On this page:
- Research Objective 1: Biological Health of Green Roof Plants in a High Elevation, Semi-Arid Climate
- Research Objective 2: Stormwater Management
- Research Objective 3: Urban Heat Island Effect
- Summary of Plant Health & Design Considerations
- Research References
Since the doors to the U.S. EPA Region 8 office building in Lower Downtown Denver first opened in 2007, one of the most popular attractions of this green building has been the vegetated or “green” roof. Rightly so, it was the first of its kind in Colorado.
The green roof at the EPA regional office is an extensive, modular design, meaning that the plants are growing in 2’ x 4’ trays made of recycled plastic. These trays cover 20,000 square feet, or almost 60% of the 35,000 square foot total roof surface, above the 8th, 9th, and 10th floors of Region 8 EPA headquarters. When installed, the trays contained 40,000 individual green roof plants rooted in 4 inches of growing medium. Four species of Sedums were selected, including Sedum album (white stonecrop), Sedum kamtschaticum (Russian stonecrop), Sedum acre (goldmoss stonecrop) and Sedum spurium (two row stonecrop). There are three varieties of Sedum spurium: ‘Dragon’s Blood’, ‘John Creech’ and ‘Red Carpet”. These plants were selected for visual appeal as well as survivability in harsh environments with temperature extremes, and for their capacity to withstand drought conditions better than most plants. The Sedums are widely used throughout the world by the green roof industry.
We anticipated that the green roof would face challenges from limited natural precipitation, increased solar radiation, high wind velocities, predominately sunny days and the thin 4 inch layer of engineered soil medium. As a result, we initiated a series of research studies led by a team of EPA researchers from Region 8 and the Office of Research and Development’s National Risk Management Research Laboratory, together with partners from Colorado State University’s Department of Horticulture and Landscape Architecture, the Denver Botanic Gardens, the Urban Drainage and Flood Control District, the Alliance for Sustainable Colorado, and the City of Denver. The studies involved evaluating the biological health of green roof plants in a high elevation semi-arid environment, green roof stormwater management, and mitigation of the urban heat island (UHI). In addition, because of the uncertainty of long term survivability of sedum plants in this green roof design and in this climate, 2000 square feet of the green roof was dedicated to research on these additional plant species, 6 of which are native to Colorado (noted with an asterisk):
- Allium cernuum (nodding onion)*
- Antennaria parviflora (small-leaf pussytoes)*
- Bouteloua gracilis (blue grama)*
- Delosperma cooperii (hardy ice plant)
- Eriogonum umbellatum aureum (Kannah Creek ® Buckwheat)*
- Sedum lanceolatum (lanceleaf stonecrop)*
- Sempervivum ‘Royal Ruby’ (hens and chicks)
- Opuntia fragilis (Brittle Prickly Pear)*
Research Objective 1: Biological health of green roof plants in a high elevation semi-arid environment
Due to the porous and well-drained nature of the typical growing medium used in extensive green roof systems, the success or failure of an extensive green roof is primarily dependent on a plant species’ ability to grow in the media. These challenges are intensified for extensive green roofs in areas characterized by high elevation and semi-arid climate since low moisture content of the growing medium is a particular challenge for the plants’ ability to survive. For this reason, plants adaptable to dry, porous soils are primarily used in extensive green roof applications. Although Sedum species have dominated the plant palette for extensive green roofs, B. gracilis, D. cooperi, A. parviflora, O. fragilis, and S. lanceolatum are appropriate, if irrigated, for extensive green roof systems. (Bousselot Dissertation, 2010, or Final Report, 2012). Diversifying the plant palette of green roofs, especially with native species, will potentially open additional habitat choices for bird and insect species in urban areas.
Succulent plant species are more likely to tolerate longer periods of drought and resume growth soon after water is made available. Based on these results, irrigation frequency is recommended for succulent species at a maximum of 28 day intervals and herbaceous species at a maximum of 14 day intervals in the semi-arid, high elevation environment of the Front Range of Colorado. (Bousselot et al., 2011).
The green roof was originally installed with a drip irrigation system. However, because the drip lines were placed on top of the trays instead of subsurface, the exposure to sunlight caused them to degrade. It is also suspected that the engineered medium did not allow for lateral transport and spreading of water across the trays, affecting the growth of the plants. As a result, the drip system was replaced in 2009 with a spray irrigation system. Current research supports that the spray system is more efficient for extensive green roofs than drip irrigation at supplying uniform distribution of water. (Craig Greenwell personal communication).
During the first 5 years, the green roof was irrigated between the months of May and October. Watering recommendations are seasonally dependent. Spring irrigation (March, April, May) may or may not be necessary depending upon the amount of natural rainfall and the temperature. In contrast, the summer months (June, July, August) are some of the most crucial months for irrigation and supplemental watering for areas that require more water. (Fusco O&M Manual, 2010)
Winter watering (December, January, February) can be very useful for maintaining healthy plants during warmer, dry periods. In very dry years watering should be considered to help prevent plant loss due to desiccation. In seasons where regular snowfall is accumulating and melting on the green roof, winter watering will not be necessary. A good example of when winter watering would have been helpful was the winter of 2009. (Fusco O&M Manual, 2010)
Shading beneath or at the edge of the photovoltaic panels produces a synergy with the green roof substrate remaining cooler and therefore the irrigation requirements are lower. This could be an important water conservation measure. (Fusco O&M Manual, 2010).
When EPA's roof was designed, it was a non-standard practice for mitigating stormwater runoff. Therefore, the building developer had to provide significant data up front without reasonable assurances that the project would be either approved by the local drainage district or effective in its designed purpose of detention and infiltration of stormwater.
