Folding, faulting, and rock deformation mechanics

Intuition Beginner

A chocolate bar snaps when it is cold and bends when it is warm. Ice shatters under a hammer but flows as a glacier over decades. The same material can be brittle or ductile depending on the conditions, and rock is no different. Whether a layer folds smoothly or cracks into faults is decided by temperature, pressure, how fast the force is applied, and how long it acts — not by the rock type alone.

Three responses appear, blending into one another. A rock loaded briefly springs back when released: that is elastic behavior, like a rubber band. Held under stress long enough it bends permanently, flowing like warm tar: that is ductile behavior. Loaded too fast or too cold it fractures: that is brittle behavior. The boundary between folding and fracturing moves with depth, because the deep crust is hot and the shallow crust is cold.

Rate matters as much as temperature. Pull a piece of Silly Putty quickly and it snaps; pull it slowly and it stretches into a thin ribbon. Rocks show the same contrast: a meteorite impact loads rock for an instant and shatters it, while the same rock creeps slowly through a mountain root over millions of years. How fast the shape changes — the strain rate — is the second dial, alongside temperature, that selects the deformation style.

This unit unpacks the mechanics behind those dials. You will meet the tools geologists use to measure how much a rock has changed shape, the criteria that predict when it will crack, the geometry that folds and faults take on once they form, and the laws that describe how rocks flow. The earlier unit 27.09.01 introduced folds, faults, and the stresses that drive them; here we ask how and why the rock responds as it does.

Visual Beginner

Variable	Rises with depth?	Favors...	Notes
Temperature	Yes (~25 °C per km)	Ductile flow	Hot rock creeps; cold rock fractures
Confining pressure	Yes (~27 MPa per km)	Ductile flow	Suppresses fracture, holds grains together
Strain rate	No (set by tectonics)	Brittle fracture when high	Fast loading breaks rock; slow loading flows it
Pore-fluid pressure	Variable	Brittle fracture when high	Lowers effective normal stress, promotes slip

Fold attribute	What it measures	Reads as...
Plunge	Tilt of the hinge line from horizontal	Direction the fold axis noses into the ground
Vergence	Direction an asymmetric fold overturns toward	Sense of tectonic transport
Interlimb angle	Angle between the two limbs	How tight the fold is (gentle to isoclinal)
Axial-surface attitude	Orientation of the plane bisecting the hinge	Upright, inclined, overturned, or recumbent

Worked example Beginner

The brittle-ductile transition is the depth at which a rock's failure style switches from fracture to flow. For quartz-rich crust the switch sits near the temperature where quartz begins to creep, about 300 to 450 degrees Celsius. Take a geothermal gradient of 25 degrees per kilometre. The shallow edge of the transition lies at $300/25=12$ kilometres depth, and the deep edge at $450/25=18$ kilometres. Above about 12 km the crust faults; below about 18 km it folds and flows. A single sedimentary layer can therefore be folded at its base and faulted at its top, because the two depths sit on opposite sides of the transition.

Now pair depth with rate. A mountain belt 200 km wide shortens to 100 km across 20 million years — a fractional shortening of 0.5. Twenty million years is about $6.3\cdot 10^{14}$ seconds, so the strain rate is $0.5/(6.3\cdot 10^{14})$, or about $10^{-15}$ per second. Natural geologic strain rates lie between $10^{-13}$ and $10^{-16}$ per second, glacially slow beside a laboratory test at $10^{-6}$ per second. That gap is why a rock that flows in a mountain belt snaps in a lab vise.

What this tells us is that the crust's behavior is set as much by how fast it is loaded as by how hot it is. Temperature fixes where the transition sits; strain rate fixes how the rock answers the load on either side of it.

