%0 Journal Article
%T Stable Boundary Layer Height Parameterization:  Learning from Artificial Neural Networks
%A Wei Li
%J Atmospheric and Climate Sciences
%P 523-531
%@ 2160-0422
%D 2013
%I Scientific Research Publishing
%R 10.4236/acs.2013.34055
%X <p class=\"MsoNormal\">
	<span lang=\"EN-US\" style=\"font-family:Verdana;\">Artificial neural networks (ANN)</span><span lang=\"EN-US\"> </span><span lang=\"EN-US\"><span style=\"font-family:Verdana;\">are employed using
different combinations among the surface friction velocity </span><i><span style=\"font-family:Verdana;\">u</span><sub><span style=\"font-family:Verdana;\">*</span></sub></i><span style=\"font-family:Verdana;\">, surface buoyancy flux </span><i><span style=\"font-family:Verdana;\">B</span><sub><span style=\"font-family:Verdana;\">s</span></sub></i><span style=\"font-family:Verdana;\">, free-flow stability </span><i><span style=\"font-family:Verdana;\">N</span></i><span style=\"font-family:Verdana;\">, Coriolis parameter </span><i><span style=\"font-family:Verdana;\">f</span></i><span style=\"font-family:Verdana;\">,
and surface roughness length </span><i><span style=\"font-family:Verdana;\">z</span></i><sub><span style=\"font-family:Verdana;\">0</span></sub></span><i><span lang=\"EN-US\"> </span></i><span lang=\"EN-US\"><span style=\"font-family:Verdana;\">from large-eddy simulation data as inputs
to investigate which variables are essential in determining the stable boundary
layer(SBL) height </span><i><span style=\"font-family:Verdana;\">h</span></i><span style=\"font-family:Verdana;\">. In addition, the
performances of several conventional linear SBL height parameterizations are
evaluated. ANN results indicate that the surface friction velocity </span><i><span style=\"font-family:Verdana;\">u</span><sub><span style=\"font-family:Verdana;\">*</span></sub></i><span style=\"font-family:Verdana;\"> is the most predominant
variable in the estimation of SBL height </span><i><span style=\"font-family:Verdana;\">h</span></i><span style=\"font-family:Verdana;\">.
When </span><i><span style=\"font-family:Verdana;\">u</span><sub><span style=\"font-family:Verdana;\">*</span></sub></i><span style=\"font-family:Verdana;\"> is absent, the
secondly important variable is the surface buoyancy flux </span><i><span style=\"font-family:Verdana;\">B</span><sub><span style=\"font-family:Verdana;\">s</span></sub></i><span style=\"font-family:Verdana;\">. The relevance of </span><i><span style=\"font-family:Verdana;\">N</span></i><span style=\"font-family:Verdana;\">, </span><i><span style=\"font-family:Verdana;\">f</span></i><span style=\"font-family:Verdana;\">, and </span><i><span style=\"font-family:Verdana;\">z</span></i><sub><span style=\"font-family:Verdana;\">0</span></sub><span style=\"font-family:Verdana;\"> to </span><i><span
%K Artificial Neural Network
%K Large-Eddy Simulation
%K Stable Boundary Layer Height
%U http://www.scirp.org/journal/PaperInformation.aspx?PaperID=37124