The role of living roofs in holistic stormwater management systems - Living Roofs in Integrated Urban Water Systems - page 41

to behave similarly to low through natural soils. The latter has been extensively
studied, but the former is only beginning to be explored experimentally. Physical
and hydraulic properties of the engineered living roof (and bioretention) media
are different from those of natural soils. For example, compared to soils, living
roof media generally have coarser particles (and particle size distributions), higher
porosity and higher saturated hydraulic conductivity ( K s ). Thus adapting models
developed for natural soils to systems with engineered media must be
approached with caution, as some parameter values may not be appropriate, or
the underlying model assumptions may be invalid.
2.7.3 Detention models
The majority of modeling and experimental research has focused on the retention
ability of living roof systems. Indeed, runoff retention (reduction in total volume
discharged) always enables detention (reduction of peak low), as there is physically
less water to discharge from the system, as was discussed in Sections 2.4 and 2.5 .
Nonetheless, signiicant permitting objectives revolve around detention
performance. Stovin et al. (2013) observed that in practice, runoff may be initiated
before ield capacity is reached due to heterogeneity in the growing media
(including composition, root development, compaction, etc.). This would cause
differences in modeled low rate and timing (i.e., detention) in water balance
approaches for retention performance as described in Section 2.7.1 and 2.7.2 .
Runoff water balance models alone may or may not account for detention
effects of the growing media, depending on model structure, and thus the ability
to predict peak low rates of a living roof's discharge. The modiied Green-Ampt
model tested in She and Pang (2010) and She et al. (2010) veriied relatively
good agreement of modeled and measured peak lows from both sites for which
the model was calibrated, as did the HYDRUS 1-D model by Hilten et al. (2008).
Kasmin et al. (2010) introduced a single storage-routing model that was reined
by Vesuviano et al. (2014) to isolate detention inluences of the growing media
and a synthetic drainage layer. The living roof assembly model's reinement was
underpinned by laboratory experiments measuring growing media layer detention
(Yio et al. 2013) and drainage layer-speciic modeling (Vesuviano and Stovin
2013).
Ultimately, most currently available, stand-alone synthetic drainage layers
provide little low resistance, as is their design intention (Bengtsson 2005; Yio et
al. 2013). Drainage layers integrated into synthetic modular trays are not
considered in the same category. Where designed for free low, the need for a
stand-alone synthetic drainage layer's discrete inclusion in a living roof model is
questionable.
2.8 discussion
Studies documenting annual, seasonal or other longer-duration performance in a
comparison against rainfall amounts or conventional roof runoff (e.g., reporting