In my last post I presented the ecocity vision, conceived by Richard Register, as one model for sharpening the urban-nature separation while embracing a degree of natural “infiltration” into the city that could avoid reinforcing the historically oppositional relationship between “nature” and “civilization.” The concept can be applied to population centers of any size, producing ecocities, ecotowns, and ecovillages. With larger population centers, the idea is that individual neighborhoods within a city could be densified to open up agricultural land and natural habitat in-between, resulting in a sort of archipelago of urban islands forming a larger metro area. Richard’s “Bigger Bay” scheme for the San Francisco Bay Area is an example.
In this post I’ll share some of my own initial thoughts on a possible second, complementary approach to densifying large metro areas while allowing that same infiltration of natural elements. I’ll call it the “wilds gradient” model, illustrated generically in the diagram below. It could incorporate most or all ecocity principles, though it’s more hierarchical than the archipelago approach in that it’s structured by a series of concentric zones ranging from the highest percentage of urbanized space in the center to the lowest at the edges. In that sense it’s superficially similar to the geography of today’s (mostly dysfunctional) cities that typically grade from more dense at the center to less dense at the outskirts. The crucial difference is that the density transition within the wilds gradient is much less granular: it only shows up looking at the city as a whole. All pieces of urban fabric (the white areas in the diagram) in-between the bits of green, if you zoomed in on them, would appear to have nearly the same, very high density, resulting in a much smaller total urban area than today’s typical metropolis (whether or not you include the green parts).
The wilds gradient model shown diagrammatically, with the three concentric zones described below. White areas represent urbanization.
The wilds gradient model divides a metro area into three concentric zones as shown in the diagram above, each named after the shape or character of the “nature” preserved or created within:
Fingers. The natural environment surrounding the city would extend as fingers into the outermost zone of the city, each the size of many city blocks up to that of an entire neighborhood. Their shape and location would be determined by topography, hydrology, or other environmental factors, functional and/or aesthetic. The urban zones in-between (again, just or nearly as dense as the center city) might be considered “ecovillages” similar to the neighborhood islands in the archipelago approach, but they’d be more like urban peninsulas than urban islands. In this zone, the area of undeveloped land would be roughly equal to, or a bit less than, the area of developed land.
Islands. Natural spaces in this zone would still be relatively large (say, several city blocks) and would have a character as natural as the “fingers” in Zone 1, but they would be generally isolated within the urban fabric. These natural islands best fit what I’ve been referring to as urban wilds.
Elements. These natural elements would represent forms rather than spaces—they could be structures like vegetated walls and roofs, or individual plants, in any configuration but importantly part of the urban fabric (or forming a layer overlaid upon it) rather than outside of it. These elements could be situated within “green spaces” like tree planting strips or parks. But these spaces themselves wouldn’t be “natural” in the sense that the configurations of natural elements within wouldn’t re-create a natural environment either aesthetically or functionally. In fact, those configurations could tend toward geometric— clearly of the city—rather than naturalistic.
The configurations of all of these components could and would be much more varied and irregular than shown in the diagrams above. For example, depending on what existing natural features are being preserved (especially waterways), the shapes might tend more toward linear, continuous corridors, varying in width. The second set of diagrams, below, shows how the same zones might look overlaid on a real city, one with a configuration shaped by a real environmental context (in this case a linear system of ridges and valleys).
The wilds gradient model as applied to a hypothetical “real” city, overlaid with the same zones.
Note that I’m oversimplifying in referring to the entire non-urbanized area (the green area in all the diagrams) as “natural” in the sense of being unmanaged natural ecosystems. The majority of it would be, particularly the green area in-between cities outside the three zones. But the green area would also include some percentage of agricultural land and active recreational space (as in more traditional urban parks).
Fitting the wilds gradient into the “island civilization” approach to planning on a more regional scale.
Zoomed out to the regional scale, cities structured on this transition from fully “wild nature” beyond the city to fully “urbanized nature” in the city center would still appear as compact, isolated urban islands floating in a sea of preserved and restored natural ecosystems, not too different from a pure application of the “island civilization” concept. But zooming in to the city itself, from outer edge to center, that oppositional relationship between nature and city—where nature is seen as something alternately noble and threatening—is softened and inverted. Moving inward from Zones 1 to 3, nature is first intermingled with city, then surrounded by it, then fully “diluted” by it—a reflection and reminder of what our modern relationship with the natural environment has generally looked like. On the level of the entire city this could be considered a “mixing” of urban and natural along the lines of what the Landscape Urbanists propose. But as I said above, this is a matter of scale and perspective: zooming in, the contrast between urban and natural spaces is still enforced. My argument has been that keeping the two spatially separate is crucial from a conservation perspective and perfectly justifiable from a cultural one, but that in order to 1) maintain a sense of ecological identity and 2) to prevent this separation from devolving into the demonization or glorification of one side, we need to think about that condition of separation with more nuance. I think this wilds gradient model could be a way to achieve an environmental version of “separate but equal.”
Like the archipelago approach of fitting large metro areas to ecocity principles of density, compactness and three-dimensionality, the wilds gradient model would take on different forms depending on the particular urban and natural geography in question. Applied to the real world the two models would probably grade into each other, and they could probably even both apply to the same place at the same time at different scales. Regardless of the formal details, though, it’s important to recognize that the overall idea of compressing and densifying cities, while idealistic, can’t just be aspirational. I’ve been focusing on the less tangible benefits of doing so—namely reinforcing the separate yet linked identities of both urban and natural environments—but of course there are much more practical and consequential reasons that are becoming more apparent every day. There are certain to be heated arguments over the degree to which existing population centers can be reconfigured. But there’s no question that the future is going to bring a lot of building and re-building, and we can’t afford to do so the same way that we have been—with low-density sprawl.
Darren