Via Switchboard of the Natural Resources Defense Council
Does suburban sprawl – spread-out, automobile-dependent strip malls, big-box stores, wide arterial roadways, and unending large-lot housing – cause flooding? Absolutely not. (Sprawl doesn’t make it rain, although I can put together a very plausible theory about increased driving, tailpipe emissions, global warming, and severe weather events.) But does sprawl aggravate flooding? Oh, yeah. Here’s how:
When it rains, the water needs somewhere to go. Ideally, that someplace is a forest or meadow, which filters and absorbs the water into the ground. But when, instead of natural vegetation, we have rooftops and pavement, the natural process is broken and the water runs off, gaining volume and velocity. If the rainfall is hard enough and/or steady enough, flooding occurs; and floodwaters increase as runoff increases. Nature, already overburdened by severe precipitation, is prevented entirely from doing its job at limiting the accumulation of flood waters when impervious surface is in the way.
What does this have to do with suburban sprawl? Spread-out, low-rise development contributes more rooftops and pavement per unit of development to the watershed than do walkable neighborhoods. Imagine a 200,000-square-foot, one-story Walmart Supercenter surrounded by 15-20 acres of surface parking. When it rains on Walmart's property, there’s no way the water can get into the ground through naturl filtration processes. Now multiply that by all the other parking lots required for strip malls and office parks, and all the widened and extended road surfaces needed to accommodate traffic heading to the retail and spread-out housing.
Now imagine a different scenario: The same amount of floor space is accommodated by a combination of even two- to four-story buildings, and housing built more compactly to a walkable scale. Imagine that the pattern reaches sufficient critical mass to support decent transit service and the substitution of walking, bicycling, and transit use for some of those car trips, thus reducing the amount of road surface needed. Where there is parking, imagine that some of it, rather than spread out on surface lots, is placed in multi-story, above- or below-ground garages such as those found in urban areas. With rainwater hitting a smaller footprint of pavement and other hard surfaces, there is less runoff.
Would the difference be great enough to prevent flooding altogether during the most severe weather events? Probably not. But it could make a difference in the volume of water running off into the flood.
EPA has done some calculations on the residential part of the issue. Suppose your metropolitan area is going to grow by 10,000 homes over the next several years. If those homes are built one to an acre, a hypothetical storm might produce 187 cubic meters of runoff; but reducing the watershed coverage to an average of four homes per acre, the runoff from those same new homes would be reduced to 62 cubic meters. Build the homes at eight to an acre, and the runoff would reduce further, to 49.5 cubic meters. The main reason for the difference is the amount of roadway required to service the homes is much greater at low densities than at moderate densities.