PS 38-103 - Modeling a hydrologically optimal green roof media mixture

Wednesday, August 9, 2017
Exhibit Hall, Oregon Convention Center
Michael A. Bollman1, Ryan Duchanois2, Montana Etten-Bohm2, Grace DeSantis2, David M. Olszyk1, John G. Lambrinos3 and Paul Mayer1, (1)Western Ecology Division, USEPA, National Health and Environmental Research Laboratory, Corvallis, OR, (2)US EPA, Greater Research Opportunities Undergraduate Fellowships Interns, Corvallis, OR, (3)Oregon State University, Corvallis, OR
Background/Question/Methods

A key environmental concern in managing urban ecosystems is controlling stormwater runoff to ameliorate pollution problems and sewage overflows. Vegetated green roofs have become an important green infrastructure tool to collect, store, and gradually release rainwater over time, with the added benefit of decreasing energy costs by acting as an insulator and increasing albedo. However, a major constraint to the survival of plants on green roofs is the lack of available water, particularly in the Pacific Northwest, where winters are cold and rainy and summers are warm and dry. The hydrologic attributes of the substrate used as growing medium strongly influence water retention, and, thus, plant survival. In this study we developed a simple spreadsheet model to optimize hydrologic performance of green roof media mixtures using data on the hydraulic conductivity (HC), wet weight (WW), and water held (WH) at saturation and after 14 days of drying for individual and mixtures of media constituents (peat moss, perlite, pumice, red cinder, sand, vermiculite) typically used in the Pacific Northwest. In addition, the proportion of processed constituents (perlite, vermiculite) was considered as a selection factor. The results of this study are intended to identify optimal green roof media mixtures for specific applications.

Results/Conclusions

We fixed the amount of organic matter (peat moss) at 20% by volume for media mixtures, so variation in hydrologic performance was driven by the composition of the inorganic fraction. Because perlite was light weight and had a high HC, yet held high amounts of water both when saturated and after 14 days, media mixtures dominated by perlite had the best hydrologic characteristics. Pumice also functioned relatively well, but was heavier. Although vermiculite performed very well in the first wetting, its water retention sharply decreased after undergoing a drying and re-wetting cycle, which resulted in an approximate 50% decrease in volume. Mixtures using perlite and/or pumice best addressed the performance criteria. This study demonstrates the potential to design green roofs with an appropriate media to enhance dry season water availability, while optimizing water release. With vegetation adapted to these media and the local environment, green roofs may be more effectively designed to not only mediate runoff, but also assist in cooling buildings and providing habitat.