The data from the Dartmouth Flood Observatory contains various geospatial representations of flood events. Some floods are depicted as points georeferenced to the nearest degree. Other floods are characterized by polygons.
To determine frequency of flood hazard, a 2.5 minute grid overlays the flood data and the number of flood events per grid cell is recorded. A flood event is defined as any portion of any type of flood representation occurring within a grid cell. Therefore, if a grid cell is crossed by a boundary of a flood polygon, then the frequency increases by one. If a point is within the grid cell boundaries, the frequency increases by one. Flooding of the same areas, but at different times, also increase the frequency.
The frequency range is classified into deciles, 10 classes of an approximately equal number of grid cells. Deciles are the chosen method of dissemination due to considerations of confidence with the data.
Areas of "No Data" are not necessarily without risk from hazards, but may rather be an artifact of a mask that excluded from analysis those areas with a population density less than 5 persons per square kilometer and without significant agriculture.