Spatial-temporal patterns of hydrological droughts in the Amazon basin are derived from drought indices computed from existing streamflow data. Principal component analysis and Monte Carlo simulations are employed to account for the uncertainty and overcome the limitations of missing data in streamflow records. Results show that northern and southern subbasins differ in drought trends and in patterns of correlation between drought indices and climate anomalies originating from the Pacific (El Nino-Southern Oscillation) and Atlantic (differences in sea surface temperature across the equator) Oceans. A significant trend toward more intense droughts is found in the southern subbasins, which is highly correlated to tropical Atlantic Ocean sea surface temperature anomalies. That drying trend might have distinct causes in each subbasin and can lead to potential intensification of regional impacts.