Socio and Bio-physical Benefits

Green Roofs Tree of Knowledge | Social and Bio-Physical Benefits

Publication Information

Title: Native Plant Performance on a Seattle Green Roof
Author Surname: Martin|Hinckley
Author Firstname: Melissa A. |Thomas M.
Journal/Conference/etc.: Fifth Annual Greening Rooftops for Sustainable Communities Conference
On pages/Number of pages: 13
Publisher: Green Roofs for Healthy Cities


Native plant survival and vigor were monitored on a recently installed 650 m2 extensive green roof at the Woodland Park Zoo in Seattle, Washington. Spatial and temporal patterns of plant performance and abiotic conditions were documented during the summer and fall of 2006. Results suggest that water availability was the most influential factor in determining patterns of plant performance during the first summer after installation. Degree of canopy shade also contributed to plant performance by controlling ground surface and substrate temperatures, thus heightening (in sun) or mitigating (in shade) drought effects. This exploratory study provided indications of both plant performance and how roof architecture and surrounding shade elements potentially affected microsite water status.


Purpose of Research

The purpose of the research was to get a general understanding about green roof performance in the Pacific Northwest, as well as to assess the palette of green roof plant species able to succeed in the mild, Mediterranean climate of the region.


Benefits Addressed

Bio-Physical Benefits:

No bio-physical benefits associated.


Research Conditions

Type of green roof studied:

Size or range of sizes of green roofs studied (square feet):

Type of Construction:

Plant Type:

Growing medium depth or range of depths (inches):

Research period:   From: JUN-2006 To: NOV-2006

Other conditions of research:

Slopes of the roof varied from 0:12 (0%) at the upper west edge to 3:12 (25%) at the lower east edge

The green roof design consists of four distinct planting zones:
1) kinnikinnick (Arctostaphylos uva-ursi) and nodding wild onion (Allium cernuum)
2) 3 blue-pod lupine (Lupinus polyphyllus) and sand strawberry (Fragaria chiloensis)
3) 2 salal (Gaultheria shallon) and sword fern (Polystichum munitum)
4) 3 blue eyed grass flower (Sisyrinchium douglasii) and kinnikinnick (Arctostaphylos uva-ursi)

The roof was divided into 2, 4, or 6 sections (depending on gardening staff) for irrigation, and each section was typically sprayed for 15-30 minutes.



Substrate moisture and temperature data from multiple locations on the roof surface were collected continuously throughout the study period. The substrate moisture and temperature probes were at -9 and -6 cm, respectively (Campbell Scientific: CS616-L water content reflectometer and 107-L temperature sensor). Sensors 1 and 3, referenced in this paper, are located in the SW corner and center of the roof, respectively.
Hemispherical photography was used to compare percent open sky across the roof surface. 180˚ photographs were taken with a fisheye lens looking skyward from a near-ground position (0.5 m tripod height). Photographs were taken at each of the 21 plant vigor subplot locations on two completely overcast days, November 3 and 6, to enhance the contrast between the sky and any shading element.
Survival and vigor of individual plants was monitored during the summer and fall of 2006. Five transects at equal 7.34 m intervals were established perpendicular to the slope of the roof (east-west). Each transect was divided into 20 – 1 m2 plots, and plant density (number of plants of each species) was measured in each plot. Sampling occurred on July 7, August 2, September 5, and October 12. A subset (3 to 5 per transect) of the plant density plots was selected to track plant vigor. Vigor was assessed approximately once each week in July and August, the hottest and driest months of the year, and once every two to three weeks thereafter until hard frost occurred.


Research Results

No research papers associated.

Other Notes

Arctostaphylos uva-ursi and Allium cernuum experienced drought stress during the summer season. No other stress, such as nutrient deficiency, was observed for either species. For Arctostaphylos uva-ursi, there was a distinct decline in survival for plants on slopes >15%, whereas survival rate was >70% at all positions of shallower slope. Allium cernuum survival was >70% for all roof slopes.
There was a relationship between plant performance and proximity to the edges of the roof, particularly the north edge.
Overall, Fragaria chiloensis vigor declined during the summer, and many individuals lost all above-ground biomass.
Gaultheria shallon and Polystichum munitum both suffered during the summer and became completely dried in roof edge areas that received minimal water, but vigor was high throughout the summer in the southwest, shaded corner of the roof. Some dried individuals of both species developed new growth in the fall, suggesting dormancy rather than complete mortality. Almost all Sisyrinchium douglasii individuals were dried by early July, and there was some vegetative regrowth in the fall.

None of the known irrigation events created an increase in substrate moisture at a depth of 9 cm, where moisture declined without interruption from late June through early September. Since only a small depth of growing medium was moist after irrigation, it is probable that the compacted, dry substrate did not readily infiltrate water; instead, water moved quickly along substrate surface, particularly in areas with steeper slope, without becoming available to plant roots.

Purchase Link:

Purchase this paper

Summary Information

Summary prepared by:

Date: 18-SEP-2008

Additional Notes:

Associated Links: