You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Spec mapping (AQA 7037): Paper 1 (Physical), §3.1.5 Hazards — the concept of hazard in a geographical context and the plate tectonic theory of crustal evolution that underpins the global distribution of tectonic hazards. This depth lesson assumes the GCSE outline of plate tectonics is known and develops the mechanisms (mantle structure, convection, slab pull, ridge push), the evidence base (palaeomagnetism, sea-floor spreading, modern GPS geodesy), and the boundary processes (Wadati–Benioff zones, hotspots, mantle plumes) at degree-foundation depth. It supplies the physical architecture within which the later vulcanicity (§3.1.5), seismicity (§3.1.5) and multi-hazard content sits, draws explicitly on the systems framing of §3.1.1 (the Earth as an energy system driven by internal heat), and links forward to the human-vulnerability material of §3.2 (population, resources, urban) where exposure converts hazard into disaster. Assessment objectives: AO1 (knowledge of structure, mechanism and evidence), AO2 (applying boundary processes to located margins) and AO3 (manipulating and evaluating plate-velocity, age and palaeomagnetic data).
Throughout this course the organising idea is the risk equation:
Risk=CapacityHazard×Vulnerability
Plate tectonics determines the Hazard term — where, how often and how violently the Earth's lithosphere will move. It says nothing about Vulnerability or Capacity, which is why two earthquakes of identical magnitude can kill 200,000 people in one place and 500 in another. Keeping the equation in view from the outset prevents the most common A-Level error: treating physical magnitude as if it were the same thing as disaster impact.
It is also worth pausing on what "hazard" actually means in geography. A natural hazard is a natural process or event that has the potential to cause loss of life, injury, property damage or socio-economic disruption. The critical word is potential: a hazard becomes a disaster only when it intersects with people and assets that are exposed and vulnerable. An identical magnitude-7 earthquake is a "hazard" whether it strikes an empty stretch of the Atacama Desert or central Port-au-Prince — but it is a disaster only in the second case. This is why disasters are sometimes described as un-natural: the trigger is natural, but the loss is overwhelmingly a product of human decisions about where and how to build, how to govern and how to distribute resources. Geographers capture this with the idea of the hazard–risk continuum and with the pressure-and-release (PAR) model, which traces a disaster back through unsafe conditions (fragile buildings, dangerous locations), dynamic pressures (rapid urbanisation, lack of training or institutions) to deep root causes (poverty, limited access to power and resources). Plate tectonics provides the geophysical "pressure" on one side; the PAR model reminds us that the human "pressures" on the other side are at least as important. Throughout this depth course, the discipline is to explain the physics rigorously and to keep asking who is exposed, who is vulnerable, and who has the capacity to cope.
This lesson concentrates on the physics — the Hazard term — but does so in a way that constantly anticipates the human geography to come. Understanding why the Pacific Rim, the Mid-Atlantic, the Mediterranean and the Himalayan front are where they are, and why each produces a characteristic magnitude, depth and frequency of event, is the indispensable foundation for everything that follows about vulnerability, impact and management.
A rigorous account of plate motion begins with the layering of the Earth, defined in two complementary ways — by chemistry (crust, mantle, core) and by mechanical behaviour (lithosphere, asthenosphere, mesosphere, outer and inner core). The distinction matters because plates are lithospheric, not crustal: a tectonic plate is the rigid crust plus the uppermost, cool, brittle part of the mantle, riding on the weaker asthenosphere beneath.
