Most of the sediment that glaciers carry comes from the bedrock they grind across — not from the sky, not from falling rocks, and definitely not from the ice itself. But the real load? That surprises people. The stuff that builds moraines, fills fjords, and shapes entire landscapes? You picture a glacier as a dirty frozen river, and sure, there's debris on top. It comes from underneath And that's really what it comes down to..
You'll probably want to bookmark this section.
I've stood at the toe of Exit Glacier in Alaska. Day to day, i've hiked the lateral moraines of the Athabasca. And every time, the same thought hits: this massive pile of rock flour, boulders, and gravel didn't fall from above. The ice made it.
What Is Glacial Sediment (and Where Does It Come From)
Glacial sediment — geologists call it till when it's unsorted, outwash when water sorts it — is basically crushed bedrock. In real terms, as a glacier moves, it drags debris across the bedrock floor. The ice acts like a bulldozer with a built-in rock crusher. That debris does the cutting. The ice just provides the weight and the motion.
Here's the breakdown: about 90% of what a glacier carries comes from subglacial erosion. So the rest? Supraglacial debris (rockfall onto the surface), englacial material (stuff trapped inside), and a tiny bit of wind-blown dust. But the heavy lifting happens at the bed Small thing, real impact. That alone is useful..
The Two Sources That Matter
Basal erosion is the big one. Plucking and abrasion — we'll get to those — chew up the bedrock and shove the fragments into the ice. This is where the silt, the sand, the striated cobbles come from.
Marginal erosion happens at the sides. Valley walls crumble. Rockfalls land on the ice edges. This contributes, especially in steep terrain, but it's secondary. The volume just isn't comparable.
Why It Matters / Why People Care
You might wonder: okay, rocks get crushed. So what?
So everything. The sediment load determines how fast a glacier erodes its bed. On the flip side, it controls the shape of valleys — U-shaped vs. V-shaped. It builds the moraines that dam lakes and create flood hazards. Think about it: it feeds deltas and continental shelves. And it locks up carbon when rock flour reacts with water and air.
Farmers in Washington's Skagit Valley grow crops on glacial outwash. Hydroelectric dams in Norway manage sediment from melting ice. Coastal planners in Alaska watch deltas shift as glaciers retreat and sediment supply changes Small thing, real impact..
And if you're reading this because you're a student staring at a textbook diagram of a glacier with arrows labeled "ablation till" and "lodgement till" — understanding where the sediment comes from is the key to actually getting it. That said, not memorizing. *Getting it.
How It Works: The Two Main Erosion Processes
Glaciers don't erode by magic. Both require debris in the ice. In real terms, they use two mechanisms. Still, both require pressure. And both leave distinct signatures That's the whole idea..
Plucking: When Ice Grabs and Tears
Imagine a block of bedrock with a crack in it. High pressure on the upstream side. Expands. But water seeps in. Now the glacier overrides it. Day to day, freezes. Because of that, the crack widens. Low pressure — cavitation, effectively — on the downstream side. The ice pulls the block loose Still holds up..
That's plucking. Also called quarrying. It works best where bedrock is jointed, fractured, or already weakened. The glacier doesn't "cut" the rock — it rips chunks out and carries them away.
You see the evidence in roches moutonnées: smooth, gently sloping upstream sides (abrasion) and steep, jagged downstream faces (plucking). Textbook. Plus, classic. And genuinely useful in the field — if you can read them, you know which way the ice flowed.
Plucking produces the big stuff. But boulders. Angular blocks. On top of that, the coarse fraction of till. It's also why glacial valleys have those steep headwalls — cirques are basically giant plucking zones And it works..
Abrasion: Sandpaper on a Massive Scale
Now picture the same glacier, but focus on the sole. Debris frozen into the base — clasts, pebbles, rock flour — gets dragged across the bedrock under tons of ice pressure. It scratches. It gouges. It polishes. It produces rock flour: silt-sized particles so fine they stay suspended in meltwater for weeks, turning rivers milky turquoise Simple, but easy to overlook..
Abrasion needs three things:
- Debris at the bed (tools)
- Normal force (ice thickness × gravity)
- Sliding velocity
No debris? Think about it: no abrasion. And clean ice slides over bedrock like a hockey puck on ice — almost no wear. In practice, this is why cold-based glaciers (frozen to their beds) barely erode. They're stuck. No sliding, no tools, no abrasion It's one of those things that adds up..
The striations on polished bedrock? Think about it: those are abrasion's fingerprints. But parallel grooves. Sometimes crossing sets — evidence of shifting flow directions. I've traced them with my fingers on outcrops in the Adirondacks. It's a direct line to ice that moved 20,000 years ago.
The Feedback Loop Nobody Talks About
Here's what most intro texts skip: erosion creates its own tools.
Plucking supplies clasts. Think about it: abrasion produces rock flour, which lubricates the bed, which increases sliding, which increases both processes. Because of that, those clasts become the abrasive tools for abrasion. It's a self-reinforcing system — until the debris supply runs out or the glacier thins.
