aragog.surface_lid
The aragog.surface_lid module holds the sub-grid surface lid: a conductive crust thinner than the top cell, carried as its thickness and its energy per area, with the temperature profile of the one-phase Stefan problem. The functions are pure; the solver evaluates the material properties and calls them in every right-hand side when subgrid_lid is on.
| Name | Role |
|---|---|
shape |
Slopes at the top and the base and the mean of the profile \(f_s\) for the shape parameter \(s\). |
surface_state |
Surface temperature, surface flux and base conduction of a lid under a grey-body surface. |
fixed_surface_state |
The same for a fixed surface temperature. |
effective_thickness |
Thickness of the linear lid with the same surface gradient. |
convective_base_flux |
Boundary-layer flux of the melt under the lid at the critical Rayleigh number. |
stefan_rate |
Growth rate of the lid base, with the growth stop and the melt-back stop. |
existence_weight |
Weight of the lid from the thickness of a steady crust. |
lid_rates |
Rates of the two lid states, the surface loss and the flux out of the melt. |
For the physics, see Surface lid.
surface_lid
Sub-grid conductive lid at the surface of the top cell.
A crust thinner than the top cell cannot be resolved by the mesh. The lid model carries it as
two states per unit area: the thickness delta [m] and the energy E_lid [J/m^2]. The
temperature profile across the lid is taken from the similarity family of the one-phase Stefan
problem (Neumann solution),
T(z) = T_s + (T_b - T_s) f_s(z / delta),
f_s(x) = erf(sqrt(s) x) / erf(sqrt(s)) (s > 0),
x (s = 0),
erfi(sqrt(-s) x) / erfi(sqrt(-s)) (s < 0),
with the shape parameter s and the surface temperature T_s fixed by the lid energy,
E_lid = delta int_0^1 H(T(x)) dx with H the heat content per volume of the lid material
(:class:LidEnthalpy), and by the grey-body balance at the surface. The family holds the
Neumann growth exactly (s = lambda^2) and reduces to the linear profile in steady state.
Material joins the lid at its base by the Stefan condition (W(S_f) - W(S_b)) d(delta)/dt =
q_b - F_b, where q_b is the conduction into the lid at its base, F_b the flux the
fluid below delivers and W(S) = int rho T dS the heat content per volume at the top-cell
pressure, at the fluid entropy S_f and the lid-base entropy S_b. The reported heat
content of the mesh uses this measure; the lid holds it as a function of temperature, the same
measure wherever the temperature rises with entropy (not inside a band of constant temperature
below the solidus).
The functions here are pure: the solver evaluates the material properties, passes them in and assembles the rates. All quantities are SI (K, m, W/m^2, J/kg, J/m^2, J/m^3, kg/m^3, s).
References
Turcotte, D. L. and Schubert, G. (2014), Geodynamics, 3rd ed., Cambridge University Press: section 4.18, the Stefan problem (eqs. 4.136 and 4.141, pp. 197-198), and section 4.15, the cooling of a half-space (eq. 4.116, p. 189).
ConstantProperties(T_ref, S_ref, rho, cp)
dataclass
The entropy-EOS calls of the lid for a constant-property column
(phase_mixed.const_properties): T = T_ref exp((S - S_ref) / cp) [K] with constant
density [kg/m^3] and heat capacity [J/kg/K], and no melt. The pressure is not used.
LidEnthalpy(H, rho, data=())
dataclass
Heat content per volume H(T) [J/m^3] and density rho(T) [kg/m^3] of the lid
material at the top-cell pressure, both vectorised in T [K].
first_knot
property
Lowest temperature [K] of the table behind H; -inf for constant properties.
linear(rho, cp, T_b, h_b)
classmethod
Constant properties: H(T) = rho (h_b + cp (T - T_b)).
Source code in src/aragog/surface_lid.py
192 193 194 195 196 197 198 199 | |
profile_density(s, T_s, T_b)
Mean density [kg/m^3] of a lid with shape s from T_s to T_b.