| Layer (mechanical) | Depth (approx.) | State / behaviour | Density (g/cm³) |
|---|---|---|---|
| Lithosphere (crust + upper mantle) | 0–100 km | Rigid, brittle; broken into plates | 2.7 (continental) – 3.0 (oceanic) |
| Asthenosphere | 100–700 km | Solid but ductile; deforms by creep; partial melt | ~3.3 |
| Mesosphere (lower mantle) | 700–2,900 km | Solid, higher viscosity | 3.3–5.5 |
| Outer core | 2,900–5,150 km | Liquid Fe–Ni; convects to generate magnetic field | 9.9–12.2 |
| Inner core | 5,150–6,371 km | Solid Fe–Ni (immense pressure) | 12.8–13.1 |
Two further contrasts are examined often. Oceanic crust (sima, basaltic, ~7 km thick, density ~3.0 g/cm³) is thinner, denser and younger than continental crust (sial, granitic, 30–70 km thick, density ~2.7 g/cm³, up to ~4 billion years old). The density contrast is the single most important physical fact in tectonics: it is why oceanic lithosphere subducts and continental lithosphere does not, and therefore why the world's most explosive volcanoes and deepest earthquakes cluster at ocean-margin destructive boundaries rather than at continental collisions.
The Earth's internal heat — the energy source for the whole system — comes from two sources: primordial heat left from accretion and core formation 4.6 billion years ago, and ongoing radiogenic heat from the decay of long-lived radioactive isotopes (uranium-238, thorium-232 and potassium-40) distributed through the mantle. The global heat flux is roughly 47 terawatts. It is this heat, escaping outward, that drives mantle convection and ultimately every tectonic hazard in this course.
It is helpful to treat the solid Earth explicitly as a system in the §3.1.1 sense, because that framing recurs across the whole specification. The input is internal heat (radiogenic + primordial); the throughput is the transfer of that heat by conduction (slow, through the rigid lithosphere) and, far more efficiently, by convection (in the ductile mantle and liquid outer core); the output is heat lost at the surface, together with the mechanical work expressed as plate motion, mountain building, vulcanicity and seismicity. The system is driven by an energy gradient — hot interior, cool surface — exactly as the atmospheric and coastal systems studied elsewhere are driven by gradients in insolation or wave energy. Recognising this lets you transfer concepts: just as a coast can sit in dynamic equilibrium and re-organise when a threshold is crossed, the lithosphere stores elastic strain until a fault reaches its failure threshold and ruptures. The vocabulary of stores, flows, feedback and thresholds is therefore not decorative — it is the connective tissue that makes the specification synoptic, and AQA rewards candidates who use it deliberately rather than treating tectonics as a stand-alone topic.
The boundaries between layers are defined by sharp changes in seismic-wave velocity called discontinuities: the Mohorovičić discontinuity (Moho) separates crust from mantle, and the Gutenberg discontinuity marks the mantle–core boundary. The very existence of these layers was deduced not by drilling — the deepest borehole, the Kola Superdeep, reached only ~12 km, a fraction of the way through the crust — but by reading how earthquake waves refract and reflect at depth. The fact that S-waves cannot pass through the outer core demonstrates that it is liquid (shear waves require rigidity), creating an S-wave shadow zone beyond 103° of arc from an earthquake; the bending of P-waves maps the core boundaries precisely. In other words, our knowledge of the Earth's deep structure is itself a product of seismic-hazard science, a neat illustration of how the study of hazards has advanced fundamental geophysics.
flowchart TD
A[Radiogenic + primordial heat at depth] --> B[Mantle convection]
B --> C[Ridge push at constructive margins]
B --> D[Slab pull at destructive margins]
C --> E[Plate motion]
D --> E
E --> F[Boundary processes: vulcanicity, seismicity]
F --> G[Hazard term in Risk equation]
Alfred Wegener proposed that the continents were once assembled in a single supercontinent, Pangaea, which fragmented from ~200 million years ago. His evidence was circumstantial but cumulative:
Crucially, Wegener could not specify a mechanism: his suggestion that continents "ploughed" through oceanic crust was physically impossible, and the theory was largely dismissed for four decades.
