How Glaciers Move Mechanisms of Glacial Flow The book's contention that glaciers always flow downhill is not exactly true. Of course, most of the time it is true but, like any fluid, ice flows in the direction of least resistance. Because ice is a viscous fluid, if it flows into a hollow, it will fill the hollow. At that point, the area of least resistance is above the hollow so pressure from the rear will push the ice up over the front lip of the hollow. If a hill is in front of an advancing glacier, the ice will flow around it first but eventually the path of least resistance will be in the uphill direction and ice will bury the hill. There are two primary mechanisms for glacial flow:

Plastic Flow Plastic flow is the result of tiny amounts of displacement between ice crystals. The sum of the individual motions between billions of ice crystals results in glacial movement. This is the dominant form of movement where all parts of the glacier are below freezing, including the base.

Basal Slip Glaciers also slide along their bases, driven by the tug of gravity and lubricated by meltwater along the base. Several possible sources for the meltwater exist. Ice under pressure melts and quickly recrystallizes along the rock-ice interface at the base. This is analogous to a skater's blade putting pressure on the ice. The ice melts and lubricates the glide but quickly refreezes once the pressure is removed. Meltwater from the upper surface of the glacier percolates downward through cracks and crevasses in the ice. Geothermal heating beneath some Icelandic glaciers provides an adequate supply of meltwater.

Thrust Faulting As the glacier slides along its base, local resistance causes the formation of thrust faults through the ice, particularly near the glacial front. Rock and soil debris from the base is carried up higher into the glacier along these faults.

	An iceberg calves off of the Perito Moreno glacier as it floats out onto Lago Argentino. (Photo by Jim Reynolds)		The brown lines running through the Perito Moreno glacier are thrust faults. Note the calving block of ice in free-fall. (Photo by Jim Reynolds)

Flow Patterns and Speeds Most glaciers move at rates measured in tens of meters per year. A few glaciers,however, move meters per day. Argentina's Perito Moreno glacier, seen in many of these photos was one of the world's premiere surging glaciers. The cause of the surge is probably the steep slope over which the ice descends from the Patagonian Ice Sheet. Until about 1990, the ice front advanced about 5 m/day. Because it floats out onto Lago Argentino, most of the advance calved off to form numerous icebergs. Eventually, the glacier crossed the narrow channel between the two major arms of the lake, forming an ice dam and, preventing circulation between the two sides. The lake's outlet into the Río Santa Cruz flows off of the north arm. The ice dam caused the water level on the south arm to rise about 20 m. Eventually, the south arm water became deep enough to refloat the ice. Once the ice rose off of the bottom, circulation between the two sides was reestablished causing a distinctive type of jokulhlaup as an enormous volume of water to gush under the ice into the north arm. The torrent caused most of the ice dam to break up and the glacier rapidly retreated back to the west shore of the lake. This cycle repeated itself about every three years for most of the 20th Century but then the glacier slowed down dramatically. The last event of the 20th Century occurred in 1989, a few months before I first visited the site. I visited the glacier again in 2001 and found that the ice was almost across the lake again. In March 2004, the first jokulhlaup of the new millennium broke through with renewed glory.

The 60 m high Perito Moreno glacier was just encroaching on the eastern shore of the channel between the two arms of Lago Argentino in December 2001. (Photo by Jim Reynolds)

Flow in Valley Glaciers Like a river, the rate of movement within a glacier is variable. The slowest motion is where the ice meets the most resistance: at the base and against the valley walls. Continuing with the analogy of a river, the fastest motion would be expected on the surface along the centerline of the glacier. This can be best envisioned as a series of parallel planes shearing past each other.

The shearing motion of the ice is easily imagined in this photo of the Skaftafelljökull in south Iceland. The dark zigzag patterns are volcanic ash layers interbedded with snow and ice at the top of the glacier. The surface elevation rises slowly to the middle of glacier so this view represents a low-angle cross-section. (Photo by Jim Reynolds)

 

The most notable erosional features are seen in mountainous regions, particularly where glaciers form. The first landform that an alpine glacier carves out of a mountain is a cirque. Cirques are large, bowl-shaped hollows gouged out by the ice. They tend to be more abundant on the side of the mountain facing away from the Sun (eg. the north side in the Northern Hemisphere). Cirques tend to have a very steep back wall, referred to as the bergschrund. Some cirques have a lip at their fronts that is higher than the bottom of the bowl. Small lakes, called tarns, occupy these cirques. Cirques and their glaciers often intersect or at least erode the same part of a mountain. This process can create several other cirque-based landforms.

