Modeling the uncertainty of potential impacts on Robust Stormwater Management from neighborhood-scale impervious cover change: a case study of population-based scenarios in Pittsburgh, Pennsylvania

ABSTRACT

Cities increasingly face challenging regulatory requirements for their aging sewer systems. Population density and can contribute to changes in impervious cover that limit infiltration, expedite the rate of water drainage, and overwhelm the conveyance system. This work presents a method for applying high-spatial resolution population-based scenarios to pre-existing stormwater models. We explore the impact of two long-term population growth scenarios for the Pittsburgh region. We find that an extreme 79% growth in population results in 24% increase in impervious area and an 11% increase in total overflows. In contrast, a moderate 15% rise in population yields an 8% increase in impervious area and a corresponding 5% increase in overflows. As the relationship between population growth, impervious area, and total sewer overflow is non-linear, it is of critical importance to understand neighborhood-level changes. Our work fills a critical scalar gap for adding robustness analysis to these projections.

KEYWORDS:

• Land cover
• stormwater management
• uncertainty analysis

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

The supplemental data for this article can be accessed here.

Notes

1. Zonal statistics in ArcGIS 10 demonstrated that the value was closer to 8–10%. Research suggests that ten percent is the threshold of urbanization where aquatic systems begin to experience degradation (Booth and Jackson 1997), though there is an ongoing debate in the literature about this value (Shuster et al. 2005; Schueler, Fraley-McNeal, and Cappiella 2009). We feel confident in the last assumption, despite recent research that shows aquatic degradation at DCIA levels closer to 2–3% (Vietz et al. 2014) as surface water courses in this urbanized area are highly degraded and our focus is on combined sewer overflow rather than sedimentation and other morphological or ecological indicators.

2. Given the wide range of land cover in the study area, it was a reasonable assumption to choose a single model relationship that spanned all conditions than more finely tuned formulae that could introduce inaccuracies at the extremes (Stankowski 1972; Graham, Costello, and Mallon 1974).

3. In order to use the LID module for additional analysis, we updated the model engine to SWMM 5.1. Despite the model being otherwise the same to what was provided by others, while validating the model, a small number of outfalls subsequently had slightly higher overflow volumes, to which we applied a percentage bias correction factor. This resulted in no net difference with the ALCOSAN RBM overflow volume baseline.

4. We wish to note that this was not an artefact of Main Rivers having a different lumping structure than the other basins. With the region assessing land cover regulations and source control projects that could help limit the negative impacts of sprawling development, we explored an alternative ‘infill’ scenario where the population was restricted to growth only on existing impervious areas. The intensification of development from the SPC population projection, however, resulted in a relatively insignificant 19 fewer acres of additional impervious surface.

5. As an epilogue, we encourage readers to explore recent policy developments, such as Pittsburgh Water and Sewer Authority’s Draft City-Wide Green First Plan as well as ALCOSAN’s revised Clean Water Plan (PWSA 2016; ALCOSAN 2019).

Additional information

Funding

This work was supported by the John D. and Catherine T. MacArthur Foundation.