fe Group.

Excellent outcrops of the Sierra Ladrones Formation can be found south of the small town of Pena Blanca. This photograph shows the sandstone facies of the formation.

Here a fault has thrown up a hill of the underlying axial gravel facies of the formation.

The San Francisco Fault passes through the notch at left. The terrain on the near side of the notch is more of the sandstone facies, while the hill on the other side is the axial gravel facies. Axial gravel is laid down in the central channel of a river, in this case the ancestral Rio Grande. These beds are difficult to date precisely but cannot be older than about 6.5 million years or much younger than 1.6 million years, based on the ages of the individual rock fragments deposited as gravel in the formation and its relation with volcanic formations in the area.

The nicely bedded red sediments are the piedmont facies of the Sierra Ladrones Formation. This is sediment brought in by the Santa Fe River and Galisteo Creek from the east. East of this location, this facies contains abundant fossils of early horses (Equus) and ancient relatives of elephants (Gomphotherium).

The gravel on top is relatively young alluvial gravel. The lower portions of the gravel contain volcanic ash that has been traced to the Lava Creek eruption of the Yellowstone caldera, 630,000 years ago, and presumably this gravel is about that age.

The gravel itself contains a mix of quartzite clasts, probably Precambrian quartzite from the Tusas Mountains to the north, and basalt clasts, likely from the Cerros del Rio southeast of White Rock. This means this gravel was brought in by the Rio Grande.

The Sierra Ladrones Formation continues far to the south; indeed, its type section is in the Socorro area in southern New Mexico. It is well exposed along I-25 just south of San Felipe Pueblo, at the southern limit of the Jemez region.

Here we see light tan silt beds atop a thick bed of coarse gravel. The gravel is axial gravel of the ancestral Rio Grande while the silt is likely overbank deposits, deposited after the main channel had shifted elsewhere when the river overlowed its banks onto its floodplain.

Terrace gravels

The ultimate fate of material eroded from the Jemez Mountains is to be carried down the Rio Grande River to the Gulf of Mexico. However, only fine clay is likely to make the trip directly, suspended in the water. Sand and small pebbles roll and bounce along the river bottom, a mode of transport called saltation. Larger clasts work their way down river channels in several steps, likely driven by infrequent but powerful flood stages of the river.

The ability of a river to carry sediment depends on how swift the river is. Turbulent motions in rapidly moving water keep the smallest particles in suspension. Swift flow at the river bottom keeps sand and pebbles rolling and bouncing. When the river changes speed, the carrying capacity also changes. Thus, when a river slows, it tends to drop sediments, a process called aggradation. Since the swiftness of a river is largely determined by how level its bed is, a river descending a steep slope quickly cuts down (incises) a deep channel, while a river that is crossing nearly level ground reaches an equilibrium in which aggradation balances erosion.

For an ancient river emptying into the sea, such as the Mississippi, the effect is that the river cuts a very broad flood plain of very low relief and elevation. The flood plain is typically underlain by very fine overbank sediments while the river channel has coarser sand and pebbles. Since sediment tends to be deposited on the inside of curves in the river, and eroded from the outside of curves, curves tend to grow and the river becomes a meandering river, with a single channel showing intricate loops. At the other extreme, very young rivers flowing across steep ground tend to form several channels that split and reconnect to form a braided river.

Curiously enough, some of the best examples of meandering streams in the Jemez area are within the caldera itself, where the flat caldera floor provides low relief for the East Fork Jemez River and San Antonio Creek, which strongly meander. The Rio Grande in the Espanola Basin is young enough, geologically speaking, that it is intermediate between a braided river and a meandering river, showing incipient meandering in some stretches and braided features in others.

A river in an area of recent uplift, such as the Rio Grande in New Mexico, may have relatively level stretches ending in knickpoints, where the river loses elevation relatively rapidly as it crosses resistant beds or other features than hinder rapid downcutting. The knickpoint is said to establish the base level for the river upstream, which becomes the level of a floodplain around the river if the knickpoint lasts long enough. When the knickpoint is cut through, or if some other process causes the river to begin cutting downwards again, it is said to have abandoned its former floodplain, and the remnants of that floodplain may remain as terraces.

Northern New Mexico has experienced regional uplift for at least the last several million years, and the larger rivers have experienced repeated drops in their base levels. This has produced multiple terrace levels along major rivers, such as the Rio Grande. These terraces are often covered with terrace gravels, deposited by the main channel of the river as it cut a flat surface onto the underlying bedrock.

The youngest terraces are obvious geomorphic features. Geomorphology is the study of how the modern topography of the land is dictated by the underlying geology. A volcanic cone is an example of an geomorphic feature; the cone-shaped mountain is a direct result of a specific geologic process. Hogback ridges are another example, formed by erosion that levels nearby ground but leaves a set of resistant rock beds as a ridge. A young terrace is distinctive flat ground away from the main flood plain of a river. Here is an example.

The low ground is the current channel and associated (small) floodplain of a small tributary of Galisteo Creek. The flat ground on the near and far sides are young river terraces, showing where a sudden drop in the base level for Galisteo Creek caused the river to abandon its previous floodplain and cut downwards to establish the current floodplain. This is the youngest of six such terrace levels in the Galisteo Creek drainage, which record repeated drops in base level.

Older terraces are eroded to where they no longer are an obvious geomorphic feature. They are identified instead from the pattern of coarse gravel resting on a flat surface cut in older bedrock near a river. There are good examples of contacts between terrace gravels and the underlying Abiquiu Formation west of Abiquiu. Here the terrace gravels are derived from red source rocks, making a stark contrast with the white Abiquiu Formation.