Check your understanding Beginner

Formal definition Intermediate+

The deformation gradient and finite strain

Let $\mathbf{X}$ be the position of a material point in the undeformed reference configuration and $\mathbf{x}(\mathbf{X},t)$ its position at time $t$. The deformation gradient is $F_{ij}=\partial x_i/\partial X_j$, with Jacobian $J=\det F$ giving the volume change ($J=1$ for incompressible deformation). The polar decomposition $F=RU$ separates a symmetric right stretch tensor $\mathbf{U}$, whose eigenvalues are the principal stretches $S_1\ge S_2\ge S_3$, from a rotation $\mathbf{R}$. A unit sphere in the reference state maps to the strain ellipsoid with semiaxes $S_1,S_2,S_3$, the canonical picture of finite strain. The infinitesimal strain tensor of 27.09.01 is the small-rotation, small-stretch limit; finite strain is required once rotations and stretches are large, as they are in mountain belts.

Rheology: elastic, viscous, and plastic constitutive laws

A constitutive law relates stress to strain or strain rate.

Elastic (Hookean): ${\sigma }_{ij}=C_{ijkl}{\epsilon }_{kl}$, recoverable and independent of strain rate; the isotropic uniaxial form is $\sigma =E\epsilon$ with Young's modulus $E$.

Newtonian viscous: ${\sigma }_{ij}^{\prime }=2\mu {\dot{\epsilon }}_{ij}$, where ${\sigma }_{ij}^{\prime }$ is the deviatoric stress, $\mu$ the viscosity, and ${\dot{\epsilon }}_{ij}$ the strain-rate tensor; the response is linear in rate.

Power-law (dislocation) creep: $\dot{\epsilon }=A{\sigma }^n\exp (-Q/RT)$, with stress exponent $n\approx 3$–$5$ and activation energy $Q$; strongly temperature-dependent and grain-size insensitive.

Diffusion creep: Newtonian ($n=1$) and grain-size-dependent, $\dot{\epsilon }\propto \sigma /d^m$ with $m\approx 2$–$3$, where $d$ is grain size.

Plastic (yield): deformation is zero below a yield stress and accumulates without bound (ideal plasticity) or hardens above it.

Viscoelastic: the Maxwell element (spring and dashpot in series) relaxes stress under fixed strain over a relaxation time ${\tau }_M=\mu /G$, with $G$ the shear modulus; the Kelvin-Voigt element (spring and dashpot in parallel) retards strain under fixed stress.

Deformation mechanisms by depth

At the grain scale, deformation is accommodated by distinct mechanisms, each with its own constitutive law. Cataclastic flow — fracturing and frictional sliding of grains — dominates the brittle upper crust. Pressure solution — dissolution at stressed grain contacts and reprecipitation in less-stressed pores — dominates compacting sediments and low-grade metamorphic rocks. Dislocation glide and climb — motion of line defects through the crystal lattice — drives power-law creep in the hot lower crust and mantle. Diffusion creep — vacancy migration through the grain interior (Nabarro-Herring) or along grain boundaries (Coble) — produces Newtonian, grain-size-sensitive flow in fine-grained rocks. The active mechanism sets the rheology: switching mechanisms, as grain-size reduction does in a shear zone, can drop the viscosity by orders of magnitude.

Failure criteria and fracture mechanics

A failure criterion predicts the stress state at which a rock breaks. The Mohr-Coulomb criterion of 27.09.01, $|\tau |=C+{\sigma }_n\tan \varphi$, governs frictional sliding on pre-existing planes. Brittle tensile fracture is governed instead by the Griffith energy balance: a crack of half-length $a$ propagates when the elastic energy released by crack growth exceeds the surface energy $2\gamma$ needed to create two new surfaces, giving a failure stress ${\sigma }_f=\sqrt{2\gamma E^{\prime }/(\pi a)}$, where $E^{\prime }=E$ in plane stress and $E/(1-{\nu }^2)$ in plane strain. Equivalently the stress intensity factor $K_I=\sigma \sqrt{\pi a}$ reaches the fracture toughness $K_{Ic}$ at failure. The rate-and-state friction law extends Mohr-Coulomb to the sliding surface itself, writing the friction coefficient as a function of sliding velocity and an evolving state variable.

