All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 170 40 5
PDF Downloads 80 41 4

Optimal Vector Interpolation of Wind Fields

View More View Less
  • 1 Department of Mathematics, Israel Institute for Biological Research, Ness-Ziona, Israel
Restricted access

Abstract

In this paper a method of optimal interpolation of the vector field is suggested. It is shown that the vector estimator and its associated error are independent of the choice of the coordinate system in which to represent the wind. The use of the method as a tool for determining the optimal number and sites of observation stations to represent the wind in a given area is demonstrated. A vector correlation coefficient is defined from the regression matrix in a manner analogous to the scalar case. It is independent of the coordinate system and is symmetric in its arguments. The method was applied to a dataset of surface wind measured in a complex terrain in Israel during the summer 1993. Comparison of the wind field interpolated by this method to that interpolated by the commonly used 1/r2 method shows the suggested method is a significant improvement both in magnitude and direction.

Abstract

In this paper a method of optimal interpolation of the vector field is suggested. It is shown that the vector estimator and its associated error are independent of the choice of the coordinate system in which to represent the wind. The use of the method as a tool for determining the optimal number and sites of observation stations to represent the wind in a given area is demonstrated. A vector correlation coefficient is defined from the regression matrix in a manner analogous to the scalar case. It is independent of the coordinate system and is symmetric in its arguments. The method was applied to a dataset of surface wind measured in a complex terrain in Israel during the summer 1993. Comparison of the wind field interpolated by this method to that interpolated by the commonly used 1/r2 method shows the suggested method is a significant improvement both in magnitude and direction.

Save