Changes in temperature and precipitation are already influencing US forests and that will continue in the future even as we mitigate climate change. Using spatiotemporally matched data for mean annual temperature (MAT) and mean annual precipitation (MAP), we used simulated annealing to estimate critical thresholds for changes in the growth and survival of roughly 150 tree species (153 spp. for growth, 159 spp. for survival) across the conterminous United States (CONUS). We found that growth of nearly one-third of tree species assessed (44 spp.) decreased with any increase in MAT (42–49 species), whereas fewer responded negatively to projected regional trends in MAP (< 20 species each in the east and west). Hypothetical increases in temperature (+1°C, +2°C) increased average annual growth in the Central East and Pacific Northwest and decreased growth over large areas of the Rockies and Southeast, while decadal survival generally decreased with temperature. Average annual growth and decadal survival had unfavorable associations with projected precipitation, generally decreasing with wetter conditions (+25%) in the east and decreasing with drier conditions (−25%) in the west. Beyond these averages, there were species that positively and negatively responded nearly everywhere across the CONUS, suggesting changes in forest composition are underway. We identified only eight species out of ~150 assessed that were tolerant to increases in temperature, and 24 species in the east and seven in the west were tolerant to regionally specific trends in precipitation (increases in the east and decreases in the west). We assessed confidence on a 5-point scale (1–5) for five aspects of uncertainty. Average confidence scores were generally high, though some species and metrics had low confidence scores especially for survival. These findings have significant implications for the future national forest carbon sink and for conservation efforts in the face of climate change.