Resumo

GOMEZ, MIRYAM et al. STATISTICAL MODELING OF ROAD CONTRIBUTION AS EMISSION SOURCES TO TOTAL SUSPENDED PARTICLES (TSP) UNDER MCF MODEL DOWNTOWN MEDELLIN-ANTIOQUIA - COLOMBIA , 2004. Dyna rev.fac.nac.minas [online]. 2007, vol.74, n.153, pp.35-40. ISSN 0012-7353.

Sand fields, constructions, carbon boilers, roads, and biologic sources are air-contaminant-constituent factors in Down Town Valle de Aburrá, among others. The distribution of road contribution data to total suspended particles according to the source receptor model MCF, source correlation modeling, is nearly a Gamma distribution. Chi-square goodness of fit is used to model statistically. This test for goodness of fit also allows to estimate the parameters of the distribution utilizing maximum likelihood method. As a convergency criteria, the Estimation Maximization Algorithm is used. The mean of road contribution data to total suspended particles according to the source receptor model MCF, is straightforward and validates the road contribution factor to the atmospheric pollution of the zone under study.

Palavras-chave : Maximum Likelihood Estimation; Expectation Maximization Algorithm; Pollutant Sources; Particulate Matter.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )