It is known that the dense forests, such as an Amazon, have dynamics of complex vegetation and, at the same time, represent a key role in carbon capture and climate change, being important completely comprehension of these ecosystems. One of the ways to understand these dynamics are the vegetation indices, such as the EVI, which has been enhanced to maximizing the response of green vegetation. Some studies done in the Amazon has presented widely divergent results with regard the EVI MODIS sensor and his possible correlation with rainfall data in dry extremes occurred in 2005 and 2010. In view of this, this work aimed to relate to different data sources remote sensing, using the sensor data MODIS monthly EVI and TRMM monthly precipitation, assessing whether, for wet and dry period and for the Purus basin located west of the Amazon, there is correlation between them. After total extraction and preprocessing the boxplot statistics were generated for the two data sources through the mean of all basin pixels for the proposed period, as well as their spatial distribution. For the period analyzed, the results show no relation between EVI and TRMM data to the working range used, however presented a spatial correlation between the data. This indicates the need for further studies on the spatial dynamics and temporal forest response in relation to rain and its forms of representation, on the scale of the basin.