Stormwater performance data collected from the green roof and compared to a conventional, gravel roof on a building across the street supports our hypothesis that our green roof is effective at detaining and infiltrating stormwater runoff. This is especially true for snowmelt events and for smaller precipitation events (generally <1" rainfall in a 24-hour period).
Data from EPA's green roof were used to establish green roofs as an approved Best Management Practice for infiltrating and detaining runoff by the Urban Drainage and Flood Control District. As an approved Best Management Practice, developers can now choose to install a green roof without additional analysis using design specifications and performance criteria provided directly by Urban Drainage.
EPA's green roof was also used to help guide how green roofs can be designed effectively in semi-arid climates. The University of Colorado at Denver, working with EPA staff and Urban Drainage, developed a more comprehensive resource called Design Guidelines and Maintenance Manual for Green Roofs in the Semi-Arid and Arid West.
By providing data and working directly with Urban Drainage, EPA was able to demonstrate the effectiveness of this technology for future development of green roofs in the semi-arid west. With the availability of design guidelines and standard software for modeling stormwater runoff reductions, developers will be better positioned to design, construct and access the many benefits of green roofs.
The term "heat island" describes urban areas that attain higher temperatures than nearby rural areas. The annual mean air temperature of a city with 1 million people or more can be 1.8–5.4°F (1–3°C) warmer than its rural surroundings. In the evening, the temperature differences can be as high as 22°F (12°C). Heat islands can affect communities by increasing summertime peak energy demand, air conditioning costs, air pollution and greenhouse gas emissions, heat-related illness and mortality, and water quality.
Green roofs provide shade and remove heat from the air through evapotranspiration, reducing temperatures of the roof surface and the surrounding air.
In order to measure temperature benefits of our green roof, we compared results against those from a conventional, gravel roof on a building across the street (the control roof). In comparison to the control roof, the temperatures on the green roof at 1595 Wynkoop are lower during hot weather and higher during cold weather than those of the conventional roof. The cooling and warming properties of the green roof center upon the retention of moisture in the growing substrate and the plant materials. Temperatures on the green roof therefore are less extreme than those of the conventional roof and thus the green roof waterproof membrane does not undergo as much physical stress as that of the conventional roof waterproof membrane, which may explain why green roofs reportedly last 2 to 3 times longer. Because of this, the amount of solid waste from not having to re-roof as frequently lowers the amount of construction debris landfill waste, and saves money on roof maintenance.
The predominant plants found on the green roof are non-native Sedum species. These Sedum species and their varieties were chosen because of their ability to hold water and they are the plant species predominantly used in the green roof industry today.
In 2010 (3 years after the green roof was planted), EPA became concerned about the green roof’s appearance and its ability to support adequate plant coverage. In some places, plant growth and overall coverage, especially on the two highest levels, had noticeably decreased and some plants had died.
Through our research with CSU and an evaluation by Green Roof expert Mark Fusco of Bison Consulting, the following conclusions were made:
- Adequate plant coverage is impacted by soil depth and loss of soil has been a problem on the EPA green room due to wind scour. As a result, plants are more vulnerable to temperature fluctuations, particularly when soil depth is 2” or less. This is especially pertinent in Denver’s climate, given the extreme fluctuations in temperature. (Fusco O&M Manual, 2010)
- Soil medium depth of at least 4” is important in supporting evaporation and transpiration of water to help maintain a constant temperature across the green roof, thus providing a fully functional green roof habitat. (Bousselot Dissertation, 2010).
- Plant death is likely due to intense UV and hot temperatures from the plastic tray edges (Fusco O&M Manual, 2010)
- Warm soil temperatures during warm winter periods provide a signal to the plant to end its dormancy cycle and should that occur during freezing temperatures, the plant becomes more susceptible to die-back. (Fusco O&M Manual, 2010)
- In the fall, it is recommended that dead flower heads and stems be cut back. All seeds should be shaken out of their seed pods onto the green roof to encourage the germination of new plants the following spring. (Fusco O&M Manual, 2010)
- Shading is beneficial to plant growth
- Reflective heat from windows and metal siding inhibits plant growth; Cacti should be planted in the areas with the highest reflective heat
- Tray edges get hot in the summer which inhibits plant coverage
- Wind scour is the main cause of soil loss; parapet walls would help prevent wind scour
- Winter watering should be implemented but only under the right weather conditions
- Exposed irrigation lines are subject to degradation from UV rays
In summary, the combination of no winter watering, desiccating winds, hot and cold temperature fluctuations and a depleted soil depth are leading to higher plant death rates and reduced plant coverage in several areas of the green roof. As a result, it is recommended that a built in place system, with a total soil depth of roughly 5”- 6” across the entire surface of the green roof, would be preferable for maximizing plant health and coverage. (Fusco Justification Report, 2010)
The knowledge gained from EPA’s green roof research has provided an opportunity to educate others in the best approach for green roofs in the West.
Bousselot, Jennifer M.; Klett, James E.: and Koski, Ronda D. Extensive Green Roof Species Evaluations Using Digital Image Analysis. HORTSCIENCE 45(8): 1288-1292. 2010.
Bousselot, Jennifer M.; Klett, James E.: and Koski, Ronda D. Moisture Content of Extensive Green Roof Substrate and Growth Response of 15 Temperate Plant Species during Dry Down. HORTSCIENCE 46(3): 518-522. 2011.
Bousselot, Jennifer M. 2010. Extensive Green Roofs in Colorado: Plants Species Performance, Growing Media Modifications, and Species Response to Growing Media during Dry Down. Ph.D. Dissertation. Colorado State University, Fort Collins, CO.
Fusco, Mark. EPA Region 8 Headquarters Operations and Maintenance Manual. Bison Green Roof Consulting Services. 2010.