Green Roof Substrates and Their Potential Effects on Plant Growth

Abstract

Green roofs consist of overlapping layers that function as waterproofing, root barrier, drainage, substrate, and vegetation. Substrate components are designed to be relatively light weight, to resist degradation, and to drain rapidly. Physical characteristics must meet industry standards (FLL Guidelines, 2002) with water retention determined using 15 × 16.5 cm (diameter × height) cylinders (cyl) containing ∼1766.3 cm³ of substrate. However, green roofs may have a depth as shallow as 4 cm and slopes that affect water-holding properties; thus, a single protocol may be insufficient. Research on green roof physical properties of substrates is lacking. In this study, we evaluated the physical characteristics of a green roof substrate using three different containers. Also, physical characteristics were determined for a preexisting green roof. Particle size distribution was determined by screening using three air-dried 100 g samples of green roof substrate placed into the top of a sieve series with mesh diameters of 9.5, 4.0, 2.0, 1.0, 0.5, and 0.053 mm followed by shaking for three minutes in a Ro-Tap shaker. Physical properties were determined at 0 kPa and following applied suction pressure (6.3 kPa) using methods of Spomer (1990) and FLL (2002). To determine substrate physical properties, Buchner funnels with removable 17 × 16.5 cm-cyl or 13 × 6.8 cm-cyl (diameter × height) were filled with 2835.8 cm³ or 902.1 cm³ of substrate, respectively. Bulk density, total porosity (TP), maximum water-holding capacity (∼container capacity [CC]), aeration porosity (AP), and AP_{-6.3 kPa} were determined. A rectangle (rec)-shaped container (∼15 × 17 × 7 cm; width × length × height) was filled with 1158.9 cm³ of substrate directly from an existing green roof (Temple University, Ambler, PA) or from prepared substrate and, subsequently, physical characteristics were determined at an approximate 13.5° slope. There were three replicates per treatment (container type). Prepared substrate consisted of heat-expanded clay with a composition of 40:50:10 fine grade:medium grade:compost. Temple University's green roof consisted of a mixture of more than one component and has supported healthy Sedum, Allium, and Dianthus genera for several years.

Substrate composition and container shape had a significant impact on physical property determinations. There were no differences for TP, CC, or AP between 17 × 16.5 cm-cyl and 13 × 6.8 cm-cyl or ∼15 × 17 × 7 cm-rec. However, TP differed between 13 × 6.8 cm-cyl (TP=38.1%) and ∼15 × 17 × 7 cm-rec (TP=45.7%). Physical characteristics on a healthy green roof were 55.8%, 49.6%, and 6.2% for TP, CC, and AP, respectively, and within FLL standards for container capacity. Particle sizes of both prepared substrate and substrate on Temple University's green roof were within FLL standards; however, the later substrate had higher TP and CC than other treatments. Except for short durations following an irrigation event, green roof substrates may be perpetually dry or nearly so. If so, then green roof substrate CC values, and their hydraulic properties, are the key determinant for plant growth. Substrate depth determines the allowable vegetation on green roofs (FLL, 2002) and an accurate evaluation of substrate physical and/or hydraulic properties is vital. In addition to water retention characteristics further testing of substrates may include water potential monitoring using mini-tensiometers, or other moisture probes, and water release characteristics.