Environmentally Friendly Transportation to Work
Kernel Density Maps
Figure 1: Non-Private Vehicle Kernel Density Map (1,500 Square Miles)
This map was created using kernel density analysis with a 1,500 square mile search radius. It shows very localized peaks in non-private vehicle commuting, especially in Oakland, Shadyside, and nearby neighborhoods. The smaller search radius highlights high-concentration areas in detail but can create a cluttered visual impression due to sharp gradients and noise from localized variation.
Figure 2: Non-Private Vehicle Kernel Density Map (3,000 Square Miles)
Using a 3,000 square mile search radius, this map represents a mid-range smoothing level for kernel density. It reduces some of the spatial noise seen in Figure 1, while preserving key clusters. The analysis helps reveal underlying regional patterns of non-private vehicle use while still reflecting moderately detailed variation across the city.
Figure 3: Non-Private Vehicle Kernel Density Map (6,000 Square Miles)
This map employs a 6,000 square mile search radius for kernel density. It offers the most generalized surface of the three, revealing broader city-wide trends rather than sharp local peaks. This wider bandwidth is helpful in showing large-scale spatial patterns, such as the dense core in Oakland to Shadyside area, and adjacent neighborhoods with consistently high non-private vehicle use.
Figure 4: Best Non-Private Vehicle Kernel Density Map with Transportation Infrastructure (6,000 Square Miles)
Figure 4 overlays the 6,000 square mile kernel density surface with transportation infrastructure layers including sidewalks, bike lanes, and Port Authority stops. Created using multiple datasets, this layout reveals a strong spatial correlation between high non-private vehicle commuting rates and available pedestrian and transit infrastructure. The integration of infrastructure highlights the urban areas most supported for alternative commuting options.
Report
Between the point symbolized map and the kernel density maps, the kernel approach provides a more informative spatial representation. While point maps offer raw values, kernel density (especially at 6,000 square miles) better illustrate trends of non-private vehicle use. The density surface simplifies interpretation, which makes it more efficient to visualize the data for analyzing urban commuting behavior.
Figure 4 is the best kernel density map representation of the NonPrivateVehicle attribute because this version of the map provides a smoother, more interpretable density surface and avoids the over-clustering seen in narrower bandwidths. Moreover, it highlights the overlaps of the high kernel density areas with local transportation infrastructure networks, effectively showcasing that the presence of walking and transit infrastructure supports increased use of non-private transportation.
In addition to the high values of the NonPrivateVehicle attribute, several other spatial factors contribute to the highest peaks observed in the kernel density maps. Areas with the greatest density of NonPrivateVehicle attribute tend to have strong public transportation infrastructure, including bike lanes, sidewalks and a high concentration of Port Authority stops. Altogether, these features support multimodal commuting options such as riding bus, walking and cycling, which contributes to the highest peaks in the kernel density.