WIT Press

Application Of Neural Networks To Model The Monin-Obukhov Length And The Mixed-layer Height From Ground-based Meteorological Data

Price

Free (open access)

Paper DOI

10.2495/AIR991021

Volume

37

Pages

10

Published

1999

Size

810 kb

Author(s)

A. Pelliccioni, U. Poll, P. Agnello & A. Coni

Abstract

The aim of this paper is to present the results of an application of a 3-layer Perceptron model with error back-propagation learning rule in order to reproduce the calculated time evolution of the Monin-Obukhov length, typically its inverse, and the Mixed-Layer height from ground based meteorological information. A summer data set of meteorological parameters have been measured in a meteo station using sonic anemometer, sensors of absolute and differential temperature and solar radiation. From these data the analytical values of Monin- Obukhov length and Mixed-Layer height have bee

Keywords