Fold geometry: plunge, vergence, and classification

Beyond the anticline-syncline distinction of 27.09.01, folds carry finer geometric information. The hinge line may be horizontal or tilted; its tilt from horizontal is the plunge. The axial surface may be vertical or inclined; an asymmetric fold overturns toward its steeper limb, and the direction of overturning is the vergence, which records the direction of tectonic transport. Ramsay's classification orders folds by how their layer thickness varies around the hinge, using dip isogons — lines connecting points of equal limb dip on the inner and outer arcs. Class 1B (parallel) folds maintain constant thickness measured perpendicular to the layer and form by flexural slip; Class 2 (similar) folds maintain constant thickness measured parallel to the axial surface and form by passive shear. The interlimb angle grades the tightness from gentle through open and tight to isoclinal.

Key result: the Griffith failure criterion and brittle fracture mechanism Intermediate+

Theorem (Griffith, 1920). A through-crack of half-length $a$ in a thin elastic plate of modulus $E$ carrying a remote tensile stress $\sigma$ perpendicular to the crack grows when

$${\sigma }_f=\sqrt{\frac{2\gamma E}{\pi a}}$$

in plane stress (replace $E$ by $E/(1-{\nu }^2)$ for plane strain), where $\gamma$ is the surface energy per unit area. The failure stress therefore scales as $a^{-1/2}$: a rock with larger flaws is weaker ^{[Griffith 1920]}.

Proof. The total mechanical energy of the loaded plate is the elastic strain energy stored in the body minus the work done by the applied loads at the distant boundaries. As each crack tip extends by $da$ (total new crack length $2da$ per unit thickness), elastic energy is released from the stress-relieved lobes flanking the crack. Using the Inglis elastic solution for an elliptical cavity, the strain-energy release rate per unit of new crack area is

$$G=\frac{\pi {\sigma }^{2}a}{E}$$

in plane stress. Creating fresh surface costs energy $2\gamma$ per unit area (two faces). Crack growth becomes energetically favorable exactly when the released energy meets or exceeds the surface-energy cost:

$$\frac{\pi {\sigma }^{2}a}{E}\ge 2\gamma .$$

Rearranging gives ${\sigma }^2\ge 2\gamma E/(\pi a)$, hence the failure stress ${\sigma }_f=\sqrt{2\gamma E/(\pi a)}$. $■$

The prediction is dramatic. The ideal cohesive strength of a perfect crystal is of order $E/10$, but natural rock fails near $E/1000$. Griffith closes that gap: a millimetre-scale pre-existing crack concentrates stress at its tip and reduces the macroscopic strength by two orders of magnitude. The same scaling explains why rock strength falls with sample size in the lab (larger samples statistically host larger flaws) and why intact wall rock is far stronger than the fractured crust that hosts earthquakes.

Bridge. The Griffith result is the foundational reason that fracture mechanics scales identically from a micron-scale crack to a kilometre-scale fault: doubling the flaw size quarters the strength, and this is exactly the size effect that leaves the intact interior of a plate far stronger than the shattered crust above it. It builds toward the rate-and-state friction law of the Master tier, where the same crack-tip energy balance reappears as the fracture energy that governs stick-slip instability, and it appears again in the earthquake-source mechanics of 27.03.01, where stress drop and rupture area set the seismic moment. Putting these together, the bridge is an energy accounting: every crack that grows pays a surface-energy tax, and the budget decides whether the crust creeps, folds, or shatters.

Exercises Intermediate+

Lean formalization Intermediate+

There is no Lean formalization for this unit (lean_status: none). Deformation mechanics is a branch of continuum mechanics coupled to empirical rheology, and the objects it manipulates are not present in Mathlib at the level this field requires. A faithful formalization would need, at minimum, the deformation gradient as a smooth map between manifolds with its polar decomposition into a rotation and a symmetric stretch tensor, the principal-stretch eigenstructure of finite strain, the Maxwell and Kelvin-Voigt viscoelastic ordinary differential equations, the Griffith energy functional over crack surfaces, and the rate-and-state friction law with its evolving internal-state variable.

Mathlib has linear-algebra and tensor vocabulary, and its differential-geometry layer is maturing, but it has no continuum-mechanics hierarchy, no physical-units layer tied to the SI pascal, and no constitutive-law library. The Griffith energy-balance derivation is essentially a variational inequality (energy release rate meets surface-energy cost), formalizable in principle once a continuum layer lands; the Flinn classification reduces to the eigenstructure of a general $3\times 3$ deformation tensor and is within reach of Mathlib's existing linear algebra. Until then the content is reviewed as quantitative prose, and the gap is tracked so that a future physics-continuum initiative can pick it up without re-deriving the motivating examples.

