The role of living roofs in holistic stormwater management systems - Living Roofs in Integrated Urban Water Systems - page 33

nitrogen concentration compared to rainfall (Aitkenhead-Peterson et al. 2010;
Berndtsson et al. 2006; Gnecco et al. 2013; Hathaway et al. 2008; Teemusk and
Mander 2007). With the exception of living roofs in Auckland, Paris and Singa-
pore (Fassman-Beck and Simcock 2013; Seidl et al. 2013; Vijayaraghavan et al.
2012), when compared against conventional roof runoff, living roofs in Michi-
gan, North Carolina, Florida, Pennsylvania, Manchester (England) and Estonia
have been shown to either absorb nitrogen, or at least not discharge more (Bliss
et al. 2009; Carpenter and Kaluvakolanu 2011; Hathaway et al. 2008; Mendez et
al. 2011; Teemusk and Mander 2007; Wanielista et al. 2007).
Phosphorus monitoring studies largely conclude that living roofs are a source
of phosphorus when compared to either rainfall or conventional roof runoff
(Berndtsson et al. 2006, 2009; Fassman-Beck and Simcock 2013; Gnecco et al.
2013; Hathaway et al. 2008; Seidl et al. 2013; Wanielista et al. 2007). The
notable exception appears to be Carpenter and Kaluvakolanu (2011), who found
phosphate concentrations in living roof runoff were lower than those for the
conventional asphalt and gravel ballast roofs, albeit not with statistical signii-
cance. It is commonly reported that phosphorus concentrations are quite sub-
stantially elevated amongst any comparison presented.
Heavy metals, namely zinc and copper, appear to not discharge at concentra-
tions of substantial concern from the living roofs monitored for these parameters
(Fassman-Beck and Simcock 2013; Gnecco et al. 2013; Seidl et al. 2013). In each
of these studies, researchers hypothesized that the presence of zinc or copper in
building materials (e.g., lashing and cladding) were likely sources of metals in
runoff, rather than the living roof itself (although in the case of Auckland there
were also suspicions about contributions from the growing media). Again, the
presence of a compound in runoff does not by deinition imply a receiving envi-
ronment problem, which seems to be the case with regard to zinc and copper in
living roof discharge, with the little evidence available to date.
Substrate composition, fertilization practices, roof age and the presence or
absence of vegetation have been consistently identiied as parameters inlu-
encing runoff water quality, but their speciic inluence has not been fre-
quently measured. Indicators for growing media physico-chemical properties
to prevent nutrient leaching have been hypothesized based on agricultural
rules of thumb, but require signiicantly more data to test the suitability
(Fassman-Beck and Simcock 2013). In addition, for each study, spatial differ-
ences in the dry deposition of dusts and atmospheric aerosols (usually catego-
rized based on local zoning, e.g., industrial, residential or commercial, or
external inluences such as trafic intensity, etc.), supplemental irrigation and
climate/drought factors may contribute to runoff quality differences. Even
within studies, dificulties were encountered interpreting water quality results
due to contrary or inconclusive results from different runoff events. Relatively
short duration studies (and thus limited sampling) render it dificult to coni-
dently characterize living roof discharge water quality, or assess the effects of
system establishment or aging.