The Effectiveness of a Constructed Vegetated Roof for Pollutant Removal and Stormwater Source Control

Date of Award

December 2017

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering


Charles T. Driscoll


Green infrastructure, Green roof, Hydrologic response, Nutrients, Stormwater management, Water quality

Subject Categories



Sustainable strategies such as vegetated roofs have been implemented as stormwater management tools to mitigate disturbance of the hydrologic cycle resulting from the urbanization. Vegetated roofs, also referred to as green roofs, can improve the urban landscape, by reducing heat island effects, providing ecosystem services, and facilitating the retention and treatment of stormwater. Green roofs receive particular attention because they do not require acquisition and development of land and represent an application of biomimicry in design and construction.

The results of a four-year study of water quantity and quality in runoff from an extensive, sedum covered, green roof on an urban commercial building in Syracuse NY, USA are summarized. Monitoring commenced seven months after the roof was constructed. The green roof was compared to a conventional (impermeable, high-albedo) membrane roof in addition to paired measurements of wet and bulk deposition at the study site. The study evaluated the effects of precipitation, evapotranspiration (ET), antecedent dry period (ADP) and seasonal variation on the runoff quantity and distribution of the green roof. Fluxes of nutrients and major solutes were estimated.

The green roof greatly facilitated retention of precipitation events without significant change over the four-year study. The green roof retained on average 95.9±3.6% of precipitation quantity on an event basis, with a range from 75% to 99.6%. However, as precipitation quantity increased the water retention decreased. This high water retention capacity was the result of the combined effects of ET, stormwater storage (plants, growth media, and roof drainage composite), as well as limited surface runoff from the roof deck due to variation in sloping of the green roof and the tapered insulation to the deck drains. The water retention capacity of the green roof did not change significantly between growing and non-growing seasons. Slightly greater precipitation during the growing season coincided with increased ET. Average reference ET during the growing season was approximately three times greater than during the non-growing season. The hydrologic performance of the green roof was not significantly impacted by an ADP greater than two days. The potential stormwater retention of the green roof, expressed as a function of a dimensionless coefficient called a curve number (CN), was estimated to be 82.

The stormwater drainage quality function of the green roof was evaluated using concentrations and fluxes of nutrients and major solutes. The seasonal and year-to-year variation in water quality of drainage from the green roof was compared with atmospheric deposition and drainage from the impermeable roof. Drainage waters from the green roof exhibited a high concentration of nutrients compared to atmospheric deposition, particularly during the growing season. However, nutrient losses were generally low because of the strong retention of water. There was marked variation in the retention of nutrients by season due to variations in concentrations in drainage from the green roof. The green roof was a net sink of total nitrogen, total phosphorus and chloride, and a source of phosphate, dissolved inorganic carbon, and dissolved organic carbon. Chloride exhibited elevated inputs and leaching during the winter due to atmospheric inputs from road salting. The drainage from the green and impermeable roofs met the United States Environmental Protection Agency (USEPA) proposed standards for freshwater for all parameters, except for total phosphorus.

This study indicated that the drainage from the green roof and the runoff from the high-albedo roof could be an adequate substitute or supplement to the potable water used in the make-up to cooling towers or stored onsite for irrigation, including reusing the stored water to irrigate the green roof.

A hydrological manipulation experiment involving an artificial irrigation of the green roof was carried out. The green roof was effective in retaining precipitation quantity from the experiment (96.9%) and the water retention was similar to the long-term monitoring results (95.9%). Discharge hydrograph for the hydrological manipulation experiment demonstrated that runoff from the green roof did not occur until the soil was saturated. Once the roof was saturated, almost instantaneous discharge was observed. Loadings of nutrients were calculated for experimental and actual storms using the concentration of nutrients and the flow data of water discharging from the green roof. The experiment supported the results from the long-term monitoring and confirmed that the nutrient retention is dependent on the water retention. The retention of total nitrogen, total phosphorus, and dissolved organic carbon during the experiment confirmed that the elevated concentrations in green roof drainage are not an issue of concern as the strong retention of water results in limited mass loss of nutrients from the green roof to the sewer system.


SURFACE provides description only. Full text may be available to ProQuest subscribers. Please ask your Librarian for assistance.

This document is currently not available here.