The Swarm satellite is equipped with the Global Positioning System (GPS) receiver and accelerometer, which can be used to restore Earth's temporal gravity field model. Firstly, the terrestrial water storage (TWS) variability and 2015—2016 drought induced TWS deficits in the Amazon basin are estimated from the Swarm time-variable gravity field model between December 2013 and December 2016. Then, the estimated results from Swarm data are compared to the results derived from Gravity Recovery and Climate Experiment (GRACE), three hydrological models (Global Land Data Assimilation System (GLDAS), Climate Predication Center (CPC), National Centers for Environmental Prediction (NCEP))and 4 virtual stations (AMZ_076, AMZ_152, AMZ_ 228 and AMZ_215). Finally, the precipitation and temperature data are employed to study the origination of the drought event in the Amazon basin. The results indicate that the TWS derived from Swarm, GRACE, hydrological models, and virtual stations are in high consistency in the space/time distribution; Swarm provided a new perspective in detecting TWS, which has a potential to substitute GRACE satellite in studying the extremely drought and flood in the Amazon basin. The results in this paper may provide a reference in understanding Earth's mass transformation by Swarm.