Open AccessArtificial neural network estimation of rainfall intensity from radar observations
Author(s) -
Orlandini Stefano,
Morlini Isabella
Publication year - 2000
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2000jd900408
Subject(s) - artificial neural network , principal component analysis , radar , computer science , position (finance) , variance (accounting) , perceptron , function (biology) , multilayer perceptron , algorithm , pattern recognition (psychology) , remote sensing , artificial intelligence , meteorology , environmental science , statistics , mathematics , geology , physics , telecommunications , finance , economics , business , accounting , evolutionary biology , biology
Volumetric scans of radar reflectivity Z and gage measurements of rainfall intensity R are used to explore the capabilities of three artificial neural networks to identify and reproduce the functional relationship between Z and R. The three networks are a multilayer perceptron, a Bayesian network, and a radial basis function network. For each of them, numerical experiments are conducted incorporating in the network inputs different descriptions of the space‐time variability of Z. Space variability refers to the observations of Z along the vertical atmospheric profile, at 11 constant altitude plan position indicator levels, namely Z T = ( Z 1 ,…, Z 11 ). Time variability refers to the observations of Z at the time intervals prior to that for which the estimate of R is provided. Space variability is evaluated by performing a principal component analysis over standardized values of Z , namely Z ˜ , and the first two principal components of Z ˜ (which describe 91% of the original variance) are used to synthesize the elements of Z into fewer orthogonal inputs for the networks. Network predictions significantly improve when the models are trained with the two principal components of Z ˜ with respect to the case in which only Z 1 is used. Increasing the time horizon further improves the performances of the Bayesian network but is found to worsen the performances of the other two networks.