Statistics for spatial data has important application in many sciences including astronomy, biology, earth, environmental, atmospheric, oceanographic and elsewhere when spatial location is important. Image algebra was developed to provide a common mathematical environment for algorithm development for images. Its operands can be identified readily within a spatial statisti- cal context allowing the strengths of both technologies to be combined. In principle, spatial statistics can be used to analyze phenomena as patterns or objects; image algebraic operations permit implementation of such analyses. For recorded phenomena as complex as patterns or objects, analysis depends on defining useful principles, in this case, spatial statistical principles which treat the recorded obser- vation as random examples of the underlying phenomena. For example, in hazardous waste site characterization, efficient sampling and inference about the size, pattern and nature of the contamination can allow efficient planning for cleanup. The particular spatial statistical methods to be developed and implemented here are natural to use when the data are in the form of images.