Advanced results Master

Rate-and-state friction and the seismic cycle

Coulomb friction is too crude to distinguish a creeping fault from one that quakes. The rate- and state-dependent friction law of Dieterich and Ruina writes the coefficient of friction as

$$\mu ={\mu }_0+a\ln \left(\frac{V}{V_0}\right)+b\ln \left(\frac{\theta V_0}{D_c}\right),$$

where $V$ is sliding velocity, $\theta$ an evolving state variable (interprettable as the average contact lifetime of asperities), $D_c$ a critical sliding distance, and $a$, $b$ dimensionless constitutive parameters ^{[Dieterich 1979]}. The state variable relaxes according to $d\theta /dt=1-V\theta /D_c$ (the aging law).

The sign of $a-b$ decides stability. Where $a-b>0$ the fault is velocity-strengthening: faster slip raises friction, slip is stable, and the fault creeps aseismically. Where $a-b<0$ the fault is velocity-weakening: faster slip lowers friction, and a small perturbation runs away into stick-slip, the laboratory analogue of an earthquake. Brace and Byerlee proposed stick-slip as the earthquake mechanism in 1966 ^{[Brace and Byerlee 1966]}; rate-and-state friction supplies the constitutive law that predicts when it occurs. The critical spring stiffness $k_c=(b-a)({\sigma }_n-P_f)/D_c$ separates stable sliding ($k>k_c$) from stick-slip ($k<k_c$), unifying lab sliding experiments with the seismic cycle on natural faults.

Deformation-mechanism maps and the lithospheric strength envelope

Frost and Ashby compiled deformation-mechanism maps plotting normalized stress $\sigma /\mu$ against homologous temperature $T/T_m$ (melting temperature), with contours of constant strain rate ^{[Frost and Ashby 1982]}. Each map is partitioned into fields — elastic, dislocation glide, power-law creep, diffusional flow (Nabarro-Herring at high temperature, Coble along grain boundaries) — in each of which one constitutive law dominates. The active mechanism at a given stress, temperature, and strain rate is read directly off the map, and the rheology follows from it.

Stacking these mechanisms with depth yields the lithospheric strength envelope. In the brittle upper crust the differential stress is bounded by Byerlee friction and rises with depth; at the brittle-ductile transition it gives way to the flow stress of power-law creep, which falls with depth as temperature rises (the creep law's Arrhenius factor weakens the rock faster than pressure strengthens it). The resulting curve — strong near the surface, a peak at the transition, weaker below — is the composite strength profile of the lithosphere ^{[Kohlstedt et al 1995]}. Whether the lower crust or the mantle is the weaker layer is the long-standing "jelly sandwich" versus "crème brûlée" debate; it turns on composition, water content, and geotherm, and it governs how mountain belts founder and how plates stay rigid.

Fold kinematics and the Ramsay classification

Ramsay's dip-isogon classification is more than taxonomy: each class is the geometric signature of a kinematic mechanism. Class 1B (parallel) folds, with constant orthogonal thickness, form by flexural slip — competent layers slide past one another, each bed maintaining its thickness, and the strain within each bed is plane. Class 1A and 1C folds, with isogons converging inward or outward, record superposed flattening or initial buckling. Class 2 (similar) folds, with constant axial-trace thickness and isogons converging to a point, form by passive shear: the layer is simply a passive marker carried by a homogeneous strain field, as in a shear zone. Class 3 folds mark the interior of a sheared multilayer where one lithology concentrates the strain.

The classification reads finite strain from fold shape. Plunge and vergence locate the fold in three dimensions; the interlimb angle grades the strain intensity; and the Ramsay class identifies the kinematic regime. A field geologist mapping a fold train is therefore recovering the strain path, the transport direction, and the competence contrast of the original sedimentary pile at once ^{[Ramsay 1967]}.

