Attribute Scale on GIS Raster Data: What to Avoid at All Costs!
Antonio Medrano, Dept. of Geography & Center for Spatial Studies, University of California, Santa Barbara
12:00 p.m. Tuesday, January 26, 2015 | Phelps Hall 3512 (map)
Abstract. Raster cost-surface data is used extensively in GIS analyses. These surfaces are created in a GIS by assembling information from numerous other raster and vector layers to generate a surface that represents the cost penalty for placing a feature at any particular location. Analysis can then be done to optimize the location of these features such that they minimize their objective impact. While such analysis may appear to be non-subjective, the decisions made in assigning costs from the data layers and, in the connectivity of the network made from the cost raster will have major impacts on the analysis results. In this talk, Medrano presents a case study using bi-objective shortest path analysis, as would be used in applications of locating transmission lines, pipelines, and other linear features over terrain. Minor differences in the attribute scale used in assigning costs from nominal features can have a major impact on the analysis results, including the number of solutions, their spatial configuration, and their objective values. Such manipulations can be used to push the solutions toward a predetermined desired outcome, while still appearing to be completely objective to the casual observer. It is important to acknowledge these issues in order to improve upon the limitations imposed by existing software, and in order to perceive when such manipulations have been used to covertly push a design agenda that does not consider aspects not beneficial to certain parties.
Antonio Medrano completed his Ph.D. at the UCSB Geography Department in December 2014. At the Center for Spatial Studies, Medrano is a post-doctoral researcher who has been working with the UCSB Library to coordinate “discovery” of spatially enabled data within the library. Initial work has focused on using linked-data to connect research publications with their associated data sets, particularly focusing on self-deposit data through Esri’s OpenData initiative.
As a Ph.D. student, he partnered with Argonne National Laboratory to research spatial optimization methods for corridor location for new transmission lines. With the increasing development of new renewable energy sources (wind, solar, and geothermal), which tend to be sited in remote locations, Argonne saw a need to improve the geospatial analysis techniques for locating the transmission lines used to deliver the energy from these new sources to consumers. In particular, Medrano developed new algorithms for the difficult computation of sets of spatially diverse of shortest path alternatives and of multi-objective shortest paths, with an emphasis on parallel computing in order to handle problems involving big data. Prior to his doctoral work, Medrano earned an M.S. in Media Arts and Technology (UCSB), and a B.S. in Engineering (Harvey Mudd College).
The objectives of the ThinkSpatial brown-bag presentations are to exchange ideas about spatial perspectives in research and teaching, to broaden communication and cooperation across disciplines among faculty and graduate students, and to encourage the sharing of tools and concepts.
Please contact Antonio Medrano (email@example.com) to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial thinking.