Temperature is one of the most important variables used in climate studies. Changes in its time series can cause various social and political impacts, affecting diverse sectors of society, such as agriculture, for example. Southern Brazil is a region that relies heavily on agriculture, contributing to national production. Therefore, it is important to check the air temperature characteristic, i.e., its spatial pattern in this region. For this, daily data of minimum and maximum temperature of the air of the National Institute of Meteorology were used in the period from 1961 to 2011. The statistical technique of cluster analysis was applied to identify the homogeneous regions of temperature and the boxplot for the characterization of each an. In general, the results showed maximum values of temperatures in the west of Rio Grande do Sul. The regions of minimum temperatures presented a more homogenous pattern, with averages around 15 ºC. The regions of the maximum temperatures showed average values oscillating between 22 ºC and 28 ºC.
