Premium
An observed relation between the macroscale local eddy flux of heat and the mean horizontal temperature gradient
Author(s) -
Tucker G. B.
Publication year - 1977
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1002/qj.49710343510
Subject(s) - temperature gradient , troposphere , stratosphere , atmospheric sciences , flux (metallurgy) , zonal and meridional , latitude , heat flux , climatology , potential temperature , physics , geology , meteorology , mechanics , heat transfer , geodesy , materials science , metallurgy
Abstract Frequent and detailed upper air observations at Laverton (38°S) for 30 days during September/October 1966 are analysed to determine the local horizontal eddy flux of heat and the mean horizontal potential temperature gradient. Components in the meridional and zonal directions are studied, but a more meaningful relation emerges when the vector eddy flux is considered in relation to the vector temperature gradient. the angle, δ, between these two vectors, and the down‐gradient and cross‐gradient exchange coefficients, K and R , are derived. A 3‐tiered structure between the surface and 30 km emerges, with down‐gradient flux ( K positive) in most of the troposphere and mid‐stratosphere, and counter‐gradient flux ( K negative) in the lower stratosphere. the cross‐gradient flux is in the direction of the thermal wind ( R positive) at nearly all heights except in the lowest layers and between 20 and 24 km. δ exhibits a fairly steady and systematic change throughout the height range analysed (2–28 km). Computations using standard upper air observations at selected Australian stations for 1970 show a mean annual pattern which is remarkably coherent with latitude (10–55°) and height (0–22km), except in the tropical troposphere. Typical derived mid‐latitude values for 1970 are:\documentclass{article}\pagestyle{empty}\begin{document}$$ \begin{array}{*{20}c} {} & {K(10^9 {\rm cm}^{\rm 2} {\rm s}^{{\rm ‐ 1}})} & {R(10^9 {\rm cm}^{\rm 2} {\rm s}^{{\rm ‐ 1}} } & {\delta ^0 } \\ {{\rm Lowerstratosphere}} & {{\rm ‐ 10}} & {{\rm 10}} & {{\rm 45}} \\ {{\rm Uppertroposphere}} & {{\rm 10}} & {{\rm 20}} & {{\rm 116}} \\ {{\rm Lower troposphere}} & {{\rm 20}} & {{\rm ‐ 10}} & {{\rm 206}} \\ \end{array} $$ \end{document}Only part of the vertical variation in the lowest layers can be accounted for by surface frictional influences.