ANTONELLO PASINI, VINICIO PELINO, SERGIO POTESTA'
A neural network model for visibility nowcasting from
surface observation: Results and sensitivity to physical input variables.
Journal of Geophysical Research, vol.106, n. D14, 2001.
A neural network model recently developed for fog nowcasting
from surface observations is summarized in its features, paying attention to its
particular learning structure (weighted least squares training), introduced
because of the nonconstant errors associated with the estimation of visibility
in Milan (Italy). The performance of this model is presented and shown to be
always better than persistence and climatology. Finally, we introduce a
bivariate analysis and a network pruning scheme and discuss the possibility of
identifying the more significant physical input variables for a correct
very short-range forecast of visibility.