Finite strain and the Flinn diagram

The strain ellipsoid, with principal stretches $S_1\ge S_2\ge S_3$ and $S_1{S}_2{S}_3=1$, is summarized by the Flinn parameter $k=(S_1/S_2-1)/(S_2/S_3-1)$ and the strain intensity $d=\sqrt{(S_1/S_2-1{)}^2+(S_2/S_3-1{)}^2}$. Constrictional (prolate) strain, in which extension outpaces shortening along one axis ($k>1$), produces rod-shaped fabrics; flattening (oblate) strain ($k<1$) produces pancake-shaped fabrics; plane strain ($k=1$) is the boundary. A coaxial deformation (pure shear, in which the principal stretch axes do not rotate) traces a path on the Flinn diagram distinct from a non-coaxial simple shear, even when the final ellipsoid is identical; the vorticity number $W_k$ ($0$ for pure shear, $1$ for simple shear) measures the rotational component that the ellipsoid alone cannot recover. Together, $k$, $d$, and $W_k$ invert the deformed shapes of originally spherical markers — ooids, amygdales, conglomerate clasts — to reconstruct the finite strain and flow regime of an orogen.

Synthesis. Putting these together — Griffith fracture, rate-and-state friction, the power-law and diffusion creep laws, and the Maxwell relaxation time — yields the central insight that a single rock samples every rheological regime as temperature, pressure, strain rate, and time vary, so deformation is not a property of the rock but of the conditions. The foundational reason is that the Maxwell relaxation time sets the elastic-to-viscous crossover and shifts by orders of magnitude with depth and temperature; this is exactly why the lithosphere is strong and the asthenosphere weak, and why the same basalt fractures in an earthquake but flows in a mantle convection cell. The framework generalises to other planets, where water content and geotherm reset the whole strength envelope, and it is dual to the fold-geometry analysis: mechanics predicts how a layer will fail or flow, and the preserved fold and fault geometry inverts the strain to recover the conditions. The bridge is that stress and strain are two faces of one constitutive relation, read across timescales from the seismic second to the orogenic aeon, and it appears again in the tectonic engine of 27.01.01 and the seismic cycle of 27.03.01.

Full proof set Master

Proposition (Maxwell relaxation time). Let a linear Maxwell element consist of a Hookean spring of shear modulus $G$ in series with a Newtonian dashpot of viscosity $\mu$, loaded in simple shear by a stress $\tau (t)$. The total shear strain $\gamma$ combines the spring strain ${\gamma }_e$ and the dashpot strain ${\gamma }_v$ and obeys

$$\dot{\gamma }(t)=\frac{1}{G}\dot{\tau }(t)+\frac{\tau (t)}{\mu }.$$

Under a step stress ${\tau }_0$ applied at $t=0$ and held constant, the strain is

$$\gamma (t)=\frac{{\tau }_0}{G}+\frac{{\tau }_0}{\mu }t,$$

and the characteristic time on which the elastic and viscous contributions are equal is the Maxwell relaxation time ${\tau }_M=\mu /G$. For loading times $t\ll {\tau }_M$ the element responds elastically; for $t\gg {\tau }_M$ it flows viscously.

Proof. In a series connection $\gamma ={\gamma }_e+{\gamma }_v$, hence $\dot{\gamma }={\dot{\gamma }}_e+{\dot{\gamma }}_v$. The spring obeys ${\gamma }_e=\tau /G$, giving ${\dot{\gamma }}_e=\dot{\tau }/G$. The dashpot obeys ${\dot{\gamma }}_v=\tau /\mu$. Adding these yields the constitutive equation $\dot{\gamma }=\dot{\tau }/G+\tau /\mu$. For a step stress $\tau (t)={\tau }_0\Theta (t)$, $\dot{\tau }=0$ for $t>0$, so $\dot{\gamma }={\tau }_0/\mu$, which integrates to $\gamma (t)=\gamma (0)+({\tau }_0/\mu )t$. The instantaneous jump at $t=0$ is the elastic strain $\gamma (0)={\tau }_0/G$, giving $\gamma (t)={\tau }_0/G+({\tau }_0/\mu )t$. The crossover when the two terms are equal, ${\tau }_0/G=({\tau }_0/\mu )t$, gives $t=\mu /G\equiv {\tau }_M$. $■$