The breakthrough came from rock magnetism. Iron-bearing minerals such as magnetite lock in the direction and inclination of the ambient geomagnetic field at the moment they cool below the Curie point (~580 °C). Two independent signals emerge:
Harry Hess proposed that new oceanic lithosphere forms continuously at mid-ocean ridges and spreads laterally — "the ocean floor as a conveyor belt." Three lines of evidence confirmed it:
Continuous Global Navigation Satellite System (GNSS/GPS) networks now measure plate motion directly, to millimetre precision, confirming the theory in real time:
The pattern in the GPS data is itself diagnostic: plates with a large fraction of their margin subducting (Pacific, Nazca, Cocos) move fast; plates dominated by ridge and collision boundaries (Eurasia, Africa, North America) move slowly. This correlation is the strongest direct evidence that slab pull, not convection alone, is the dominant driving force.
A further strength of the GPS evidence is that it is independent of, yet consistent with, every older line of argument. It confirms the direction and rate implied by sea-floor magnetic stripes, matches the absolute motion derived from hotspot tracks, and even resolves intra-plate strain — for instance the slow squeezing of the Himalayan front or the locked-and-loading state of the Cascadia and Nankai megathrusts, where GPS detects the surface bulging inward as strain accumulates between great earthquakes. This convergence of completely different methods — fossils, palaeoclimate, palaeomagnetism, ocean-floor dating, hotspot geometry and satellite geodesy — onto a single, coherent model is what makes plate tectonics one of the most robustly evidenced theories in the Earth sciences, and it is exactly the kind of triangulated, multi-strand argument AQA rewards in evaluative answers.
The original "mantle conveyor belt" model — plates passively rafted on convection cells — is now considered incomplete. Modern understanding ranks the driving forces as follows.
At a subduction zone, cold, dense oceanic lithosphere sinks into the asthenosphere. As it sinks it undergoes mineral phase changes (e.g. basalt to denser eclogite) that make it denser still, so gravity pulls the descending slab — and the entire attached plate — downward and forward. Slab pull is now estimated to account for the majority of the net driving force, which is exactly why slab-rich plates move fastest (the GPS correlation above).
Newly formed lithosphere at a mid-ocean ridge is hot, buoyant and topographically high. As it cools and thickens moving away from the ridge, it subsides, and gravity causes it to slide down the gentle flank of the ridge — a body force ("gravitational sliding") that pushes the plate horizontally. Ridge push is real but several times weaker than slab pull.
Convection of the whole mantle, driven by the radiogenic/primordial heat budget, organises the broad pattern of upwelling and downwelling. Basal drag — coupling between the moving asthenosphere and the underside of the plate — can either help or resist motion depending on whether the flow beneath is faster or slower than the plate. Convection sets the stage; slab pull and ridge push do most of the work.
Fnet≈Fslab pull+Fridge push±Fbasal drag−Fresistance
| Boundary | Relative motion | Driving force | Landforms | Hazard signature |
|---|---|---|---|---|
| Constructive / divergent | Plates move apart | Ridge push, upwelling | Mid-ocean ridge, rift valley, shield volcanoes | Frequent shallow quakes (low Mw); effusive basaltic eruptions |
| Destructive / oceanic–continental | Oceanic subducts under continental | Slab pull | Ocean trench, fold mountains, composite volcanoes | Deep Wadati–Benioff quakes (up to Mw 9+); explosive andesitic eruptions; tsunami |
| Destructive / oceanic–oceanic | Denser plate subducts | Slab pull | Trench, island arc | Deep quakes; explosive eruptions; tsunami |
| Collision / continental–continental | Convergence, no subduction | Slab pull (former ocean) | Fold mountains (Himalayas), high plateau | Powerful shallow thrust quakes; no volcanism |
| Conservative / transform | Plates slide past | Differential drag | Fault scarps, offset features | Shallow but powerful quakes; no volcanism |
At a constructive (divergent) margin, decompression of the rising mantle (it melts because pressure falls, not because temperature rises) generates copious low-silica basalt. On the ocean floor this builds the mid-ocean ridge and erupts as pillow lavas; on land an incipient rift such as the East African Rift produces fissure eruptions and basaltic shield/scoria volcanoes (Erta Ale, Nyiragongo). Earthquakes are frequent but shallow and low-magnitude, because the lithosphere here is thin, hot and weak, so it cannot store large amounts of elastic strain before it fails.