When the high walls of two cirques intersect, an arête is formed. It is a steep, knife-edge ridge that plunges precipitously down the bergschrunds of both cirques. When three or more cirques cut into a mountain from different sides, a steep, prismatic mountain results. This is known as a horn with Switzerland's Matterhorn being an exemplary model for this feature.

Cirque Formation: A) Snow accumulates in a hollow, building to a thickness where ice forms and starts to flow. B) The moving ice carves out the cirque. Ice flowing down the steep wall gouges the back of the cirque deeper than the front. C) After the ice melts this cirque is left with a bergschrund and a tarn.

 

 

 

 

 

 

 

 

 

 

 

The mountain behind the bridge, located in the Norwegian Arctic, is dominated by a large cirque on its north-facing side. (Photo by Jim Reynolds)	Numerous horns dominate the scenery in Chile's Torres del Paine National Park. (Photo by Jim Reynolds)
Several horns can be seen in the glacially-sculpted Caledonide Mountains of the Norwegian Arctic from the ferry crossing one of the fjords. (Photo by Jim Reynolds)	The snow-covered cirque, between these two towers in Torres del Paine National Park, Chile, backs up on another cirque to form a spectacular arête.(Photo by Jim Reynolds)

Once the glacier begins to flow out of its cirque, it tends to follow the most efficient downhill path. Eventually, it enters an existing stream valley and begins to fill it with ice. Because the most active point of erosion in a stream valley is the river bed, these valleys tend to have a V-shape. When one of these valleys is filled with ice, erosion takes place wherever the ice is in contact with the ground. As a result the once V-shaped valley eventually becomes a U-shaped valley. When a U-shaped valley is flooded by the sea, a fjord is formed. Where tributary glaciers join the main glacier, they are usually at about the same elevation at the surface of the ice but the larger glacier erodes much deeper than the tributaries do so there is usually a substantial difference in basal elevation. After the ice melts, a waterfall drops from the hanging valley into the main valley.
California's Yosemite Valley is a beautiful example of a U-shaped valley. It was once entirely filled with ice. The walls on both sides are nearly vertical and rise more than 1,000 m. Bridalveil Fall, on the right, is a classic example of a hanging valley. (Photo by Jim Reynolds)	Lago Huechelaufquén, in the Argentine Andes, is situated in a glacially-carved, bedrock basin. Note the broad, smooth U-shape that marks the base of the glacier in the left background. Lanín volcano is the high peak on the right. It straddles the border with Chile. (Photo by Jim Reynolds)
Fjords cut deep into the Norwegian coastline. The city of Narvík, north of the Arctic Circle, is situated near the mouth of this fjord. (Photo by Jim Reynolds)	Hundreds of hanging valley waterfalls drop into Chile's Ultima Esperanza fjord. The few glaciers that remain around the fjord are in rapid retreat. (Photo by Jim Reynolds)

 

Glacial Sedimentation and Sedimentary Landforms

All glacial sediment is referred to as drift. Drift is deposited in three basic formats:

1. Unstratified Drift
   1. also referred to as till or moraine
   2. exhibits a complete lack of stratification and sorting: clay may be deposited along with house-sized boulders
2. Stratified Drift
   1. exhibits well-defined stratification suggesting that it was deposited by water or wind associated with glaciation
3. Ice-Contact Stratified Drift
   1. water-borne sediment was stratified when deposited in an area surrounded by ice walls
   • the walls supported the sedimentary pile
   2. when the ice melted the support disappeared
   3. the sedimentary pile collapsed outward
      • these deposits typically have numerous small folds and normal faults on their outer edges

Ice-Laid Deposits

Most ice-laid deposits are composed of unstratified drift.

