Quantifying the Complete Hydrologic Budget for an Extensive Living Roof

Abstract

The main aim of this work is to quantify the ability of an extensive living roof to provide onsite source control for stormwater mitigation. Stormwater mitigation ability was tested by quantifying the components of the living roof hydrologic budget. These included: runoff volume and peak flow rates, water storage within the substrate, and evapotranspiration (ET). Additionally, this research aimed to determine if common methods used to predict ET in agricultural applications could be used to predict living roof ET. Runoff response to rainfall was continuously monitored for one year from an established 235 m2 field installation living roof in Auckland, New Zealand. The extent of stormwater control was quantified by comparing rainfall and runoff across three different substrate types (Pumice, Zeolite and Expanded Clay), at two different substrate depths (50 and 70 mm), in a side-by-side comparison. ET was measured directly in greenhouse and field environments using bench-scale trials. Three tray configurations were tested: unplanted, planted with Disphyma australe (a native species) and planted with Sedum mexicanum (a non-native plant). ET was modelled using ten existing agricultural models. The living roof helped restore an impervious-site runoff hydrograph to a shape and magnitude more representative of a vegetated-site runoff hydrograph. This was achieved by delaying the onset of runoff, delaying peak runoff, and extending the duration of the hydrograph when runoff occurred. The living roof operated effectively year-round, exhibiting consistent volume retention and peak flow reduction. Annual cumulative retention of rainfall was 66% and median peak flow reduction was 93%. Living roof response could not be linked to one factor alone; multiple climatic parameters influenced stormwater mitigation. Plants provided substantial contribution to storage recovery within the system, with transpiration comprising up to 45% of total ET. The influence of water availability overrode the influence of climatic parameters on living roof ET rates. When water was available ET rates were up to 5.4 mm.d-1, but dropped to <0.7 mm.d-1 when water became limited. Commonly used agricultural models failed to accurately estimate daily ET from a living roof. Instead, empirical regression models for estimating ET are presented. They provide a simple and practical solution to allow use of ET in continuous hydrologic simulations. Overall, extensive living roofs demonstrate significant stormwater mitigation benefits through onsite source control.