Greenhouse gas exchange from green roof ecosystems

Abstract:

Greenhouse gas (GHG) surface-atmosphere exchange is an important indicator of ecosystem function, and is of relevance for global climate change and related ecosystem feedbacks. A large and growing research effort has been put towards quantifying GHG exchange from the earth’s surface. Yet, there are important urban ecosystems about which we know next to nothing regarding their GHG exchange rates. One of the notable knowledge gaps is for green (vegetated) roofs, an increasingly common land cover in cities as municipalities work to green dense urban landscapes to promote storm water runoff and other ecosystem services. In this study, we used static chambers to characterize the net exchange of the greenhouse gases CH₄ and N₂O, as well as respiration rate, for green roofs in the Cincinnati, OH region. A full year of measurements were made at intervals of approximately every other week, for six green roofs varying widely in characteristics, including four shallow-substrate extensive green roofs, and two deep-substrate intensive green roofs. Respiration, measured as CO₂ emission rate in dark chambers, showed a strong seasonal pattern as with natural ecosystems, with emission rates positively correlated with temperature. Estimated annual ground-layer respiration varied 7-fold among the different roofs, from 119-837 g C m^-2 yr^-1; similar to sparsely vegetated natural ecosystems in the region. CH₄ exchange was negligible except for one intensive green roof that had soil, shrubs and small trees – this ecosystem functioned as a CH₄ sink (averaging -14.5 mg C m^-2 h^-1) similar to that measured in other urban forested ecosystems. N₂O exchange was negligible from the four shallow-substrate extensive green roofs, but the two intensive green roofs emitted N₂O at average rates ranging from 6 - 12 mg N m^-2 h^-1; though still much lower than rates from fertilized managed ecosystems. On balance, green roofs were a minor sink for CH₄ and a minor source for N₂O, with variability among roofs likely driven by variation in vegetation and substrate characteristics. Our results suggest that the direct exchange of GHG from these ecosystems is a relatively minor contributor to their overall GHG impact, compared to energy costs associated with construction, and energy savings associated with enhanced insulative and cooling properties of the roofs.