I evaluated whether package-theft incidents in Charlotte are random or clustered and at what scales. The dataset began with 773,278 CMPD incidents (2012–present) and was filtered to 47,252 records (2023+; near roads; within city jurisdiction; theft from a residence outdoors). All layers were projected to NAD 1983 StatePlane North Carolina FIPS 3200 (US feet). Methods included a 1-mile quadrat grid for dispersion, ANN for inter-event spacing, and Ripley’s K with Monte Carlo envelopes across 250–13,750 ft inside a residential-street mask.

All three tests agree: thefts are clustered. Quadrat counts show strong over-dispersion (VMR ≈ 32.7; χ² p < 0.0001). ANN on raw points indicates extreme clustering; after collapsing co-located events, clustering remains (observed ≈ 390 ft vs expected ≈ 442 ft; z ≈ −23.9; p < 0.001). Ripley’s K stays above confidence envelopes at every distance tested, meaning clustering persists from the shortest measured ranges through city-scale distances. The takeaway is operational: focus prevention in high-density pockets and adjacent blocks (e.g., targeted outreach, secure pickup options) and monitor whether clusters persist or shift over time.

This GIS project explores where a person could most safely survive in North Carolina during a zombie apocalypse. Starting with more than 37,000 parcels statewide, the analysis applied Boolean masking, vector scoring, and suitability modeling using data on wildfire hazard potential, flood zones, temperature severity, urban boundaries, hydrography, road proximity, and resupply points. The process identified clusters of parcels with the best combination of safety, accessibility, and resource availability, resulting in five primary safe zones: Western NC North, Western NC South, Central, Southeast, Inner Coastal Plain, and Albemarle Sound.

The findings are visualized through a set of thematic maps, individual zone profiles, and charts showing acreage and average suitability scores. The project is designed to be both methodologically rigorous and accessible to a broad audience, blending technical GIS methods with a creative survival theme to engage viewers while demonstrating spatial decision-making in action.

North Carolina’s unique alcohol laws—centralized pricing, no Sunday liquor sales, and exclusive distribution through state-run ABC stores—prompted this investigation into whether state control actually reduces public harm. As a newcomer to the state, the limited access raised questions about how alcohol flows differently here, and whether public safety benefits as a result. Using Charlotte as a study area, we mapped census tracts, ABC stores, and permitted locations, alongside crash records and population data.

Through spatial joins, attribute queries, and calculated ratios, we found that liquor access is overwhelmingly driven by permitted locations rather than ABC stores. Tracts with the highest number of permits per 1,000 people were linked to increased rates of intoxicated crashes, while ABC presence showed little correlation. Removing major commercial zones (airport, Uptown) improved the R² to 0.31, reinforcing a moderate link between permit density and crash frequency. While the state regulates liquor sales tightly, these findings suggest harm may shift—not vanish—under a controlled system.

In this final for the Raster Analysis class, I conducted a multicriteria suitability analysis to determine the most personally fitting places to live in North Carolina. The project integrated real-life needs and values—such as remote work, veteran healthcare access, commuting distance to a hair salon, and neighborhood diversity—into spatial layers. Using distance rasters, reclassification, and interpolation (IDW for broadband speeds), they created standardized suitability rasters across ten variables, including land cover, home prices, schools, fishing docks, and airports. Each layer was ranked and weighted to reflect lifestyle importance.

The user combined all layers using a weighted overlay approach in ArcGIS Pro, producing a final suitability map scored from 0 to 1000. The results were summarized using pixel counts above threshold values within municipal boundaries, helping to identify the top ten candidate towns. Two counties were excluded due to data gaps, and a Limitations section addressed methodological trade-offs and assumptions. The final product was shared as a StoryMap structured around purpose, methods, data, results, and personal context.

As Charlotte experienced record-breaking heat, this project explored how GIS can help residents find cooler, nearby outdoor spaces. I hand-selected and scored fifteen potential "chill zones" based on elevation, tree cover, water access, and shade potential. Using ArcGIS Pro, I conducted four types of network analysis: an origin-destination cost matrix to identify reachable cool spots from each city area, a route analysis to optimize visiting all five zones, a closest facility search to show which chill zones are nearest to each neighborhood, and a location-allocation to find three ideal sites that offer broad coverage. Each method built on the last, offering insights not only into distance and time, but also equitable access across Charlotte. I used public roads, neighborhood centroids from CAP Cardinal, and hand-curated points of interest to power the analysis. The final maps show different perspectives on accessibility, utility, and relief options during extreme heat. 

This project leveraged vector and raster analysis to identify prime stargazing locations across the Carolinas. Using clipping, projection, reclassification, buffering, and Euclidean distance, key datasets like light pollution, elevation, land cover, and road networks were processed and combined in a weighted overlay analysis to determine suitability. Additional tools like zonal statistics and raster calculator refined the results, showcasing GIS-based spatial modeling for outdoor recreation planning.