Spatial Prediction based on the Third Law of Geography

press release: Spatial Prediction is one of the most important spatial analytical tasks for geographers and anyone who conducted analysis related to phenomena of spatial variation because it provides the needed information on spatial variation with a discrete set of field observations. However, existing theories (the first Law of Geography and statistical theories) for spatial prediction require the set of samples to be of certain size with special distribution as well as the relationships extracted from the samples to be spatially stable (stationary). These requirements render existing techniques unsuitable for spatial prediction over large and complex geographic areas at high spatial resolution which is a norm for geographic analysis in this digital era. This talk presents a new theory (the Third Law of Geography) which does not require samples and the relationship from these samples to meet the stated requirements. Case studies suggest that the new theory will transform spatial prediction to meet the need of this new digital era.

The Yi-Fu Lecture Series features a wide variety of U.S. and international guest lecturers from all geographic disciplines. Lecturers at these Friday seminars also often speak at brown-bag lunches, one-on-one student sessions, and breakfast meetings with student interest groups as part of their visit. Doctoral students are invited to present their final research. The lecture series was initiated by Dr. Tuan and receives enthusiastic support as a department and campus tradition.

All lectures are presented on Friday at 3:30pm in Science Hall - Rm 180 unless otherwise noted. Alumni, friends and the public are always invited to attend.