1. The glacier acts as a conveyor belt moving ice and sediment forward to the snout of the glacier.
2. The rubble is piled at the front of the glacier.
   1. If the ice advances over the rubble pile, the process starts again.
   2. If the glacial budget remains at equilibrium for a period of years, a significant pile may build up along the glacial front.
      1. The pile left by the farthest advance of the glacier is called the terminal moraine. These features can be associated with both continental and valley glaciers.
      2. If, during glacial retreat, the climate cools and the glacier becomes stagnant or readvances a similar moraine may form. Because this moraine is not at the farthest advance limit, it is referred to as a recessional moraine.
         1. Numerous recessional moraines can form at the end of an ice age as the glacial front gradually retreats.
      3. During rapid retreat, sediment is just dropped where the ice melted. This is called ground moraine.
      4. See an animation of the formation of a terminal moraine and ground moraine: http://craton.geol.brocku.ca/faculty/rc/teaching/1F90/glaciers/transportanimation.htm
3. Valley glaciers also abrade a lot of material off of the sides of the valleys.
   1. After the glacier melts, the remaining rubble is called lateral moraine.
   2. When two tributary glaciers merge, their lateral moraines on the confluence side of the valley merge as well and are carried out into the glacier in the valley below, looking like a racing stripe.
4. Drumlins--Drumlins are the most enigmatic of the ice-laid deposits. They are streamlined hills that look like inverted spoon. Drumlin orientation is parallel to the ice flow direction and they taper in the down-flow direction. Drumlins usually occur in fields, seldom as isolated features. They may be several tens of meters high and up to a kilometer long.

Drumlins on the north side of the Straits of Magellan, in southern Chile, are oriented east-west, reflecting glacial flow off of the east side of the Patagonian Andes. (Photo by Jim Reynolds)	Argentina's Perito Moreno Glacier descends from the Patagonian ice cap and crosses a narrow arm of Lago Argentino. (Photo by Jim Reynolds)
This is the recessional area in front of the Skaftafellsjökull glacier descending from the Vatnajökull ice cap in southern Iceland. The glacier (left) is very dirty due to the high ablation rate at the snout. Because this glacier has been retreating, most of the gray area is ground moraine. Another valley glacier is seen emerging from the next valley to the east. The North Atlantic Ocean is on the horizon. (Photo by Jim Reynolds)	The prominent medial moraine, farther up-glacier on the Skaftafellsjökull, is formed from the confluence of two smaller glaciers at the nunatak in the upper left of the photo. (Photo by Jim Reynolds)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Water-laid Deposits Water-laid deposits are characteristically stratified. Their grain size is an important indicator of the amount of energy present during deposition. There are several important types of stratified drift.
Outwash deposits--Rivers that breach the terminal/recessional moraine wash sediment onto a broad, usually barren plain known as the outwash plain or by the Icelandic term sandur. This sediment behaves like any river-borne material, being sorted and deposited in accordance with the energy regime of the river. Glaciolacustrine Deposits--Because glaciers dump vast amounts of sediment as they retreat, the local drainage becomes highly disrupted. This results in the formation of numerous lakes. Sediment flowing into these lakes can be divided into two categories: Deltas--Rivers flowing into these lakes deposit the coarser sediments in deltas which prograde outward into the lake. These strata typically exhibit topset, foreset, and bottomset. beds. Lake level is assumed to be at the intersection of the topset and foreset beds. Varved Clays--The fine component of inflowing sediment is carried far out into the lake where it slowly settles to the bottom. Clays in glacial lake regions are distinctive in that they typically exhibit thin, alternating, dark and light beds. These beds represent an annual sedimentation cycle. In the summer months, large amounts of light-colored fine sediment are washed into the lake and are deposited on the bottom. There is a corresponding algal bloom but the amount of algae deposited is miniscule compared to the amount of clay. The influx of sediment shuts down almost completely in the winter months. Dark, organic algal material is deposited in much greater abundance than the light clays. Varved clays serve as a valuable chronometer, preserving an annual record in the lake sediments. They also preserve an important climatic record because the relative thicknesses of the dark and light bands give an indication of the annual temperatures. These variations can be correlated between different lakes across a broad region. Glaciomarine Deposits--Glaciers discharging into the ocean deposit sediments on the ocean floor. These are most easily identified by the presence of marine invertebrate fossils contained within the strata. Dropstones are another glaciomarine and glaciolacustrine feature. These are cobble to boulder size rocks contained within fine-grained sediment. These rocks were rafted out into the ocean in icebergs and fell to the bottom when the ice melted. Loess--Loess is a stratified sedimentary deposit but it is not water-deposited. It is composed of silt- and clay-sized particles deposited by the wind. Strong, cool catabatic winds flow downhill off of glaciers. They pick up the finest grains and carry them outward. Some of these particles may be carried around the world. Most, however, are deposited away from the glacial front where surface wind speeds die down. Loess deposits tend to make very fertile soils. Much of the North American midwest, the Eurasian steppes, and the South American pampas is covered with loess deposits.
Ice at the nose of a glacier is very dirty but it is still mostly ice. I learned the hard way. After walking about 3 steps, I fell hard on my butt! (Photo by Jim Reynolds)	A subglacial river flows out from beneath the Skaftafelljökull glacier in Skaftafell National Park, Iceland. Its sedimentary load will most likely be deposited as stratified drift. (Photo by Jim Reynolds)