The terrace gravel is the red bed about halfway up the hill side. The white beds beneath and in back are Abiquiu Formation. The terrace gravel thus is perched on the side of the hill, at what was once the level of a side channel of the ancestral Rio Chama.

Terrace gravels associated with the ancestral Rio Chama extend further south, and are well-exposed along Highway 84 between Abiquiu and Medanales. Here the gravels form low hills west of the highway.

Terrace gravels are found in the area south of Black Mesa, with a particular well-defined contact with the Chamita Formation.

This is the younger of two terrace levels identified along the Rio Grande in this area, with an age of around 140,000 years.

The terrace gravels are also well-exposed along Thirtyone Mile Road west of Espanola.

The large, well-rounded clasts show that this deposit is an axial deposit, representing the major river channel in this area. The geologic map for this area identifies this as terrace deposits of the Rio Chama-Rio Grande confluence, with an estimated age of 350-650 thousand years. The age is based in part on the presence of Lava Creek B ash beds near the base of the deposit. The deposit in the second photograph has been heavily calichified.

There are beds of terrace gravel along the main highway to Los Alamos near the point where it crosses the Rio Grande.

These can be difficult to distinguish from the Totavi Lentil of the Puye Formation, but their location close to the river, without any overlying beds, hints at their true nature. The Totavi Lentil has, in fact, been interpreted as terrace gravels from successive stands of the Rio Grande as it cut downward and shifted eastward due to the growth of the Jemez volcanic field. These were subsequently buried by fanglomerates of the Puye Formation.

There are some particularly fine examples of terrace gravels near Cochiti Dam. One example is so well sorted, with a very distinct upper contact, that my first thought was that this was artificial, placed by highway workers in the road cuts for erosion control or some other purpose. It was not until I saw the beds exposed in arroyos well away from the road that I was convinced they were a natural geologic feature.

This is one of the younger terrace gravels in this area, probably less than 100,000 years old. It is at the fourth of five distinct terrace levels that have been mapped in the area. The silty, tan beds above the gravel are mapped as eolium, which is silt and fine sand brought in largely by the wind.

Similar old gravels are found in the Abiquiu area and probably represent gravel deposited by the ancestral Rio Chama.

This is apparently an old axial gravel of the Rio Chama, deposited on Chama-El Rito Member, Tesuque Formation. The flat ground above has ruins of an Ancestral Puebloan city, Poshuinge.

Some chapters back, we saw that there is a distinctive bed of gravel between the Tesuque Formation and overlying terrace gravels throughout the area west and north of Espanola. This marker bed is particularly well exposed at Arroyo Largo.

This bed is very distinctive, consisting of well rounded clasts of quartzite typical of the Tusas Mountains to the north. It mark a geologically brief interval sometime in the last two million years when the ancestral Rio Grande and its tributaries in the Espanola area carried a heavy load of gravel from the mountains to the north. One wonders if it correlates with the beds we've just seen at Abiquiu and Cochiti and, if so, what event it records.

Alluvium

Finally, we reach the modern flood plains, where alluvium is being transported today or has been transported in the recent past. River channels and their floodplains are among the more obvious and recognizable of land forms. Alluvium often fills canyon and arroyo bottoms even where there is no permanent stream.

Arroyos such as this one can become very wet indeed during heavy rain. With their extremely flat sandy bottoms, they’re an important land form in the Jemez area.

Many arroyos in the Jemez Area are characterized by very steep banks, subject to occasional rockfalls.

Here the occasional flash flood has undercut the banks, leaving them unstable against collapse.

Some of the most impressive alluvium deposits are associated with the valles of the Valles caldera. These are broad valleys winding among the ring domes that are devoid of timber but support lush grass. We've seen many photographs of these areas in previous chapters.

The lack of timber arises from two causes. First, the valles are floored with thick beds of clay in which trees have considerable difficulty putting down roots. Second, the winter climate in the valles can be brutal.

This is a photograph of Valle Grande, the largest of the valles, on an October morning. The layer of fog is not unusual. The caldera accumulates cold air, producing a temperature inversion that often leads to the formation of fog or a low cloud layer. During the winter months, when there can be long stretches of dry and sunny weather, the snow cover evaporates off the valles floors and leaves tree seedlings exposed to bitter night cold. This is more than they can take.

Alluvium is also found along river channels, but here there is a delicate balance between deposition and erosion. Where fire has stripped cover and increased the power of water, erosion can predominate, as in the areas burned by the 2011 Las Conchas fire. Here the channel is being rapidly cut downwards to produce extensive gullies.

They’re real, and they’re spectacular. This section was at least ten feet deep, and it cut right across the old jeep road I was following. I had to work upstream a significant distance to find a spot where I could cross.

These channels represent concentrated flow. If the watershed of such a channel was a square kilometer, and erosion in the watershed averaged just a centimeter per thousand years, this would correspond to a cube of debris more than two meters on a side moving down the channel each year. Most likely the rate is much lower in ordinary years and much higher in years following a fire or other disturbance. There is evidence that the canyon bottoms of the Pajarito Plateau experienced a burst of alluvium deposition between 12,000 and 8000 years ago, based on carbon dating of charcoal fragments. A borehole in Ancho Canyon found a floodplain deposit 10,000 years old five meters (16.5') below the present canyon floor.

Sheetwash

Alluvium tends to be concentrated in river channels, but, in arid climates, areas of low relief may become covered with sheetwash.

Sheetwash is deposited where flood waters run across a large surface without being channeled. It is distinct from a floodplain, formed by regular seasonal overflow of a river from its channel. Sheetwash deposits occur where there is no nearby river channel.

Our story has now reached the point where it becomes our story. In the next chapter, Homo sapiens arrives on the scene.