The USC Spatial Sciences Institute (USC Spatial or SSI) has announced the 2019 prizes for the top three outstanding theses by graduating students in the M.S. in Geographic Information Science and Technology (GIST) program.
This year’s M. S. in GIST thesis prize winners are:
- Christopher Marder, first place, for his thesis “The Role of Precision in Spatial Narratives: Using a Modified Discourse Quality Index to Measure the Quality of Deliberative Spatial Data”;
- Cass Kalinski, second place, for his thesis“Building Better Species Distribution Models with Machine Learning: Assessing the Role of Covariate Scale and Tuning in Maxent Models”; and
- Brian Jeantete, third place, for his thesis“GeoBAT: Crowdsourcing Dynamic Perception of Safety Data Through the Integration of Mobile GIS and Ecological Momentary Assessments.”
In his thesis, Marder examined how digital forms of participatory mapping known as soft geographic information systems can be a valuable policy crafting tool. Soft geographic information systems encourage discussion as to the meaning of a particular location, as described in spatially precise terms. Yet despite soft geographic information systems’ potential, policy developers remain fixated on the assumption that affective responses to spatially precise locations undermine the quality of the quantitative justifications for crafting policy. To better understand how affective responses to spatially precise locations affect argument quality, a measure is needed to quantify the qualitative spatial data.
Marder amended a quality index from the field of discourse ethics to also quantify spatial narratives. Namely, he assessed the degree to which mentioning specific spatial locations changes the overall discourse quality. The modified discourse quality index was used to code public comments for a national forest plan revision at the Chugach National Forest in Alaska, U.S.A. The results of his analysis suggest that the precision of spatial narratives changes the discourse quality during deliberative activities. These results leave room to explore the empowerment of individuals or political groups when incorporating precise spatial thinking into their arguments for certain policies.
Dr. Jennifer Bernstein was Marder’s thesis advisor. Dr. Robert Vos and Dr. Elisabeth Sedano also served as his thesis committee members. For winning first prize, Marder receives a cash award of $1,500 from the Spatial Sciences Institute, a certificate, and his name inscribed on a plaque in the USC Spatial office honoring each year’s top thesis prize winner.
In his thesis, Kalinski investigated Maxent as a widely-used tool in the area of species distribution modeling (SDM). Maxent has shown good to superior performance compared to other SDM methods in studies using presence-only species data only when the tool is used properly. Often, however, due diligence with the selection of input data and model parameters is neglected, resulting in models of questionable quality.
Kalinski examined two factors of Maxent modeling, performance impact of covariate scaling and the results of model tuning on Maxent species distribution models, evaluating two questions related to these factors. Do higher resolution covariates yield a better performing Maxent model of potential habitat extent? Does a tuned Maxent model yield a better performing model of potential habitat than a model using the default Maxent settings? Two approaches to Maxent modeling, default parameters and tuned parameters, were used at two different covariate resolutions, yielding four evaluation models. Presence data for bristlecone pines (Pinus longaeva) provided the species example for the evaluation. Covariates were selected that are relevant to the species. These were scaled to match the two study resolutions. Model tuning was performed using the ENMeval R package. Quantitative and qualitative evaluations of the resulting models demonstrated improvements in the model performance in the tuned models. Results from the resolution aspects of the study were less conclusive. Issues with the quality of certain aspects of the climate and elevation data raises questions about the certainty of results at either resolution.
Dr. Karen Kemp served as Kalinski’s thesis advisor. Serving on his committee were Dr. Travis Longcore and Dr. Laura Loyola. In recognition of the second place prize, Kalinksi received a cash award of $1,000 and a certificate.
In his thesis, Jeantete explored how perception of the surrounding environment influences personal behaviors and the way humans interact with each other over time. The fear of crime and perceptions of safety are major contributors to these behaviors, and these perceptions influence the decisions made by law enforcement and city planners. Over time, a wide range of studies have been performed to understand the triggers that accentuate the fear of crime, and the possible solutions to alleviate these fears. Most of these studies have been static in nature and rarely included a dynamic geospatial component.
Jeantete detailed the process used to integrate a dynamic geospatial component by developing an ecological momentary assessment (EMA) mobile GIS application that: (1) pushes notifications to users at specific times; (2) enables users to collect spatiotemporal and perception data in real time; (3) distributes this information to a Realtime database; and (4) provides values for integration into multiple GIS platforms for subsequent GIS analysis. Once the mobile application was ready for release, testers were distributed throughout the city of Albuquerque, where they collected data and provided feedback on application functionality. At the conclusion of testing, all requirements to develop a functional EMA mobile application were achieved. Future work includes adding additional application features, external data sets for further analysis, and iPhone OS development for wider distribution.
Dr. Jennifer Bernstein chaired Jeantete’s thesis committee. Also serving on his thesis committee were Dr. Yao-Yi Chiang and Dr. Laura Loyola.
The theses which are nominated for consideration by the faculty are evaluated on the basis of depth of content, uniqueness of the research problem, quality of writing, overall thesis structure and relevance to solving real-world problems.
“On behalf of the Spatial Sciences Institute, I am delighted to offer congratulations to this year’s prize winners,” said Dr. Robert Vos, assistant professor (teaching) of Spatial Sciences and director of graduate studies. Vos added, “Every year, the faculty selection committee faces extremely tough decisions because of the high quality of the theses competing for the prize. The faculty selection committee let me know that the consideration this year was as tough as ever, which points to the ever-increasing academic quality and multidisciplinary strength of our students.”
To learn more about the value of a master’s thesis from the USC Spatial Sciences Institute GIST Graduate Programs, visit https://gis.usc.edu/curriculum/masters-thesis/. To see the roster of the M.S. GIST master’s theses which have been successfully defended, visit https://spatial.usc.edu/m-s-in-gist-theses/.