The Skeidarársandur is a large, braided outwash plain between the glaciers and the sea (background) at Skaftafell National Park, Iceland. After the subglacial volcanic eruptions in 1997, a large jokulhlaup completely flooded the area to the right of the dike system seen on the left side of the photo. (Photo by Jim Reynolds)	Most of the Patagonian pampas in Argentina are covered with a thin veneer of loess. (Photo by Jim Reynolds)

 

 

 

 

 

 

 

 

 

 

 

Ice-contact Deposits Ice contact deposits are typically found around the margins of glaciers or in hollow areas within glaciers. The sediment is initially deposited, against an ice support, by flowing water leaving stratification. This becomes disrupted when the ice melts. There are several important features that are formed by ice-contact deposits.

Kettles-- As a glacier retreats, large ice blocks calve off. Some of these may be rapidly buried in in either stratified or unstratified drift. When the ice finally melts, months, or even years, later, the strata sag into the depression that is formed to form a kettle. If the bottom of the kettle is below the water table, it will fill with water to form a kettle lake. This is why Minnesota calls itself the "Land of 10,000 Lakes". Note the kettle lakes in front of the glacier in the photo above. Kames--Particularly during summer months, vast volumes of meltwater flow across the surfaces of glaciers. Naturally, they flow in the lowest areas. Numerous crevasses, fractures, and thrust faults cut the glacial surface. Where two or more of these intersect, more surface area of the ice is exposed. As a result, faster melting occurs at these intersections, often creating a vertical abyss that descends deep into the glacier.This is called a moulin. Water carries sediment from the glacial surface into the moulins. Although some of it flows out beneath the bottom of the glacier, the coarse material tends to pile up in the moulin as a thin column of crudely stratified material. When the ice melts, the column collapses and leaves a conical mound called a kame. Kame and kettle topography is typical of recessional areas. Kame Terraces--Water flowing off of the side of a glacier can deposit sediment on top of the ice next to the valley wall. When the ice melts, a kame terrace is left, sometimes precariously plastered against the valley wall. Eskers--Water flowing within or beneath a glacier, particularly during the summer months, melts the ice along its path creating a glacial tunnel. When winter comes, the flow is slowed to a trickle. Any sediment that was carried by the water is deposited on the tunnel floor. This process is repeated year after year. As sediment builds up on the floor, the summer water melts away water on the ceiling, gradually building a sinuous ridge of sediment up into the glacier. Eventually, the ice melts. The oversteepened ridge responds through the formation of normal faults along the sides that cause the sedimentary layers to sag. Most eskers form along the base of the glacier. They wind through the countryside, following the downhill course of the subglacial river. Those that formed higher in the glacier can be identified by the fact that they are draped over the topography. In places they appear to have flowed uphill! Occom Ridge, in Hanover, NH, is an example of this type of esker.

