Air temperature influences a variety of environmental processes, having a significant impact on the conditions of living of humans and other life forms. The Brazilian Northeast is a region that comprises a diversity of ecosystems, but it is most known as a semi-arid area characterized by severe environmental conditions. This work proposes to model air temperature in Brazilian Northeast to evaluate changing patterns between 2000 and 2017 using two interpolation techniques and comparing them. Monthly average temperature data from meteorological stations were gathered and used to compute the average annual temperature. Then, the timeframe was divided into 2 periods: 2000 to 2008 (1) and 2009 to 2017 (2), and the average temperature of each period was computed based on the annual average. Descriptive statistics analysis and exploratory spatial data analysis were performed, providing insights on the temperature patterns and distribution. In addition, interpolated surfaces were generated using the Inverse Distance Weighting and Ordinary Kriging methods for each period, and results were compared using error statistics derived with cross-validation. The results revealed that for both periods the highest temperatures are exhibited in the northern and central regions, whereas the lowest values occur in the south and east. In terms of change, an overall increase in the average temperature was noticed from period 1 to 2, although in some areas the increase was greater than in others. There was an increase of 0,37ºC in the mean, 1,37ºC in the maximum and 0,18ºC in the minimum temperature over the study region. Furthermore, Ordinary Kriging produced better results in terms of the bias of the predictions. The interpolated surfaces allow to visually notice the change in the average temperature between the periods. This study contributes to a better understanding of the temperature variability in the Brazilian Northeast in the 21st century.