At a destructive (convergent) margin involving oceanic lithosphere, the descending slab carries water-bearing minerals down into the mantle; as it heats, dehydration releases water into the overlying mantle wedge, lowering its melting point (flux melting) and generating gas-rich, intermediate (andesitic) magma that rises to feed explosive composite volcanoes and island arcs. The locked, then suddenly slipping, interface between the two plates (a megathrust) stores enormous strain over centuries and releases it in the planet's largest earthquakes (Mw 9+), which also displace the sea floor and generate tsunami.
At a collision margin, two buoyant continental masses meet; neither will subduct, so the crust crumples and thickens into fold mountains and a high plateau (the Himalayas and Tibet). There is no volcanism because no oceanic slab descends to flux-melt the mantle, but powerful shallow thrust earthquakes are common (Nepal 2015). At a conservative (transform) margin, plates grind past one another along a fault; there is no creation or destruction of lithosphere and therefore no volcanism, but friction locks the fault until it ruptures in shallow, high-magnitude earthquakes (San Andreas; North Anatolian).
At destructive margins, earthquake foci are not random with depth. They define an inclined plane — the Wadati–Benioff zone — that dips beneath the overriding plate at 30–70°, tracing the descending slab. Quakes occur from the surface down to ~700 km (the deepest possible, below which the slab loses its brittleness because increasing temperature and pressure make it deform plastically rather than rupture). The systematic deepening of foci away from the trench, beneath the continent, was historically one of the clinching pieces of evidence for subduction, and it is the reason destructive margins produce the full depth range of earthquakes while constructive and conservative margins produce only shallow ones. The zone also has direct hazard relevance: the shallow part of the Wadati–Benioff zone, where the megathrust locks, hosts the great tsunamigenic earthquakes, whereas intermediate and deep events, though they can be very large in magnitude, are usually less destructive at the surface because their energy attenuates over the long path to the ground.
flowchart LR
T["Ocean trench"] --> S0["Shallow foci 0-70 km"]
S0 --> S1["Intermediate 70-300 km"]
S1 --> S2["Deep foci 300-700 km"]
S2 --> M["Slab loses brittleness"]
A minority of volcanism occurs within plates, far from any boundary, above mantle plumes — narrow columns of anomalously hot material rising from deep in the mantle (possibly the core–mantle boundary). Because the plume is broadly fixed while the plate moves over it, a time-progressive chain of volcanoes forms, oldest furthest from the present hotspot. Hotspots matter for hazard geography because they explain the existence of dangerous, productive volcanoes (and the islands they build) in places the plate-boundary model alone would leave blank — Hawai'i, Iceland (which sits on both a ridge and a plume, explaining its extraordinary volcanic productivity), Réunion and Yellowstone are all hotspot-related. They are the principal "anomaly to explain" in exam questions on the distribution of volcanic activity, and a candidate who can account for an intra-plate volcano by invoking a mantle plume, while noting that most volcanism is boundary-controlled, demonstrates exactly the kind of secure, nuanced spatial reasoning the specification rewards.
The Hawaiian–Emperor seamount chain is the type example: Hawai'i (the Big Island) sits over the plume today and is volcanically active; islands to the north-west are progressively older and more eroded. Dividing the distance between dated volcanoes by their age difference yields a plate velocity that independently matches the GPS rate (~9–10 cm/yr north-west), and the sharp ~60° bend in the chain (~47 million years ago) records a change in Pacific Plate direction. Hotspots therefore provide a measure of absolute plate motion against a deep-mantle reference frame, complementing the relative motions read at boundaries.