 

Icebergs that calved off of Argentina's Upsala Glacier float on Lago Argentino. The steep, light-colored slopes plastered to the base of the mountain in the background are recently-exposed kame terraces. They mark the approximate level to which the glacier filled the valley as recently as 1998.	This Landsat TM image (29/27/1997/LT5029027009727710) shows numerous kettle lakes in the state line area between North Dakota and Minnesota. http://www.landsat.org/index.html

 

 

Ice Ages: the Pleistocene Glaciation Louis Aggasiz is regarded as the father of glacial geology. His pioneering studies on the rate of movement of a glacier in the Swiss Alps, in the 1830's, catapulted him into prominence. He later emigrated from Switzerland to the United States. As a professor at Harvard University, he traveled widely in the northern regions of Europe and North America. It seemed that everywhere he went he saw landforms that were reminiscent of those he had seen left behind by the glaciers in his native land. The extent of the glacial drift and moraine deposits was so great that he concluded that great ice sheets had flowed out of the north across these continents. It took time for him to convince other geologists of the validity of his hypothesis but by the 1870's the term "ice age" was well known. Aggasiz's studies also demonstrated that the youngest glacial deposits were very young. Through the use of carbon-14 dating, we now know that the last ice age ended about 12,000 years ago.
Multiple Glacial Ages: Continental and Oceanic Records As more people studied glacial deposits, it became obvious that the ice age was not a single event. Stratigraphic and end moraine studies show that in North America there were four major ice advances during the Pleistocene. These advances are named for a state in which their terminal moraine is located. They are listed below with the youngest on top and the oldest on the bottom. Wisconsinan Illinoisian Kansan Nebraskan It became obvious that these might not be the only ice age events since the glaciers that traveled the farthest, such as the Kansan and Nebraskan would obliterate almost all evidence of previous ice ages. Since ocean sediments provide a much more complete record than anything on the continents, various techniques were attempted to see if they could help resolve how many glacial periods there were. A group of geochemists hit upon an idea that works quite well. They decided to look at the record of the concentration of oxygen-16 (O-16) vs. oxygen-18 (O-18) in ocean sediments. This was easily accomplished by collecting fossils with CaCO3 shells from the strata obtained in cores. The idea behind the experiment is that because O-18 is slightly heavier than O-16, it will be somewhat less likely to evaporate as water vapor (H2O). Because evaporation is much greater during warm times than during cool times the O-16/O18 should be greater during cooler periods than warmer periods. This proves to be an effective low-temperature geothermometer. It is likely that there were many more glacial advances in the past 800,000 years than we find recorded on the continents. Figure 16.25 in your book illustrates changes in the O-16/O-18 through time. It suggests that there may have been as many as 9 glacial periods within the last 800,000 years. But the really important finding, which for some reason your book does not include, is that this pattern can be extended back more than 2 million years! Recent paleontological data suggest that the earliest ice age occurred between 2.75-3.0 Ma. More than 30 ice advances may have occurred in the interim.
Climate Geosystem: Causes of Glacial Epochs Milankovitch Cycles Your book attributes much of the reason for multiple ice ages in the Pleistocene to the tectonic configuration of the continents in conjunction with three types of perturbations in Earth's orbit. These perturbations were first hypothesized by John Croll, a Scotsman, in 1867. In the 1930's they were shown to be valid by Milutin Milankovitch, a Serbian mathematician/astronomer. These perturbations occur reoccur with a regular periodicity. We'll discuss each of these cycles below:
I. Eccentricity--Period of ~100,000 years Earth's orbit is not a perfect circle. It is more like a shallow ellipse. The shape of the ellipse stretches and contracts at a regular interval. High eccentricity causes temperature variation due to distance from the Sun to be more extreme. Low eccentricity minimizes temperature variation due to distance from the Sun. At present, Earth's orbit is closest to the Sun in January and farthest from the Sun in July.
