Domesday horsepower

Claim (verbatim)

Domesday horsepower. Joins stream-power hydrology to the fiscal cadastre: the 1086 Domesday survey unwittingly recorded a physics meter, because a mill's render capitalized the hydraulic power of its site; feudal rent should therefore scale with drainage-area-times-slope like an engineering formula.

Prediction clause (verbatim)

Matching >=2,000 Domesday mill renders to modern DEM-derived stream power at the recorded vill (discharge proxy times channel slope), log render regresses on log stream power with elasticity between 0.4 and 0.8 and within-county R^2 >= 0.3; multiple mills on one manor split their render in proportion to available head; and render-per-unit-power is systematically 30-80% higher in grain-rich arable counties than in pastoral uplands, reflecting demand rather than physics.

Kill-dataset (verbatim)

Kill: Open Domesday mill renders joined to DEM- and gauge-derived stream power. Elasticity indistinguishable from 0, or R^2 below 0.1 everywhere, kills it.

Novelty / leakage triage

Adjacent (closely related prior work exists)

Mill siting has been linked quantitatively to stream-power-adjacent geomorphology (the 2023 knickpoint study), Darby's tradition maps the ~5,600 Domesday mills descriptively, and Langdon's economics treats mills as rent assets for 1300-1540; the render-vs-DEM-stream-power elasticity regression on Domesday data was not located.

• 'Shaping landscapes and industry: linking historic watermill locations to bedrock river knickpoints' (2023) — Mill-siting vs geomorphic power proxies
• Langdon, 'Mills in the Medieval Economy: England 1300-1540' (OUP) — Mills as quantitative economic data, later period

Predictions

No prediction registered yet.