“Ontological considerations in GIScience” by Agarwal states the issues of ontologies and the communication of the data contained within the various ontology types is problematic to understanding. One way I have worked with in the past to resolve the issue of relating different ontologies is setting a accessible reference grid over the different ontology of the same area and creating a cross-referencing database that links both the human and scientific data. For example, if I selected a grid a side bar would open that would display human based data (impressions, oral histories, etc.) and any other data for that same location. In addition, neighboring grid cells can be linked to see if they represent similar data, and a referencing system based on type of ontology characteristic created. This last part is similar to what I believe Agarwal maybe alluding to in the last part of his article.
Although grouping of data may be useful, the variety of ontologies and the evolution of how humans see the world and use of GIS makes this difficult and a problem for GIScience to resolve. I believe the problem is not how we group data but how databases are mostly static in form and unable to expand to new information. That is one reason why creating grids for an area to set a standard may be the best solution, where they are related to dynamic ontology data sets. To simplify, ontologies are dynamic with many ontology layer types, set to a geo-referenced grid.
C_N_Cycles