Studies of turbulence in the surface layer over water (Lough Neagh). Part I. Instrumentation, programme, profiles

Abstract

Abstract Instrumentation is described which has been used over a body of deep water, with fetches of 8 km to 22 km, to provide (i) vertical profiles of mean wind u , potential temperature ø and specific humidity q from the water surface to 16 m height and (ii) turbulent fluctuations of temperature and of the longitudinal and vertical components of wind and hence the vertical fluxes of momentum and sensible heat. Logarithmic wind profiles in near‐neutral conditions lead to a drag‐coefficient C D of the water surface as a function of the wind speed u 10 at 10 m height given by 10 3 C D = 0.36 + 0.10 u 10 (3 m s −1 < u 10 < 16 m s −1 ). Corresponding bulk transfer coefficients for heat and water vapour are not distinguishably different from C D in the observed range of u*z 0 /ν < 10 2 (conventional nomenclature). An analysis, following Webb (1970), of profiles in stable conditions (0 < Ri 4 ≤ 0.16) lends support to the utility of the log‐linear form. Using the Monin‐Oboukhov relation ∂ X /∂ z = X *ø ( z / L )/( kz ) where × is either u, ø , or q and taking ø( z / L ) = 1 + α z / L , so that z/L = Ri/(1− α Ri), we find α ≃ 6. In unstable conditions (0 > Ri 4 ≥ −0.6) the profile shape factor S x = ( X 4 − X 1 )/( X 16 − X 1 ) is greater for ø than for u , implying ø H < ø M (K H > K M ) but S q has less well defined behaviour. The data on S u ,ø are not inconsistent with ø M = (1 − 16 z/L) −1/4 , ø H = ø , as proposed by Dyer and Hicks (1970) and others but the shape factor is very insensitive to the numerical factor multiplying z/L in these relations. Analysis of the measurements of turbulent fluctuations and derived fluxes will be presented in later papers.