J. Tuzo Wilson recognised that ocean basins open and close over hundreds of millions of years in a repeating sequence, linking the assembly and break-up of supercontinents (Pangaea, and before it Rodinia):
The cycle explains the distribution of ancient mountain belts, ophiolite complexes (slivers of ocean floor obducted onto continents) and the periodicity of supercontinent assembly.
A recurring AQA demand is to characterise tectonic hazards in terms of their spatial distribution, magnitude, frequency, regularity and predictability — and plate tectonics explains all five. The distribution is overwhelmingly linear and boundary-controlled: the "Ring of Fire" around the Pacific accounts for ~75% of the world's active volcanoes and ~80% of large earthquakes because it is an almost continuous girdle of subduction; the Mid-Atlantic Ridge traces a line of constructive activity; the Alpine–Himalayan belt marks the collisional suture from the Mediterranean to Indonesia. The clear exceptions — intra-plate hotspots (Hawai'i) and intra-plate earthquakes (such as the 1811–12 New Madrid sequence in the central USA) — are the cases that prove the rule and most often appear as "anomalies to explain" in exams.
Magnitude is also boundary-controlled: the largest earthquakes (Mw 9+) occur only on subduction megathrusts because only there does a fault large enough and locked enough exist to store the necessary strain; the most explosive eruptions occur at subduction zones for the magma-chemistry reasons developed in the next lesson. Frequency is inversely related to magnitude — small events are common, great events rare — a relationship formalised for earthquakes by the Gutenberg–Richter law, log10N=a−bM, where N is the number of events of magnitude ≥M in a given period and b is typically close to 1, meaning each unit increase in magnitude corresponds to roughly a tenfold decrease in frequency.
Regularity and predictability are where tectonic theory delivers a sobering message. Plate boundaries make the where highly predictable, and probabilistic hazard maps can state long-run likelihoods, but the when of any individual rupture remains essentially unforecastable in the short term. The seismic-gap and elastic-rebound ideas suggest that segments which have not slipped for a long time are "overdue", but strain does not accumulate at perfectly constant rates and faults interact, so precise timing eludes us. This asymmetry — confident spatial prediction, near-zero temporal prediction — is the single most important governance fact in this course and the reason later lessons argue that managing vulnerability and capacity, not predicting the Earth, is where lives are saved.
A frequent skills task supplies plate velocities and asks you to describe, manipulate, explain and evaluate. Consider:
| Plate | GPS velocity (cm/yr) | Fraction of margin subducting |
|---|---|---|
| Pacific | 9.4 | High |
| Nazca | 7.6 | High |
| Indian | 4.8 | Moderate |
| North American | 2.3 | Low |
| Eurasian | 1.1 | Low |
| African | 2.2 | Low |
Describe: velocities range from 1.1 to 9.4 cm/yr — a factor of ~8.5 between the slowest (Eurasian) and fastest (Pacific) plates.
Manipulate: the mean velocity is (9.4 + 7.6 + 4.8 + 2.3 + 1.1 + 2.2) / 6 = 4.6 cm/yr. Express the Pacific rate over a human lifetime: 9.4 cm/yr × 80 yr ≈ 7.5 m — about the length of a classroom — emphasising that plates are imperceptibly slow on human timescales yet relentless on geological ones.
Explain: the data correlate strongly with the fraction of margin subducting — the three fast plates (Pacific, Nazca, Indian) all have extensive subduction zones, supporting the dominance of slab pull over convection/ridge push.
Evaluate: the relationship is correlational, not a controlled experiment; African and North American plates have similar low velocities despite different boundary configurations, so additional factors (plate area, continental keel "anchoring," basal drag) modulate the simple slab-pull story. The data are also present-day snapshots — they cannot capture the episodic acceleration of plates during slab break-off events recorded in the geological past. A robust judgement is therefore that slab pull is the primary control on first-order velocity differences, but not the sole control.
