Floating plastic is pushed down by breaking waves and rises back at velocity w_r, so its concentration decays exponentially with depth (Kukulka et al. 2012; Reisser et al. 2015):
c(z) = c0 exp(-z / L), L = K / w_r
K = 0.4 u*w Hs (vertical eddy diffusivity)
u*w = sqrt(rho_a Cd / rho_w) U10 = 1.25e-3 U10
Hs = 0.0214 U10^2 (fully developed sea)
A barrier with a skirt of depth d therefore intercepts the fraction 1 - exp(-d/L) of the plastic in the water column. A passive U-shaped barrier moves slower than the water because of a sea anchor; the relative speed is v_rel = a U_c with U_c = 0.035 U10 + 0.05 m/s. Plastic spills over when v_rel is too large: retention r = max(0, 1 - (v_rel/0.8)^2). Surface particles also random-walk horizontally with D_h = 0.1 + 0.02 U10 m2/s.
efficiency = r * (1 - exp(-d/L)) (captured / particles that reached the barrier)
Honest scope. A teaching model of one passive U-shaped system: constants (0.8 m/s overflow speed, 3.5 % wind drift, 0.4 mixing factor) are order-of-magnitude values, not engineering data. No Langmuir circulation, fouling, bycatch, or real fleet logistics. Real cleanup also needs source reduction; barriers alone cannot clear the ocean.