GIS Graduate Seminars: Jacob Eeling & Angela L. Bowman
Analyzing, Modeling and Visualizing Submersed Vegetation in Treatment Wetlands
Jacob Eeling, GIS Certificate candidate, Department of Ecology, Evolution and Organismal Biology
Adviser: William Crumpton
Submersed vegetation has a major impact on characteristics of wetland systems. Percent cover of submersed vegetation data was collected bi-weekly at three treatment wetlands in central Iowa. On average 150 data points were collected in each vegetation survey. ArcGIS was used to investigate normality and stationarity. There were significant issues with normality in the dataset, leading to selection of universal Kriging method as the method for spatial interpolation.
GPS point data from vegetation surveys was interpolated to submersed vegetation cover surface. Autocorrelation analysis (Nearest Neighbor Analysis) was used to determine if the sampling method achieved significant dispersal of samples over the sampling area. Local Moran’s I and Getis-Ord G was used to analyze the patterns of vegetation growth by analyzing clustering and hotspots with results from interpolated surface rasters. Kriging interpolation was done with R, which, compared to ArcGIS, allowed for greater transparency and flexibility in scripting and fitting of the model to the dataset. ArcMap 10.4 was implemented as a visualization tool. A time series animation of interpolated surfaces shows changes in wetland vegetation from May to November. ArcScene 10.4 was used to model and animate the vegetation in the wetlands in 3D.
Hydrologic Process Understanding through Satellite Observations
Angela L. Bowman, GIS Certificate candidate, Department of Geological & Atmospheric Sciences
Adviser: Chris Harding
Improving the spatial and temporal representation of the surface water balance in streamflow forecasting models is critical for increasing the utility of water supply, flood and drought predictions. Bowman investigated the application of satellite remote sensing data in operational streamflow prediction — specifically, the consequence of the role hydrologic model structure accuracy has on streamflow simulations through input of satellite-derived evaporation data.
In her GIS application, Bowman applied several techniques to process and analyze satellite-derived and model-simulated data. Examples include automation of geoprocesses to convert time series spatial model output to files that can then be converted to raster files; raster analysis for comparison of the model simulated data to satellite data used as a validation tool; and, upscaling or downscaling, as necessary, of the satellite data to match the scale of the model simulations. Evaluation statistics of the overall streamflow simulations show mixed results with the use of the satellite-derived data. Analysis of model states indicates the water balance components progress through the model differently while simulations of streamflow discharge do not reflect this.