This worked sequence is a template you can transfer to any quantitative resource in the exam: never stop at describing the pattern (the most common under-performance), always manipulate the figures (a mean, a ratio, a rate over a stated period, a percentage), then explain the pattern using the underlying process, and finally evaluate — interrogate the data's reliability, representativeness and the strength of the inferred relationship. The mark schemes for the data-response questions in §3.1.5 explicitly reserve the upper band for candidates who evaluate the evidence rather than merely report it, so building the habit of asking "how good is this dataset and what can it really tell me?" is worth doing from the first practice question.
"Assess the relative importance of the different mechanisms driving plate movement." (9 marks — AO1 4 / AO2 3 / AO3 2)
Plates move because of convection currents in the mantle. Heat from radioactive decay makes the mantle rise at constructive boundaries and sink at destructive boundaries, dragging the plates along. Slab pull happens at subduction zones where the heavy oceanic plate sinks and pulls the rest of the plate with it, and ridge push happens at mid-ocean ridges where new crust slides off the high ridge. All three mechanisms work together to move the plates.
The traditional model emphasised mantle convection, but evidence increasingly shows that the gravitational forces of slab pull and ridge push dominate. Slab pull operates at destructive margins, where cold, dense oceanic lithosphere sinks and becomes denser still through phase change to eclogite, pulling the attached plate. This is supported by GPS data: plates with extensive subducting margins (Pacific, ~9–10 cm/yr) move far faster than ridge-dominated plates (Eurasia, ~1 cm/yr). Ridge push is a weaker body force from gravitational sliding off the elevated ridge. Convection organises the broad pattern but cannot alone explain the velocity differences.
Quantitatively, the velocity data are the decisive discriminator. If convection were dominant, plate speed should relate to mantle heat flux rather than boundary type, yet the observed ~8.5-fold range correlates with the fraction of margin subducting (Pacific 9.4 vs Eurasia 1.1 cm/yr) — direct evidence that slab pull is the first-order control, plausibly the majority of the net force. Ridge push is genuinely secondary, and basal drag is dual-signed — it can resist as readily as it drives, which is why continental plates with deep "keels" (Eurasia, Africa) move slowly despite their size. However, the importance is context-dependent: at a young ocean (Red Sea) with no subduction, ridge push and convective upwelling necessarily dominate because no slab yet exists to pull. The defensible judgement is therefore conditional: slab pull dominates mature, subduction-girt plates and explains the global velocity pattern, but the mix of forces shifts systematically through the Wilson cycle, so no single mechanism is universally "most important."
The Mid-band answer lists the three mechanisms accurately (secure AO1) but treats them as equal partners and offers no evidence or judgement — it never assesses. The Stronger answer ranks the forces, supplies the GPS evidence that discriminates between them (AO3) and explains the slab-pull mechanism with the eclogite phase-change detail (deeper AO1). The Top-band answer is distinguished by quantified, evaluative reasoning: it uses the magnitude of the velocity range as a test of competing hypotheses, recognises basal drag as dual-signed, and — crucially — makes the importance conditional on the Wilson-cycle stage, reaching a defensible, nuanced judgement rather than a flat ranking. That conditional, evidenced judgement is the top-band discriminator.
The deep-Earth mechanism remains an active research frontier. Seismic tomography — imaging the mantle with earthquake waves — now resolves subducted slabs sinking to the core–mantle boundary and giant "thermochemical piles" (LLSVPs) beneath Africa and the Pacific that may anchor long-lived plumes. Debate continues over whether plumes truly originate at the core–mantle boundary or are shallower features, and over whether whole-mantle or layered convection prevails. For hazard geography the frontier question is timescale and predictability: tectonics sets the long-term Hazard term with confidence (we know where the Ring of Fire is), but the timing of individual ruptures remains unforecastable, which is precisely why later lessons stress that managing Vulnerability and Capacity — not predicting the Earth — is where lives are actually saved.
This content is aligned with the AQA A-Level Geography (7037) specification.