Predicting the magnitude and locations of unusual climate activity continues to be a critical scientific endeavor. These hotspots may indicate increased risk of disease outbreak, extreme crop vulnerability, infrastructure breakdown, or other events of interest. The analysis of climate-related data sets requires special care, as they are typically quite large, and traditional methods for inference easily break down. We propose methodology for making simultaneous inference about the locations of climate hotspots. The methodology links a number of popular approaches for analyzing large geostatistical datasets, and results can often be obtained in a matter of minutes on a personal computer. We illustrate use of the methodology using state-of-the-art climate model data.