	II. Tilt of the Orbital Plane--Periodicity of ~41,000 years Earth's rotational axis is tilted with respect to its orbital plane around the Sun. The angle of this tilt steepens and shallows at a regular interval. The axial tilt gives us seasons. Axial tilt varies between 21.5º and 24.5º. The greater the tilt of the planet, the more extreme the climate will be: colder winters and warmer summers. With a lower tilt angle, the annual climate is more equitable. At present the tilt angle is 23.5º.
III. Precession of Earth's Axis--Periodicity of ~23,000 years Because mass is not evenly distributed around the planet, Earth's axis does a slow wobble as it spins. A kid's top begins to wobble in a circular motion as its rotation slows down. It takes about 23,000 years for one complete wobble to occur. In 11,500 years, the northern hemisphere winter will start in June and summer will start in December. Since, at present, we are closest to the Sun in January and farthest away in June, we can expect a more extreme climate when this occurs.
Conjunction of the Milankovitch cycles occurs about every 200,000 years. Glacial cycles appear to conform well with this 200,000 cycle. The coldest parts of each cycle occur with a ~200,000 year period. Coldest maximum eccentricity maximum tilt farthest from Sun during winter And, offset by half a cycle, every 200,000 years there is an interglacial period. Warmest: minimum eccentricity minimum tilt closest to Sun during summer
A growing number of scientists are concerned that global warming, exacerbated by industrial pollution, could be paving the way for the end of the present interglacial period and the initiation of a new cycle of dramatic global cooling. This could begin within a matter of decades. To understand how this counterintuitive scenario could occur, see: http://www2.brevard.edu/reynoljh/climatechange/introduction.htm.
Evidence from Boreholes in the Ice Deep drilling in the Antarctica and Greenland ice caps has produced a nearly continuous climatic record for the last 750,000 years. The 3 km long core from Antarctica provides a distinct stratigraphic record of snow/ice accumulation on the southern continent. Since the ice contains air bubbles, there is also a concise stratigraphic record of atmospheric chemistry. One of my former students spent two summers drilling the Greenland ice cap. He was working on a highly successful Master's thesis, at the University of New Hampshire, in which he discovered several previously unrecorded volcanic eruptions from ash layers interbedded in the ice.
Climate Since the Last Glaciation Because an average annual temperature difference of only a few degrees can determine whether or not Earth is in an ice age or interglacial climate, it is not surprising that several dramatic fluctuations have occurred since the end of the Wisconsinan Ice Age. The “Little Ice Age” occurred between 1300-1850 A.D. (Longer than your book states.) Washington crossed an ice-covered Delaware River. The Delaware almost never freezes today. Hans Brinker skated to victory on Dutch canals that rarely freeze today. Until the early 19th Century, people regularly walked across the frozen Hudson River to get to New York City from New Jersey. This portion of the river has not frozen in decades. In Europe, t he Baltic Sea and Thames River froze regularly. Abrupt warming about 1,000 years ago allowed the Vikings to establish settlements in Greenland, Iceland, and Vineland. The Greenland settlements were abandoned during the Little Ice Age. The reasons for the failure of the Vineland settlements are unclear but they failed around this time too. 8,200 years ago • Another abrupt cooling event took place. It lasted about a century and was a relatively mild change. A change of this magnitude today would be catastrophic. The growing season across Eurasia and North America would shorten dramatically. The Younger Dryas--about 12,700 years ago Temperatures in the North Atlantic dropped about 5ºC causing the region to slip into a cooling period. The Younger Dryas lasted about 1,300 years.
The Record of Ancient Glaciation The late Pliocene and Pleistocene has been a time period dominated by cooler climates. Only four other ice age times are known in Earth's history. Pennsylvanian-Permian: southern Gondwana was centered over the South Pole. Ordovician: northern Africa was centered over the South Pole. The two Proterozoic ice ages are extremely interesting. 600-750 Ma--evidence suggests that ice may have extended to the equator. There were several glaciations and interglacial periods. Some geologists speculate that the entire planet was covered in ice. This is known as the Snowball Earth hypothesis. see http://ecosystems.wcp.muohio.edu/studentresearch/climatechange02/snowball/articles/home.html The oldest confirmed ice age occurred around 2.4 billion years ago (Ga). There is some suggestion of a 3.0 Ga ice age as well.
Websites http://www.glacier.rice.edu/