A fellow urbanist named Kent Lundberg, toiling away in New Zealand, fighting the good fight, put together this image correlating street frontage and real estate value within the context of downtown. Red is the highest value area and as you can expect, the highest dollar real estate is the most central within the shown network.
We wanted to compare his real estate value map with spatial integration as produced by the depthmap software. Using a bit more context, because I wanted to show or at least examine the effect the highway ring has, on overall interconnectivity and therefore land value, I created the following two diagrams illustrating global connectivity and local connectivity.
One caveat, I don't have the highways in perfectly as they're immensely complicated in Auckland with various on-, off-ramps, as well as parallel highways, etc. I also haven't taken the time yet to go through and "unlink" certain segments where there is a 3-dimensional over- or under-pass. So the following maps read the links as still "connected."
The maps also don't reflect the connectivity added by the port (industry) and the marina (recreation) have on the overall system.
The first map is global connectivity:
Unsurprisingly, the downtown core lights up red (you may have to click to embiggen in order for the colors to read). Perhaps the most telling thing about Auckland is the striking difference when comparing global (above) and the local connectivity map below:
Theoretically, the local connectivity map will point out areas of local centers, places that act as neighborhood centers, the heart of generally walkable, livable areas. Downtown nearly disappears entirely from this map, which is not dissimilar to other places with mostly commercial/corporate office downtowns, the donut hole downtowns. Then with a few relatively high quality neighborhoods shortly outside the core.
The highways set up this situation where walkable areas can exist only in isolated bubbles and much of downtown has to function effectively as a buffer for the core from the highway loop. Much like Dallas really.
I'll have to spend some more time examining this when I have the time, as well as fine tuning the various linkages and de-linkages. Hopefully, Kent and I can share some of the data points as well and see what kind of regression analysis we can put together.