in the news

Back to Guest Features

September 2012
guest feature article

Using Seeds as an Installation Method on Green Roofs


By Kathryn McDavid, Ph.D.
This is a short summary of author’s Ph.D. research at Penn State University.

All Photos and Graphics Courtesy of Kathryn McDavid






Seeds are rarely used as the primary installation method on a green roof and only used occasionally as an overseeding technique with cuttings and/or plugs.  Sedum species are the most common plant material used on green roofs in the Eastern United States.  Although there are recommended planting densities for the installation of cuttings or plugs, there are no specific recommendations as to the seeding intensity of green roofs.

Currently, very little published research exists on Sedum seed viability at various temperatures or specifically how the seeds mature over time.  These germination rates, time of the year, and the age of the seed significantly impact the success of a greenroof.

Sedum seeds range in size from small to extremely small (Figure 1).

Figure 1:
Example of Sedum seeds in a Petri dish.

This author’s research specifically investigated whether five Sedum seed species could be a viable option for plant installation for green roofs.  In particular, research was performed to test the germination rates of these Sedum species at various temperatures and seed maturations.  The five species used throughout the studies were Sedum acre, Sedum forsterianum ‘Oracle’, Sedum reflexum, Sedum selskianum ‘Spirit’, and Sedum spurium ‘VooDoo’.


Germination rates of the five Sedum seed species were tested at an optimal temperature (70 degrees Fahrenheit) and at a supra-optimal temperature (90 degrees Fahrenheit).  Both temperatures were selected because green roofs are often subjected to a range of optimum and supra-optimum temperatures during the planting season in Pennsylvania.  To determine the germination rates of aging seeds, the seeds were tested at these temperatures over the course of a year.  Seeds were placed in Petri plates in growth chambers at the respective temperatures.  An ideal species would demonstrate high germination rates at both temperatures and at all seed ages and would also have a short time to 50% germination.

Note: it was unknown how old the seed was when ordered; thus, all ages were based on storage time after the seed was received from Benary Seed.  Additionally, seeds subjected to the 90 degree Fahrenheit temperature were tested up to only nine months in storage; whereas, the 70 degree Fahrenheit seed studies were run through eleven months of storage.

For Sedum acre and Sedum reflexum seeds, the younger seeds generally produced faster and higher germination rates in both temperatures.  As the seeds aged, germination rates decreased and the time to reach 50% germination significantly increased.  Therefore, using fresh seed is ideal for getting the best germination rates at various temperatures for both species.

Sedum forsterianum seeds germinated at high percentages at all seed ages when subjected to the 70 degree Fahrenheit growth chambers but were very sensitive to increased temperatures and thus would be unlikely to produce adequate coverage on a green roof at any seed age (Figure 2).  It is possible that Sedum forsterianum could be successful in cooler climates or during cooler months of the year.

Figure 2: Sedum forsterianum germination percentages in both 70 degree and 90 degree Fahrenheit growth chambers.

Sedum selskianum seeds germinated at high rates at all ages tested in both the optimal and supra-optimal temperatures (Figure 3).  This research showed that Sedum selskianum seeds were unaffected by storage time or temperature during seed germination.

Figure 3: Sedum selskianum germination percentages in both 70 degree and 90 degree Fahrenheit growth chambers.

Although Sedum spurium ‘Voodoo’ seeds germinated at high rates at all storage ages in the 70 degree Fahrenheit growth chamber, the 90 degree Fahrenheit germination data showed the seeds germinated at the highest rates when stored less than five months (Figure 4).  There was an increase in germination data in the 90 degree Fahrenheit growth chamber at nine months.  More research, however, is required to determine whether this increase is significant.

Figure 4: Sedum spurium ‘Voodoo’ germination percentages in both 70 degree and 90 degree Fahrenheit growth chambers.


