The data preparation of primary shapefiles and feature classes is the first step. The OpenStreetMap goes at the bottom, followed by the California state shapefile and then the relevant files at the top in order of if and when I need to view them. The fire hazard point files of the Great Basin needs clipping with the state shapefile in order to retain only the required Californian data.

A very important step is the normalization of the fire occurences shapefile by the cities point file. This required the cities point file to undergo a Kernel density fix, which then was clipped by the state boundary shapefile yet again to retain only the required records. Once that is done the fire occurences shapefile can be normalized to create a fire hazard to people index map, which is more relevant to the findings of the project as it is an index of where the fire threatens the safety of the people rather than purely the fire originating locations.

The Kernel density map was rather poor in visual detail and representation however, and I wanted a better correlation between where the cities and fire stations are located. Specifically, is there roughly an equal spatial distribution of fire stations across the state? Is there a good spatial distribution of fire stations to serve areas of highest population, or the areas of greatest fire risk? Therefore, a Thiessen Polygon of both variables was performed. Two maps of this kind was procured, where one had the Cities point shapefiles overlaid on top of a Fire Stations Thiessen Polygon, and the other effectively vice-versa where Fire Stations point shapefiles overlaid on top of the Thiessen Polygon of Cities. The two maps are with inverse data input, yet is complementary to each other as it points out one’s relationship to another both ways.

The six variables within the CES3 shapefile is needed in conjunction with the Fire hazard to people index map to create another analysis through the Hotspot Getis-Ord Gi* tool. This is done with each and every variable chosen to see where areas of high intensity within the variables of the CES3 shapefiles correlate to areas of high fire hazard to people. The hotspot maps are the most visually representative as

There is also consideration given to finding the Moran’s I index and the Nearest Neighbor analysis of the variables. For the Moran’s I, the six variables will be performed, whereas for the nearest neighbor index, we will take the PM 2.5 values from the CES3 and perform the analysis with the cities point file, fire stations point file and fire hazard areas. The results will lead to the Moran’s I value and the z-scores of each variable, as well as the Nearest Neighbor ratio of the other 4 tested separately from the Moran.