“…The wavelength of the ripples has been estimated to 4.5 mm in the present case but for the water-layer depth of 20 mm and 30 mm, the wavelength increases to 5.4 mm and 7.5 mm, respectively. However, from the classical theory of Kelvin-Helmholtz instability [6,7], in presence of density gradient ∇ρ = ρ water − ρ air and surface tension η, the wavenumber k of the first wave to go unstable is k = √ g∇ρ η and the air flow velocity necessary to drive waves at the air-water interface can be calculated by U = √ ρwater+ρair ρwaterρair (ηk + g k ∇ρ) If we put g = 9.8, ρ water = 1000, ρ air = 1.25 and η = 0.074 appropriate for air above water, we find the critical wavelength of the waves 2π/k = 1.7 cm travelling at velocity U = 6.6 m/s. Conversely, applying the same theory, for air blowing over water at 60 m/s, the wavelength of the disturbances should be of about 0.1 mm which is not in agreement with experimental observations.…”