Source code in src/aragog/surface_lid.py
228 229 230 231 | |
profile_heat(s, T_s, T_b)
Mean heat content [J/m^3] of a lid with shape s from T_s to T_b.
Source code in src/aragog/surface_lid.py
223 224 225 226 | |
table(T, H, rho)
classmethod
From a table on increasing T: monotone cubic H, linear rho, both
continued outside the table with the end slopes of H and constant rho.
Source code in src/aragog/surface_lid.py
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 | |
LidRates(ddelta, dE, F_top, F_fluid, w, stop, H_lid)
dataclass
Rates and fluxes of the lid at one state.
Attributes:
| Name | Type | Description |
|---|---|---|
ddelta |
float
|
d(delta)/dt [m/s]. |
dE |
float
|
dE_lid/dt [W/m^2], internal heating of the lid included. |
F_top |
float
|
Flux the planet loses at the surface [W/m^2]. |
F_fluid |
float
|
Flux out of the fluid at the top of the top cell [W/m^2], from energy conservation. |
w |
float
|
Lid weight [0, 1]. |
stop |
float
|
Factor of the growth or melt-back stop [0, 1]. |
H_lid |
float
|
Internal heating of the lid [W/m^2]. |
LidSurface(T_s, F_s, q_b, s, clipped, T_m)
dataclass
Surface solution of the lid at one state.
Attributes:
| Name | Type | Description |
|---|---|---|
T_s |
float
|
Surface temperature [K]. |
F_s |
float
|
Surface flux [W/m^2], positive out of the planet. |
q_b |
float
|
Conduction into the lid at its base [W/m^2], positive upward. |
s |
float
|
Shape parameter of the profile [-]. |
clipped |
int
|
0 when the root lies inside the shape bounds; +1 when the lid energy is too high for any
profile ( |
T_m |
float
|
Mean temperature of the profile [K]. |
convective_base_flux(dT, k, rho, cp, alpha, g, eta, ra_crit)
Boundary-layer flux [W/m^2] of a fluid dT [K] hotter than the lid base.
q = k dT^(4/3) (alpha rho g / (Ra_crit kappa eta))^(1/3) with kappa = k / (rho cp):
the flux of a boundary layer at the critical Rayleigh number. Zero for dT <= 0.
Source code in src/aragog/surface_lid.py
462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 | |
effective_thickness(delta, s)
Thickness [m] of the linear lid with the surface gradient of a lid of thickness delta
and shape s: delta / f'(0). NaN at the lower shape bound, where f'(0) is 0.
Source code in src/aragog/surface_lid.py
234 235 236 237 238 | |
existence_weight(F_b, T_b, k, emissivity, T_eq, delta_on, T_surf=None)
Weight [0, 1] of the lid from the quasi-steady crust thickness.
A crust in steady state carries F_b by conduction, delta_eq = k (T_b - T_s) / F_b,
with grey(T_s) = F_b, or T_s = T_surf [K] at a fixed surface temperature. The lid
acts where delta_eq exceeds delta_on:
w = (1 + tanh(log10(delta_eq / delta_on) / LID_EXIST_WIDTH)) / 2, set to 0 below
LID_WEIGHT_FLOOR. A fluid that delivers over the grey flux at T_b, or a base not
hotter than T_surf, carries no lid.
Source code in src/aragog/surface_lid.py
538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 | |
fixed_surface_state(delta, E, T_b, k, T_s, delta_floor, hv)
Lid with a fixed surface temperature T_s [K]: the shape follows from
Q(s) = E / delta at that surface temperature, held inside the shape bounds.
Parameters and the result as :func:surface_state.
Source code in src/aragog/surface_lid.py
431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 | |
floor_flux(phi_integral, h_top, delta)
Steady flux [W/m^2] of the eddy-diffusivity floor from the fluid to the lid base.
