On the properties of stochastic intermittency in rainfall processes

In this work we propose a mixed approach to deal with the modelling of rainfall events, based on the analysis of geometrical and statistical properties of rain intermittency in time, combined with the predictability power derived from the analysis of no-rain periods distribution and from the binary...

Full description

Saved in:
Bibliographic Details
Published in:Water science and technology 2002, Vol.45 (2), p.35-40
Main Authors: MOLLNI, A, LA BARBERA, P, LANZA, L. G
Format: Article
Language:English
Subjects:
Aggregation
Analysis
Atmospheric precipitations
Distribution
Earth, ocean, space
Educational institutions
Environmental Monitoring
Exact sciences and technology
External geophysics
Extreme values
Filtration
Intermittency
Italy, Genova
Meteorology
Modelling
Models, Statistical
Probability distribution
Probability theory
Properties
Rain
Rainfall
Reproducibility of Results
Time Factors
Water in the atmosphere (humidity, clouds, evaporation, precipitation)
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work we propose a mixed approach to deal with the modelling of rainfall events, based on the analysis of geometrical and statistical properties of rain intermittency in time, combined with the predictability power derived from the analysis of no-rain periods distribution and from the binary decomposition of the rain signal. Some recent hypotheses on the nature of rain intermittency are reviewed too. In particular, the internal intermittent structure of a high resolution pluviometric time series covering one decade and recorded at the tipping bucket station of the University of Genova is analysed, by separating the internal intermittency of rainfall events from the inter-arrival process through a simple geometrical filtering procedure. In this way it is possible to associate no-rain intervals with a probability distribution both in virtue of their position within the event and their percentage. From this analysis, an invariant probability distribution for the no-rain periods within the events is obtained at different aggregation levels and its satisfactory agreement with a typical extreme value distribution is shown.
ISSN:0273-1223
1996-9732
DOI:10.2166/wst.2002.0024