A Study of the Dependence of Surface Wind Direction on the Gradient Wind

M. E. Johnson Division of Water and Land Resources, CSIRO, Canberra, Australia

Search for other papers by M. E. Johnson in
Current site
Google Scholar
PubMed
Close
,
J. D. Kalma Division of Water and Land Resources, CSIRO, Canberra, Australia

Search for other papers by J. D. Kalma in
Current site
Google Scholar
PubMed
Close
, and
D. Pederson Canberra College of Advanced Education, Canberra, Australia

Search for other papers by D. Pederson in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Measured surface wind patterns in the Canberra region of southeastern Australia have been used to derive a mean regional surface wind based on three-hourly means from 17 sites over an area of 600 km2. The mean direction was calculated using the definition of the mean of an angular variate given by Mardia, and the circular variance was calculated as a measure of dispersion. This technique treats each directional observation as a vector of unit length. Additionally, three-hourly observations of the wind at 850 mb for one location in the region have been taken to represent the gradient wind over the region. Data wore taken over a 12 month period.

With such input data, linear regression for an angular variate has been used to develop equations between the regional surface wind direction and variables defining diurnal and seasonal changes under varying synoptic conditions for each of the eight compass directions of gradient wind.

These equations show the generally close relationship between the gradient wind direction and the surface wind direction during daytime, but during nighttime the surface wind direction is often independent of the upper level wind. It is suggested that on such occasions the wind is a local drainage flow.

Abstract

Measured surface wind patterns in the Canberra region of southeastern Australia have been used to derive a mean regional surface wind based on three-hourly means from 17 sites over an area of 600 km2. The mean direction was calculated using the definition of the mean of an angular variate given by Mardia, and the circular variance was calculated as a measure of dispersion. This technique treats each directional observation as a vector of unit length. Additionally, three-hourly observations of the wind at 850 mb for one location in the region have been taken to represent the gradient wind over the region. Data wore taken over a 12 month period.

With such input data, linear regression for an angular variate has been used to develop equations between the regional surface wind direction and variables defining diurnal and seasonal changes under varying synoptic conditions for each of the eight compass directions of gradient wind.

These equations show the generally close relationship between the gradient wind direction and the surface wind direction during daytime, but during nighttime the surface wind direction is often independent of the upper level wind. It is suggested that on such occasions the wind is a local drainage flow.

Save