Geographic Information Systems
Hierarchical Models for Large Goestatistical Datasets with Applications to Forestry and Ecology
Measuring and Modeling Positional Uncertainty in Geographic Vector Data
Using Twitter as Volunteer Geographic Information to Predict Unit Specific Influenza Hospitalizations in New York City and London
Ligmann-Zielinska, A. (2013). “Spatially-Explicit Sensitivity Analysis of an Agent-Based Model of Land Use Change.” International Journal of Geographical Information Science online (DOI:10.1080/13658816.2013.782613).
G. O’Neil and A. M. Shortridge (2012) Quantifying local flow-direction error. International Journal of Geographical Information Science, doi:10.1080/13658816.2012.719627.
Shih, C. and S. Nicholls. 2011. “Modeling the Influence of Weather Variability on Leisure Traffic.” Tourism Analysis. 16 (3), 315-328.
Langley, S., and J. P. Messina. 2011. “Embracing the Open-Source Movement for Managing Spatial Data: A Case Study of African Trypanosomiasis.” Journal of Map and Geography Libraries. 7 (1) 87-113. Doi:10.1080/15420353.2011.534693.
Grady, S. C. 2011. “Housing Quality and Racial Disparities in Low Birth Weight: A GIS Assessment.” Chapter 15 in Geospatial Analysis of Environmental Health, J. Maantay and S. McLaffferty, eds. Springer Publishing.
Geographic Information Systems
The Department of Geography at Michigan State University has a long tradition of excellence in GIScience. GISci-related research and teaching at MSU focus on both theory and application of GIS technology to the spatial analysis of a variety of physical and human phenomena. Below are just a few examples of GISci activities we’re involved with.
Ancient Lake Algonquin. Ten thousand years ago the northern Michigan shoreline was much different than it is today. Digital analysis of terrain survey data employing global positioning systems (GPS) and baseline elevation data, we can reconstruct the boundaries of this ancient lake and better understand the region’s dynamic past.
Land Surface Change Monitoring. Dynamic terrain like dune fields can change rapidly. This map shows the results of a process quantifying the extent of this change between two high-resolution elevation models.
Multiresolution data models. Here a regularly gridded dataset is partitioned into regions of variable interpolation error based on a recursive compression algorithm. The partition can assist in identifying areas that are difficult to model, or inform appropriate remediation.
Error Propagation via Monte Carlo simulation. A process is detailed in this diagram to determine the effect of input spatial data error on a GIS operation. We have investigated a variety of different components for this process.
Modeling spatial distribution of grain prices in West Africa. Advanced statistical techniques can be used to identify spatial pattern. Such patterns may inform policy and management decisions for the region.
Compact Spatial Data Models. The DEM shown here consists of 4,096 different elevations. Employing a flexible discrete cosine transform allows us to characterize the surface with only a few coefficients. The red contours represent a surface with 64, 32, 16, 8, and 4 coefficients, respectively.
Geometric Probability. The use of spatial partitions in a data set implies that some features will intersect multiple tiles. Here the probability of such intersections is analyzed for equilateral triangular tiles.