SS 18-8
A photovoltaic array on a green roof in Denver, Colorado, improves resilience of green roof plants by expanding habitat heterogeneit

Tuesday, August 6, 2013: 9:05 PM
L100C, Minneapolis Convention Center
Thomas J. Slabe, Environmental Protection Agency
Jennifer Bousselot, Centenial, CO
Background/Question/Methods

The first modular, extensive green roof in Denver, Colorado, USA, was installed on the ten-story U.S. EPA office building in 2007. This high elevation, semi-arid, short grass prairie region has frequent “false springs”, hard freezes, and soil moisture deficits that challenge plants, especially those on roof surfaces rooted in 10 cm depth (4 in) porous engineered substrate. Five-minute readings of temperatures, solar irradiance, infrared radiation, precipitation, and other parameters, using 14 sensors produced high resolution data for two full years. Another study employed 12 temperature dataloggers (Onset Hobo Tidbits) spaced vertically and horizontally on and beneath the substrate surface. From June through October, 2009, 30 minute average temperature values were logged from exposed areas and areas beneath photovoltaic panels, which are on a support rack 1.5 meter above the roof surface. Studies were conducted to quantify spread (growth) and overwinter survival of eight plant species in exposed areas verses those in the AM shadow of PV panels. Due to the unique nature of these efforts conducted in situ on two buildings and resource constraints, study designs suggest a generalized yet comprehensive approach, verses a rigorous design with multiple replicates.

Results/Conclusions

Plants within the shadow of PV panels spread to a greater extent, reducing bare substrate. Overwinter survival of plants in shaded areas was 95% versus 60% for plants in exposed plots. Temperature variation beneath PV panels was markedly lower, e.g. for substrate surface beneath the panels and in exposed areas, respectively, standard deviations were 3.92°C and 6.36°C while that beneath the substrate in sheltered and exposed areas were 1.18°C and 3.94°C. The panels appear to function somewhat as canopy structure, mimicking natural ecosystems. The shadowing effect from panels serves to establish gradients in substrate moisture, solar irradiance, and temperatures. Results indicate that shading structures on green roofs expand habitat heterogeneity, improve performance of the plant community and expand species diversity potential on green roofs. Transforming temperatures into power (W/m2) to estimate net thermal radiation (using Stefan-Boltzmann’s equation), shows that beneath the panels and substrate surface is a thermal sink relative to exposed roof surfaces. The results indicate that green roofs with structural features that increase thermal sinks will improve resilience. Photorespiration and evapotranspiration rates of shaded plants are likely more optimal relative to those of exposed plants.