COS 81-6
The influence of ecological design principles on semi-arid green roof ecosystem services

Wednesday, August 7, 2013: 3:20 PM
M100HC, Minneapolis Convention Center
Mitchell Pavao-Zuckerman, Biosphere 2, University of Arizona, Tucson, AZ
Sarah Kovachich, University of Florida
Margaret Livingston, School of Landscape Architecture, University of Arizona
Ronald Stoltz, School of Landscape Architecture, University of Arizona
Steven E. Smith, School of Natural Resources, University of Arizona
Background/Question/Methods

With the use of landscape planning and ecological design, the urban environment can be manipulated to enhance ecosystem services. Many approaches to ecological design of green infrastructure originated in mesic environments, leaving questions about their ability to function as desired when transferred to semi-arid environments and the potential for ecosystem disservices (such as high water costs) to outweigh any benefits that result. This study was conducted to test the function and appropriateness of green roofs in a semi-arid environment. We test whether using native species with local-environmentally appropriate irrigation regimes if the ecosystem service benefits associated with green roofs will result. A set of miniature model houses (1 m2) were constructed at Biosphere 2, outside Tucson, AZ with either shingled or green roofs (8” soil bed) to evaluate green roof performance in an arid setting. We used a nested design comparing soil composition, plant species, and irrigation approaches. Soils were composed of expanded shale, sand, and composted organic matter in a light or heavy soil mix (by maniupating the proporion of sand and organics). The roofs were planted with one of three species native to the Sonoran Desert that represent different functional types (a sub-shurb, forb, and succulent). One set of roofs received ambient rainfall, while the other received minimal supplemental irrigation based upon precipition inputs needed to meet potential evapotranspiration. Roofs were planted in October 2010, and soil and interior house temperatures were monitored through March 2012. Hyperspectral imaging was used to assess plant function and health in the dry and monsoon summer season in 2011.

Results/Conclusions

All plants were able to survive and function in the green roof soil mixes with both minimal and ambient watering treatments. Houses with the green roofs had reduced internal temperatures (3°C mean daily temperature and 1.2°C mean nighttime tempeartures) relative to the shingled roofs. Soil treatments and irrigation approaches had minimal but significant impact on temperatures. Significant species effects on house temperatures were detected (the sub-shrub consistently resulted in cooler house and soil temperatures), although this varied by season. Plant status and function varied by species, and was positively correlated with soil moisture content, although this relationship varied by species. This small-scale study suggests that arid-adapted species and ambient precipitation can bring green roof ecosystem services into arid cities. Environmentally appropriate design choices can thus impact local climates and support urban ecosystem functioning.