The coastline is a long, narrow hazard zone vulnerable to storm surge or tsunami as well as to the much slower rise in relative sea level associated with climate change and crustal subsidence. Effective planning for coastal hazards depends upon reliable climatological and tidal data, monitoring networks able to detect large rotating storms or distant undersea earthquakes, and refinement of computer simulation models like SLOSH (Sea, Lake, and Overland Surges from Hurricanes), developed in the 1970s and used for three decades by the National Weather Service to identify areas requiring evacuation. SLOSH, which accommodated 1970s computers with a coarse systematic grid unable to capture the effects of irregular shorelines and small features that can accelerate or attenuate sudden inundation by wind-driven seas, epitomizes the need for better models and more refined elevation data. Reliable modeling is important because the time-consuming evacuation of large coastal populations is thwarted by the unnecessary clearing of areas not at risk. Improved planning for severe coastal storms is possible with more computationally efficient models based on high-resolution, unstructured grids and able to account for wave action as well as topographic irregularities. Although these advanced models require refined elevation data—as do credible projections of sea level rise—for many parts of the United States coast the best current elevation data are derived from obsolete topographic maps with a five- or ten-foot contour interval. Topographic lidar, which can provide a half-foot vertical resolution, is a promising solution to the pressing need for better coastal elevation data. Integration with bathymetric lidar, which can image the sea floor in water as deep as 20 to 30 meters, yields a seamless topographic/bathymetric dataset useful for the high-resolution modeling of storm surge.