F = Phi / (0.5 (h_top - delta)), with
Phi(S) = int_{S_sol}^{S} rho T kappa_floor f(phi) dS' [W/m], the Kirchhoff integral of
the floor over entropy (f its melt-fraction weight), across half of the fluid part of
the top cell. Lengths in m.
Source code in src/aragog/surface_lid.py
494 495 496 497 498 499 500 501 502 | |
floor_weight(S_f, S_sol)
Weight [0, 1] of the base flux above the top-cell floor of the mush base.
The quintic ramp in x = (S_f - S_sol - LID_FLOOR_GAP) / LID_FLOOR_WIDTH, entropies in
J/kg/K: 0 at and below the floor S_sol + LID_FLOOR_GAP, 1 from LID_FLOOR_WIDTH above
it, and C2 at both ends.
Source code in src/aragog/surface_lid.py
483 484 485 486 487 488 489 490 491 | |
flux_surface_state(delta, E, T_b, k, F_atm, delta_floor, hv)
Lid under a prescribed surface flux F_atm [W/m^2]: for a shape s the conduction
k (T_b - T_s) f'(0) / d = F_atm fixes the surface temperature, at least 1 K, and with it
the lid heat content, so Q(s) = E / delta gives the shape. At 1 K the lid carries less
than F_atm. The surface stays LID_DT_MIN below T_b, so that the profile keeps a
temperature drop; :func:flux_weight gives such a lid no weight for F_atm <= 0.
Parameters and the result as :func:surface_state, with F_atm for the grey body.
Source code in src/aragog/surface_lid.py
369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 | |
flux_weight(F_b, T_b, T_s, k, emissivity, T_eq, delta_on)
Weight [0, 1] of the lid under the surface T_s of the atmosphere.
It is :func:existence_weight of the steady crust under that surface,
delta_eq = k (T_b - T_s) / F_b: 0 for T_s at or above T_b, 1 for F_b <= 0.
For F_b > 0 the weight falls to 0 as T_s rises to T_b, so the lid has no step at
the onset; for F_b <= 0 it steps from 1 to 0 at T_s = T_b.
Source code in src/aragog/surface_lid.py
572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 | |
grey_flux(T_s, emissivity, T_eq)
Grey-body flux [W/m^2] of a surface at T_s [K] against T_eq [K].
Source code in src/aragog/surface_lid.py
262 263 264 | |
grey_surface_temperature(f0, d, T_b, k, emissivity, T_eq)
Surface temperature [K] where grey(T_s) = k (T_b - T_s) f0 / d.
Newton from T_b: the residual is convex and increasing in T_s and non-negative at
T_b, so the iterates fall monotonically to the root. T_b when T_b <= T_eq.
Source code in src/aragog/surface_lid.py
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 | |
lid_rates(*, delta, E_lid, surf, F_b, F_grid, w, rho, H_f, T_b, H_b, hv, h_top, delta_lo, heating)
Assemble the lid rates and the fluid flux at one state.
Fluid of heat content H_f [J/m^3] per volume joins the lid at T_b with heat content
H_b [J/m^3] per volume, so the Stefan rate releases H_f - H_b per volume; rho is
the density at the lid base, which scales the floor of that jump. The lid acts with weight
w; the hidden part 1 - w relaxes to the phantom lid (thickness delta_lo, linear
profile) over LID_TAU_HIDDEN, and the surface then sees the grid flux F_grid. The
fluid flux follows from the conservation of the energy of the fluid and the lid:
F_fluid = F_top + dE - H_lid - H_f d(delta)/dt. Exact for any w and any stop.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
heating
|
float
|
Internal heating per mass [W/kg] of the top cell, applied to the lid mass. |
required |
Source code in src/aragog/surface_lid.py
622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 | |
linear_lid_surface(delta, T_b, k, emissivity, T_eq)
Surface temperature [K] of a steady lid: grey(T_s) = k (T_b - T_s) / delta.
