Atmospheric Chemistry and Physics (May 2021)

Fifty-six years of surface solar radiation and sunshine duration over São Paulo, Brazil: 1961–2016

  • M. A. Yamasoe,
  • M. A. Yamasoe,
  • N. M. É. Rosário,
  • S. N. S. M. Almeida,
  • M. Wild

DOI
https://doi.org/10.5194/acp-21-6593-2021
Journal volume & issue
Vol. 21
pp. 6593 – 6603

Abstract

Read online

A total of 56 years (1961–2016) of daily surface downward solar irradiation, sunshine duration, diurnal temperature range and the fraction of the sky covered by clouds in the city of São Paulo, Brazil, were analysed. The main purpose was to contribute to the characterization and understanding of the dimming and brightening effects on solar global radiation in this part of South America. As observed in most of the previous studies worldwide, in this study, during the period between 1961 and the early 1980s, a negative trend in surface solar irradiation was detected in São Paulo, characterizing the occurrence of a dimming effect. Sunshine duration and the diurnal temperature range also presented negative trends, in opposition to the positive trend observed in the cloud cover fraction. However, a brightening effect, as observed in western industrialized countries in more recent years, was not observed. Instead, for surface downward irradiation, the negative trend persisted, with a trend of −0.13 MJ m−2 per decade, with a p value of 0.006, for the 56 years of data and in consonance with the cloud cover fraction increasing trend, but not statistically significant, of 0.3 % per decade (p value = 0.198). The trends for sunshine duration and the diurnal temperature range, by contrast, changed signal, as confirmed by a piecewise linear regression model. Some possible causes for the discrepancy are discussed, such as the frequency of fog occurrence, urban heat island effects, horizontal visibility (as a proxy for aerosol loading variability) and greenhouse gas concentration increase. Future studies on the aerosol effect are planned, particularly with higher temporal resolution, as well as modelling studies, to better analyse the contribution of each possible cause.