Transepidermal water loss is a Fickian flux through the stratum-corneum barrier: it crosses twice baseline when thickness halves (here at 7 of 15 layers), then collapses discontinuously when a chemical insult drives the keratinocyte switch past its spinodal. The absolute insult a layer tolerates scales as γ1.5, so keratin is the most insult-resistant.
TEWL rises as 1/N with barrier thickness, crossing twice baseline at half thickness, then collapses discontinuously when an insult drives the keratinocyte switch past its spinodal. Insult tolerance scales as γ1.5, ranking keratinocyte highest.
Water loss is a Fickian flux through a living barrier
The stratum corneum is a diffusion barrier; transepidermal water loss (TEWL) is the steady flux across it. For a barrier of N layers the flux scales as 1/N, so thinning the barrier raises TEWL smoothly and monotonically.
On a 15-layer reference barrier the flux crosses twice its baseline once thickness halves — here at 7 layers. This is the ordinary, graded failure mode: gradual barrier loss, gradual water-loss rise.
Insult is different: a discontinuous collapse at the spinodal
A chemical insult does not thin the barrier; it drives the keratinocyte switch field h toward its spinodal. The barrier integrity (the depth of the ON basin) is robust until the drive reaches the spinodal — up to a fraction of about 1.0033 of it — and then collapses discontinuously, TEWL jumping rather than drifting.
The absolute insult a layer can absorb before that snap scales as γ1.5, so the four organs rank in resistance as KRT14 > MITF > EDAR > TP63: the stiff keratinocyte tolerates the most, the epidermal field the least. The keratinocyte absolute tolerance here is 0.70 (dimensionless drive units).
What is verified and what is open
The two regimes — smooth threshold under thinning, discontinuous collapse under insult — and the γ1.5 resistance ordering are simulation-verified shapes [V] Simulation-verified. The conversion to an absolute clinical TEWL in g·m−2·h−1 is open [O] Open (obstacle stated): it needs the stratum-corneum lipid permeability and the water activity gradient as cited inputs.