Numerically, the asthenospheric upper mantle with $\mu \approx 10^{20}$ Pa·s and $G\approx 5\times 10^{10}$ Pa has ${\tau }_M\approx 2\times 10^9$ s, about 60 years: short on the convection timescale (so it flows) and enormous on the seismic-wave timescale (so it is elastic). Dry cratonic lithosphere with $\mu \approx 10^{23}$–$10^{24}$ Pa·s has ${\tau }_M$ of $10^{12}$–$10^{13}$ s, comparable to or longer than orogenic timescales, so it transmits stress elastically over hundreds of millions of years.

Proposition (Flinn parameter and the plane-strain condition). For a volume-conserving finite strain with principal stretches $S_1\ge S_2\ge S_3$ and $S_1{S}_2{S}_3=1$, the Flinn parameter $k=(S_1/S_2-1)/(S_2/S_3-1)$ equals $1$ if and only if $S_2=1$ (plane strain). It exceeds $1$ on constrictional (prolate) paths and lies in $[0,1)$ on flattening (oblate) paths.

Proof. Forward direction: suppose $S_2=1$. Volume conservation gives $S_3=1/S_1$. Then $S_1/S_2-1=S_1-1$ and $S_2/S_3-1=1/(1/S_1)-1=S_1-1$, so $k=1$. Reverse direction: suppose $k=1$, so $S_1/S_2-1=S_2/S_3-1$, i.e. $S_1/S_2=S_2/S_3$. Cross-multiplying gives $S_1{S}_3=S_2^2$. Substituting volume conservation $S_1{S}_2{S}_3=1$ gives $(S_1{S}_3)S_2=S_2^2\cdot S_2=S_2^3=1$, hence $S_2=1$. On constrictional paths the numerator grows faster than the denominator (the long-axis stretch $S_1$ outpaces the intermediate), giving $k>1$; on flattening paths the denominator dominates, giving $0\le k<1$, with $k=0$ at uniaxial flattening ($S_1=S_2$). $■$

The boundary $k=1$ is therefore the locus of plane-strain deformations, which includes simple shear — the kinematic model of a ductile shear zone — and pure shear shortened along one axis with no intermediate stretch. Measuring $k$ from deformed markers thus separates constrictional tectonics (transpression, extrusion) from flattening tectonics (thrust stacking, gravitationally spreading orogens).

Connections Master

• The stress and strain tensors, Anderson's fault classification, and the Mohr-Coulomb criterion on which this unit builds are all developed in the companion unit 27.09.01. This unit assumes that vocabulary and deepens it: where 27.09.01 catalogues the structures a stress field produces, here we ask by which mechanism the rock arrived at that structure and what its geometry records about the conditions. The two units are most useful read in sequence.

• The rate-and-state friction law derived here is the constitutive input to earthquake source theory in 27.03.01. The sign of $a-b$, the critical stiffness $k_c$, and the fracture energy that Griffith balances against surface energy all reappear as the parameters controlling nucleation, rupture speed, and seismic moment. Earthquake seismology and rock friction are two readings of the same sliding surface.

• Deformation and metamorphism are coeval: the temperature, pressure, and fluid conditions that fold and fault a rock also drive the mineral reactions catalogued in 27.02.03 pending. A mylonitic foliation in a shear zone is simultaneously a finite-strain gauge and a metamorphic index of peak temperature, and the grain-size reduction that switches the deformation mechanism from dislocation to diffusion creep is itself a metamorphic recrystallization event.

• The far-field stresses and the thermal structure that set the rheology come from the plate-boundary forces of 27.01.01. Ridge push and slab pull set the principal stresses that Anderson converts into fault regimes; the geotherm and the plate's age set the depth of the brittle-ductile transition and the viscosity of the asthenosphere that lets the plate move. Plate tectonics supplies the engine; this unit describes how the engine's forces are absorbed by the rock.

