
ID	0331
Class	BUG
Severity	1
Status	blocked

Models: RbCable/Cable, TrawlCable, MooringCable, RundDorg

Found: 2026-10-01, while checking the constant of MARE-0322 against reference data.

Decision needed: Owner: move the Re = 14 branch edge, or refit the first two branches, so that Cd is continuous. A source for the correlation is also still needed.

Evidence

src/RMCable/Cable.h:186-193 (ElementDragCoefficients) and src/trawl_mooring_interaction/CableSegmentWaterForce.h:89-96 use three branches: 1.3 + 8.85/(Re^0.85 + c) for Re < 14, 0.75 + 2.35/(Re^0.38 - 1.67) for Re < 2515 and 1.15 - 5.05/(Re^0.5 - 31.6) for Re < 1e6. At Re = 14 the first gives 2.24 (c = 0.01646) or 2.22 (c = 0.1646) and the second 2.98, a step of about 0.75. At Re = 2515 the second and third agree (0.881 against 0.878). Against the digitised Wiley cylinder-drag curve (github.com/peterdsharpe/CylinderDragFits, Fig_from_incompressible_flow_wiley.csv, from White/Wiley Incompressible Flow, a hand-digitised figure, so noisy) the second branch is within about 10% in log terms for 14 < Re < 2515, but the first branch is too low from Re of about 5 to 14 (2.99 against about 3.8 at Re = 7).

Effect

A cable whose Reynolds number crosses 14 sees its normal drag change by about 30% at once. Only very slow or thin cables are affected.

Possible fix

Refit the first branch so that it meets the second at the branch edge, or move the edge to where the two curves cross (between Re = 20 and 50 they are within 10%). Either changes results at Re < 50. The constant itself is settled by MARE-0322.

Test that would prove it

A unit test of the shared drag-coefficient function that Cd is continuous across each branch edge.

Risk

Low: a result change for cables at Re < 50, in both model families.