Outline service area drive times to “shelters” and provide several possible turn-by-turn evacuation routes for people within the Piney Point mine “impact zone”, to utilize in the event of a major collapse in the mine’s wastewater gypstacks

Additional sub-questions (used to solve the prime research question )
•What areas would serve best for evacuation gathering points “shelters”?
•What road segments may be at risk of tree fall from the event, and cause traffic obstructions? (For example: what areas adjacent to roadways may pose problems as choke points and may require detours including heavily forested areas, one lane roads, dirt or gravel roads, and areas with few road outlets and higher population density).
•What tracts of land will be affected by the flood?
•Where are the highest population densities within each of the three evacuation extents?

Data Processing Techniques
•Buffer (“impact extents” 0.5, 1, 2-mile intervals)
•Merge (both county road layers)
•Summary Stats (count populations in census blocks within the buffer
•Clip building layer to > zoning layer > census tracts to > buffer intervals
•Supervised Classification of Landsat land use raster. Find tree patches and
intersect with road layers to highlight downed tree blockages.
•Elevation tin with contours shows flood extent above or below 22 feet. The
official worst-case scenario.
•Network Analysis (find routes/find obstructions). The start node will be dense
population nodes (derived from the kernel density population analysis).

Network Analysis Service Areas & Routing for Evacuees
•I was not very surprised by the result at a
macro/generalized level. However, on a smaller scale, I was
surprised by some of the nuanced discoveries. I was a bit
surprised that some road segments nearest to the mine
leak location took the longest to evacuate. Also, some of
the tree canopy dispersion data were counterintuitive
there were higher densities in urban than in “natural” areas in
some spots. Also, the closer to the mine you were, the less
the tree fall hazard

•Overall, I think the methods I choose would fare well in a
basic analysis of this topic. With some dialed-in function
execution, and possibly some more defined parameters ( i.e.
what are the “real” treefall hazard extent radii), the study
outputs may become more valuable and useful.