• The deformation styles analyzed here are the building blocks of the orogenic cycles traced in 27.08.01. Each Wilson-cycle closure stacks thrust sheets, folds sediments, and exhumates shear zones that this unit's tools can read; the resulting mountain belt is then eroded into the basin fill of the next cycle. Structural geology thereby turns the rock record into a quantitative history of planetary stress through deep time.

Historical & philosophical context Master

The theoretical strength of a perfect crystal, set by the work needed to pull two atomic planes apart, is of order $E/10$. Real rock fails at roughly a thousandth of that value, a discrepancy that frustrated engineers for a generation. A. A. Griffith resolved it in 1920 by postulating that every solid contains submicroscopic cracks, each of which concentrates stress at its tip. Balancing the elastic energy released by crack growth against the surface energy of the new faces created, he derived a failure stress scaling as $a^{-1/2}$ and showed that millimetre-scale flaws quantitatively account for the weakness of glass and metal ^{[Griffith 1920]}. The paper founded fracture mechanics and reframed strength as a property of defect populations rather than of ideal matter.

The rheological half of the subject descends from James Clerk Maxwell. In his 1867 dynamical theory of gases he introduced the idea that a material's response to stress depends on the timescale of loading: at short times it stores energy elastically, at long times it dissipates energy by flow. The Maxwell element — a spring and dashpot in series — and its relaxation time ${\tau }_M=\mu /G$ became the prototype of every later viscoelastic model and remain the conceptual bridge between the elastic seismology of the mantle and its viscous convection ^{[Maxwell 1867]}.

Quantitative fold geometry arrived with John Ramsay's 1967 Folding and Fracturing of Rocks. By defining dip isogons and the five-fold classification, Ramsay turned field mapping from description into measurement: the shape of a fold became a record of its kinematic mechanism and of the finite strain it had absorbed. His methods remain the working language of structural analysis and underpin the modern inversion of deformed markers for strain and vorticity ^{[Ramsay 1967]}.

The frictional and creep halves of rheology converged in the late twentieth century. Brace and Byerlee in 1966 proposed that earthquakes are stick-slip instabilities on faults rather than continuous slip ^{[Brace and Byerlee 1966]}; Dieterich in 1979 gave the phenomenon its constitutive law, rate-and-state friction ^{[Dieterich 1979]}; Frost and Ashby in 1982 compiled the deformation-mechanism maps that place each mechanism in its stress-temperature field ^{[Frost and Ashby 1982]}. The synthesis of these threads — a brittle upper crust governed by Griffith and Byerlee friction, a ductile lower crust and mantle governed by power-law creep, joined at a relaxation-time boundary — is the strength envelope against which every modern tectonic model is tested.

Bibliography Master

1. Griffith, A. A. (1920). The phenomena of rupture and flow in solids. Philosophical Transactions of the Royal Society of London, Series A, 221, 163–198.

2. Maxwell, J. C. (1867). On the dynamical theory of gases. Philosophical Transactions of the Royal Society of London, 157, 49–88.

3. Brace, W. F., and Byerlee, J. D. (1966). Stick-slip as a mechanism for earthquakes. Science, 153, 990–992.

4. Dieterich, J. H. (1979). Modeling of rock friction: 1. Experimental results and constitutive equations. Journal of Geophysical Research, 84, 2161–2168.

5. Ramsay, J. G. (1967). Folding and Fracturing of Rocks. McGraw-Hill, New York.

6. Frost, H. J., and Ashby, M. F. (1982). Deformation-Mechanism Maps: The Plasticity and Creep of Metals and Ceramics. Pergamon Press, Oxford.

7. Kohlstedt, D. L., Evans, B., and Mackwell, S. J. (1995). Strength of the lithosphere: Constraints from laboratory experiments and the rheology of deformation mechanisms. Journal of Geophysical Research, 100, 17587–17602.

8. Turcotte, D. L., and Schubert, G. (2014). Geodynamics (4th ed.). Cambridge University Press, Cambridge.

9. Karato, S.-I. (2008). Deformation of Earth Materials: An Introduction to the Rheology of the Solid Earth. Cambridge University Press, Cambridge.

10. Scholz, C. H. (2019). The Mechanics of Earthquakes and Faulting (3rd ed.). Cambridge University Press, Cambridge.