Results/Conclusions: Geoengineering implementation is predicted to increase precipitation velocities substantially, with drying in biotically vulnerable areas over the Amazon, Siberia and southeast Africa. Under RCP4.5 (an emissions stabilization scenario, with total radiative forcing stabilized by 2100), temperature velocities are predicted to increase to a global median of 4.0 km yr-1. However, velocities could increase to 10.1 km yr-1 after sudden termination of geoengineering, with 29% of the land surface predicted to have more than double both temperature and precipitation velocities predicted for RCP4.5. In multiple regions including biodiversity hotspots, we project that sudden termination will cause temperature velocities to more than triple those predicted for RCP4.5 (i.e. ‘business as usual’ with moderate CO2 emission reduction). Finally, for many regions, termination would cause temperature and precipitation velocities to pull species and communities in divergent directions, with high potential for fragmentation of communities and ecosystems. Geoengineering is not an easy fix for greenhouse gas climate change, but represents a very serious threat to species and community persistence, including those in the most biodiverse regions on Earth.