Located in the tropics, the climate of the Amazon basin is marked by the strong seasonality of the tropical rain belts that migrate between the northern and southern hemispheres. During the summer rainy season, severe convective storms sweeping across large areas of the Amazon basin can induce strong winds that uproot trees, with important consequences for forest dynamics and carbon storage. In contrast, during spring before the onset of the rainy season, the Amazon basin experiences the warmest and driest season of the year, so the Amazon forest may be particularly vulnerable to variability and changes that occur during the spring season. Here, we explore future changes in severe convective storms during summer and the hot-dry conditions during spring under global warming to motivate research in understanding tropical forest response to climate change in the Amazon basin. Here we performed cloud-resolving simulations of the 2005 squall line that traversed almost the entire Amazonia from south to north between January 16^th and 18^th, causing about 0.5 million tree deaths over Manaus alone. To explore future changes in windthrows caused by convective storms, we simulated the same 2005 squall line but under the future large-scale environment to investigate how the storm may unfold with warming. Our simulations suggest a larger than 5-fold increase in the storm area and significant increases in the maximum wind speed that could amplify the impacts of the storm on tropical forest. We also analyzed the spring temperature and precipitation in a multi-model ensemble of global climate simulations for the present day and future. Driven by the delayed onset of rainfall, a robust signal across many tropical land regions under global warming, the Amazon will experience drier spring and as a result, much larger increase in mean and extreme temperatures in spring compared to the annual average. The warming and drying during spring, a consequence of the delay onset of the rainy season, may have compounded impacts on tropical forest.