Runoff is (mostly) a transportation problem

Over the last decade, some much-needed attention has been directed at the problem of stormwater runoff in metro areas. Heavy rains rush down hard surfaces eroding stream banks, intensifying flooding, burying aquatic biodiversity, and carrying every variety of pollutant and unwanted nutrient into our waterways. While the nature of the problem has been studied very thoroughly, we're still in the process of figuring out how to manage it well, both at the level of regulation and individual practice. I think we'd do well to remember Bruce Ferguson's words in his classic Introduction to Stormwater from 1998:

"The runoff and pollution from a contemporary city result not so much from the number of human beings, as from the lavish support given to their automobiles in land use and land development."

Therefor,

"A given population can reduce its needs for pavement by reducing its dependence on the automobile. Development that provides for non-automotive transportation has compact mixed of land use, where many of people's everyday needs can be met within small distances, and safe, convenient paths for biking among the various neighborhoods. When people are not using cars, there are no emissions and there is no demand for parking pavement when they arrive at their destination."

There a couple of lessons to draw from this. First of all, rooftops are less important than pavement, particularly in urban areas where multiple floors of activity can occur under one roof. Green roofs and rain barrels get lots of press for a good reason. They can promote food production, help conserve water for human use, and are just cool. But as any glance at Google Earth will confirm, parking lots and roadways create far more impervious surfaces (up to a quarter of all land used in metro areas), and importantly more interconnected impervious surfaces, than rooftops. Rooftops also do not generate or convey pollution like ground-level impervious surface does.

The second observation to make is that stormwater management is as much a part of an interconnected system as transportation is. Evaluating nutrient load on a site-by-site basis makes as much sense as considering only the transportation needs internal to a proposed development when assessing its impact. Often a residential subdivision can get a gold star for using low-impact development techniques to virtually eliminate on-site runoff, but if it's still auto-dependent and generating a significant number of vehicle trips the problem is only pushed somewhere else. What roadways are these motor vehicles using? Where are the three parking spots for each vehicle going to be located?

My intent is not to downplay all of good techniques being developed to mitigate the impacts of our transportation system, but let's not lose sight of simplicity when thinking about regulations. The best solution is to create and promote the kinds of places where driving is less necessary in the first place.