出版社：Grupo de Pesquisa Metodologias em Ensino e Aprendizagem em Ciências
摘要：This article aims to find the statistical distribution that best adesits to the set of atmospheric carbon monoxide (COatm) data recorded in micrograms per cubic meter (μg.m-3) by four automatic air quality monitoring stations (hourly measurements) installed in the city of Ipatinga, Vale do Aço, Minas Gerais (MG), Brazil. The historical data series comprises periods, from October-2013 to October-2017 (4 years), recorded in the neighborhoods Bom Retiro, Cariru and Cidade Nobre, and from January-2014 to March-2016 (2 years) measured in the Venice neighborhood. To reduce the huge discrepancies between the hourly data, the one was moved to the analysis of weekly amounts of COatm capture. With the application of variance analysis (ANOVA) after statistical data processing, it was found that the weekly data of COatm collected in the three stations together in the biennium 2015-16 were statistically equal, and that the statistical distribution that best adhered to the data set was beta: [135,000 + 170,000 * BETA(1.47 , 1.72)] μg.m-3. When the approximation was made by normal distribution (NORM), the following value was obtained: [NORM (213,000, 41,100)] μg.m−3, where the hypothesis that the data followed this distribution with the COatm accumulated in the three stations analyzed was not rejected. Historical data show that the concentration of carbon dioxide (CO2) has remained constant between 172 and 300 parts per million (ppm). But this amount has increased exponentially in recent years.