This is why fast, thick, temperate glaciers in fractured bedrock are erosion monsters. And why cold-based ice sheets in shield terrain (think Laurentide over the Canadian Shield) left behind... But not much. Just a thin veneer of till and a lot of polished rock Surprisingly effective..
Common Mistakes / What Most People Get Wrong
Mistake 1: "Glaciers carry sediment from the top down."
Nope. Supraglacial debris exists, sure. But it's a rounding error. The volume from rockfall onto the surface is tiny compared to what's quarried from the bed. If the glacier thickens, that surface debris gets buried — becomes englacial — and eventually reaches the bed. But it started as basal material in a previous cycle No workaround needed..
Mistake 2: "All till looks the same."
Lodgement till (plastered down under active ice) is dense, overconsolidated, clast-rich, with a strong fabric. Melt-out till (let down as ice stagnates) is looser, more matrix-rich, often stratified. Flow till (remobilized by deformation) looks different again. If you're logging a core or mapping a quarry, knowing the difference tells you how the ice behaved.
Mistake 3: "Glacial erosion is constant."
It's wildly variable. Seasonal
Seasonal swings in meltwater production are the most obvious driver of that variability. During warm months, surface melt penetrates crevasses and moulins, pressurizing the subglacial drainage system. Elevated water pressure reduces effective normal stress at the bed, allowing the ice to slide faster and to entrain more debris. The resulting spike in sliding velocity amplifies both plucking and abrasion, often leaving a distinct summer‑layer of coarse clasts embedded in the till. Conversely, when winter freezes the meltwater network, basal drag rises, sliding slows, and the glacier shifts toward a more stagnant, deformation‑dominated regime. In that state, erosion rates can drop by an order of magnitude, and the till that accumulates tends to be finer‑grained and more matrix‑rich.
Beyond the annual cycle, longer‑term climatic oscillations modulate the same feedbacks. Multi‑year warm phases thin the glacier, decreasing the normal force that drives abrasion, yet they also increase meltwater flux, which can lubricate the bed enough to maintain high sliding speeds despite reduced thickness. The net effect depends on the balance between these opposing influences, which is why some temperate glaciers show accelerated erosion during periods of modest thinning, while others experience a slowdown when thinning outpaces meltwater generation.
Spatial heterogeneity adds another layer of complexity. Fault zones and shear zones act as natural conduits for subglacial water, focusing high‑pressure flow and creating localized hotspots of erosion that can carve overdeepened basins or tunnel valleys. Bedrock lithology controls how readily clasts are plucked: heavily fractured, jointed granites yield abundant angular debris, whereas massive, unfractured basalts resist quarrying and favor abrasion of existing clasts. Glacial geologists often use the spacing and orientation of striations, crescentic gouges, and chatter marks to infer whether a particular patch of bed was dominated by plucking, abrasion, or a hybrid regime during a given advance.
The official docs gloss over this. That's a mistake.
Temporal spikes in erosion also arise from glacier dynamics unrelated to climate. In real terms, surge events, driven by internal instabilities in the ice flow, can momentarily boost sliding velocities by an order of magnitude, transporting huge volumes of basal debris downstream in a matter of weeks. The deposits left behind — often coarse, poorly sorted, and showing strong fabric alignment — serve as recognizable signatures of these short‑lived, high‑energy episodes. Similarly, subglacial outburst floods (jökulhlaups) can scour the bed with turbulent, sediment‑laden waters, producing erosional features that mimic those made by ice but are actually fluvial in origin That's the whole idea..
People argue about this. Here's where I land on it.
All of these controls interact in a non‑linear fashion. A surge may expose fresh bedrock, increasing the supply of pluckable clasts; the ensuing rise in abrasive efficiency can then sustain high erosion rates even after the surge wanes, until the debris reservoir is exhausted or the glacier thins enough to reduce basal shear stress. Recognizing these feedback loops is essential for interpreting the geological record: thick, stratified till sequences often record alternating phases of high‑energy, debris‑rich flow and quieter, melt‑out dominated deposition, while polished bedrock surfaces preserve the integrated signature of long‑term abrasion punctuated by episodic plucking spikes The details matter here..
Conclusion
Glacial erosion is far from a steady, uniform wear‑and‑tear process. It emerges from the interplay of debris availability, normal stress imposed by ice thickness, and basal sliding speed — each of which fluctuates with seasons, climate trends, bedrock properties, and internal glacier dynamics. The system is self‑reinforcing: plucking generates the tools that abrasion uses, while abrasion’s product, rock flour, can lubricate the bed and accelerate sliding, thereby fueling further erosion. Recognizing the variability and feedbacks inherent in this system allows geologists to decode the complex histories encoded in tills, striated bedrock, and glacially carved landscapes, and to predict how contemporary glaciers will sculpt the Earth under a warming climate.