Source code in src/aragog/surface_lid.py
451 452 453 454 455 456 457 458 459 | |
mush_base_flux(phi_integral, h_top, delta, F_in, g, q_conv)
Flux [W/m^2] from the mush into the lid base: the floor flux plus the boundary-layer
flux q_conv, blended by the floor weight g to the flux from below F_in, given
per unit area of the top face of the cell, so that g = 0 holds the fluid entropy.
Source code in src/aragog/surface_lid.py
505 506 507 508 509 510 511 | |
profile(s, x)
The similarity profile f_s(x) on [0, 1] (series below |s| = 1e-6).
Source code in src/aragog/surface_lid.py
113 114 115 116 117 118 119 120 121 122 | |
profile_nodes(s)
Quadrature nodes and weights on [0, 1] for integrals over the profile f_s.
Two segments split at c(s), which follows the steep part of the profile:
min(1/2, 6 / sqrt(s)) for s > 0 (erf layer at the surface) and
1 - min(1/2, 15 / |s|) for s < 0 (exponential layer at the base); c is
continuous in s. Each segment holds PROFILE_PANELS 16-point Gauss-Legendre panels:
the table enthalpy has kinks at the EOS nodes, so the panels, not the order, set the error
(about 3e-6 relative at worst on the test tables).
Source code in src/aragog/surface_lid.py
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 | |
robin_surface_state(delta, E, T_b, k, C, Fp, T_held, delta_floor, hv)
Lid under the Robin flux C + Fp (T_s - T_held): for a shape s the conduction
k (T_b - T_s) f'(0) / d equals that flux, which fixes the surface temperature (in the
bounds of :func:flux_surface_state), and Q(s) = E / delta gives the shape. C is the
cap [W/m^2], Fp the slope of the atmosphere flux [W/m^2/K] and T_held the held surface
[K]. Fp = 0 is :func:flux_surface_state at C.
Parameters and the result as :func:surface_state.
Source code in src/aragog/surface_lid.py
397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 | |
shape(s)
Slopes at the top and the base, and the mean, of the similarity profile f_s.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
s
|
float
|
Shape parameter [-]; |
required |
Returns:
| Type | Description |
|---|---|
tuple of float
|
|
Source code in src/aragog/surface_lid.py
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 | |
stefan_rate(q_b, F_b, rho, dh, delta, h_top, delta_lo)
Base growth rate [m/s] of the lid and the factor of the stop that acts.
r = (q_b - F_b) / (rho max(dh, LID_DH_MIN)). Growth stops linearly over
[LID_CAP_FRAC - LID_CAP_WIDTH, LID_CAP_FRAC] h_top; melt-back stops linearly over
[delta_lo, 2 delta_lo].
Returns:
| Type | Description |
|---|---|
tuple of float
|
|
Source code in src/aragog/surface_lid.py
514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 | |
surface_state(delta, E, T_b, k, emissivity, T_eq, delta_floor, hv)
Solve the grey-body surface of a lid with energy E [J/m^2].
For a shape s the grey-body balance grey(T_s) = k (T_b - T_s) f'(0) / d fixes the
surface temperature, and with it the lid heat content Q(s) = int_0^1 H(T(x)) dx. Q
rises with s, so Q(s) = E / delta has one root in the shape bounds.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
delta
|
float
|
Lid thickness [m]. |
required |
E
|
float
|
Lid energy [J/m^2]. |
required |
T_b
|
float
|
Lid-base temperature [K]. |
required |
k
|
float
|
Lid thermal conductivity [W/m/K]. |
required |
emissivity
|
float
|
Surface emissivity [-]. |
required |
T_eq
|
float
|
Equilibrium temperature of the radiation balance [K]. |
required |
delta_floor
|
float
|
Conduction floor [m]: the flux uses |
required |
hv
|
LidEnthalpy
|
Heat content and density of the lid material. |
required |
Returns:
| Type | Description |
|---|---|
LidSurface
|
|
Source code in src/aragog/surface_lid.py
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 | |