The first outdoor trials were performed on the Root Cellar Roof at The Pennsylvania State University, University Park campus, during the Summer of 2010.  Seed species were placed in randomized grids (Figure 5).  Unfortunately, due to various environmental factors and the large water droplets from an oscillating sprinkler, many seeds did not remain in their respective grids.  As a result, the seeds required a treatment to secure them near the surface of the roof.  Among the many treatments available, hydroseeding was the only option investigated in outdoor trials in 2011 (Figure 6).

Figure 5: One section of the Root Cellar Roof prepared for trials in 2010.
Figure 6: One section of the Root Cellar Roof prepared for hydroseeding in the 2011 trial.

Due to the number of treatments, seed species, and replications required for the studies, seeds were not added to a hydroseeder.  Instead, a thin layer of hydromulch was sprayed onto the surface of the roof media.  Seeds were then applied in respective grids.  Finally, another layer of hydromulch was sprayed on top of the seeds (Figures 7-10).

Figure 7: Scott Burk of Scott’s Landscaping in Centre Hall, Pennsylvania spraying the first layer of hydromulch.
Figure 8: Section of the Root Cellar Roof after the first layer of hydromulch; area is ready for applying seed treatments.
Figure 9: Applying the seed treatments to the hydromulched roof.  Some seeds were previously primed (priming research not discussed here) and both the treated and the untreated seeds were applied to respective grids on the Root Cellar Roof.  Sand was used only to facilitate the spreading of the seed.
Figure 10: Scott Burk spraying the top layer of hydromulch over the seed treatments.

This procedure allowed for encapsulating the seeds in the hydromulch as if they had initially been mixed with the hydromulch in the sprayer.  Additionally, rather than use an oscillating sprinkler as was utilized in the 2010 trials, misting irrigation heads were employed (Figures 11-12).  Irrigation was supplied for 30 minutes per day for the first three weeks after seed installation.

Figure 11:
Root Cellar Roof 2011 trial with seeds and irrigation installed.
Figure 12: Close up of the misting irrigation heads.

The main advantage of having a recommendation for seeding Sedum seeds is the reduction of roof installation costs.  According to Scott Burk of Scott’s Landscaping in Centre Hall, Pennsylvania, there is a significant difference in labor costs for installing a green roof through the use of plugs versus the use of seeds.

For example, one laborer can typically plant 50 plugs per hour or hydroseed a 2,500 square foot area in the same time.  If the cost for that individual (including wages, insurance, etc.) is $50 an hour, just the labor costs for planting a 40,000 square foot roof, using two plugs per square foot, would total around $80,000.  By contrast, labor costs to hydroseed a 40,000 square foot roof would be significantly reduced to about $800, due to mechanization.  As expected, though, actual cost savings will vary depending on hourly wages, cost of insurance, and productivity of workers.

Using the hydromulch to stabilize both the seeds and media improved germination rates and plant coverage in outdoor experiments.  The recorded germination data led to a preliminary seeding guideline per species, although, more research is necessary to establish a recommendation for a proper mixture of adding the seed species to the hydroseeder.  Furthermore, research is required to determine the longevity of plant coverage when installed through seeds.

Kathryn McDavid

Kathryn McDavid
Washington, DC

Katie recently completed her Ph.D. in Horticulture at The Pennsylvania State University where she conducted research on green roofs with Dr. Robert Berghage.  Her research focused on utilizing Sedum seed species as an installation method on green roofs.  She also has a B.S. in Horticulture from The Pennsylvania State University.  This is a short summary of author’s Ph.D. research at Penn State University.  For more information, please contact author.

She currently lives in the D.C. area and is looking for a horticulturist position, ideally one that would involve green roofs and incorporate sales, research, maintenance, or project management; however, she is open to any opportunities which would utilize her skills. 

Contact Katie at:

The opinions expressed by our Guest Feature writers and editors may not necessarily reflect the beliefs of, and are offered to our readers to simply present individual views and experiences and open a dialogue of further discussion, debate and research.  Enjoy, and if you have a particular comment, please contact the author or send us an email to:


Back to Top



Ad: American Hydrotech