Using QGIS, Photoshop, InDesign and Rstudio, I analyzed, modeled and rendered several datasets to provide the strongest deliverable I could. 

This project had a very loose prompt, with the only requirements being spatial analysis and modeling on the Rouge River Watershed. I took advantage of this freedom to target what I find to be the most optimal venue for restoration in metropolitan areas: golf courses. 

Using spectral imagery, I isolated the exact color of golf course greens from satellite imagery uploaded to QGIS. I cross referenced these data with National Land Cover Database (NLCD) mapping, using Google Earth to fill the discrepancies between both data sets. I repeated the same process in order to find the cumulative acreage of watershed golf courses, making a feature class with existing NLCD from 2021 and shapefiles I made for missing courses. 

Once I had an accurate course inventory, I began spatial analysis within my three analytical frameworks: social, economic and environmental.

I used two to three ancillary spatial datasets, analyzing each individually, before applying them to a spatial modeling calculation from data I found most pertinent to the spatial relevance of golf courses and/or the need for a different use value. 

I found Stanford University's InVEST geospatial modeling plug in to be especially valuable in this step. InVEST takes site specific data, analyzes it and indexes it spatially.

After modeling these data, I reformatted them into more legible values and rendered this work as shown below. 

To create the spatial models below, I used a nearest neighbor functions to score which watershed golf courses were best candidates for use change based on their proximity, or lack thereof, to the three analytical framework calculations. 

I also analyzed the golf courses as an isolated variable, using nearest neighbor functions in combination to course type (public or private) and proximity to each other to score spatial redundancies. 

Using Google Earth, I made axonometric views from each golf course best suited for each respective spatial need. I then modeled its associated intervention concept on top of the real-time satellite imagery. 

This project refined my geospatial data science skills. It reinforced how much data can provide in planning and design. It also showed me that objective precedents like math and science can be used in creative ways to promote unique solutions. I hope to be able to use such creative applications of advanced analyses in my future endeavors.