Soil-less Soil




By Rebecca Popowsky, Research Associate at OLIN


The Soil-less Soil project’s goal is to create healthy green infrastructure soils for the urban forest and parks of Philadelphia and the Delaware Valley that are 100% renewable sourced from a combination of recycled glass fines, Class A Biosolids, on-site mineral soils, and other rapidly renewable materials that achieve consistent results and preserve the region’s topsoil for farmland and natural preservation. The Soil-less Soil project aims to divert one of the region’s largest, most expensive waste streams—glass—into a rapidly reusable local component of urban soil blends. By ending the need to disruptively mine sand from pits throughout the region, this project will both close a major regional waste loop and reduce the peri-urban development pressures that accompany this extractive industry. Second, this project is focused on designing an urban soil and crushed glass blend that optimizes water quality and plant health. Finally, this project is ultimately about accelerating the green stormwater infrastructure development process. By developing an optimal, standardized urban soil blend, this project promises to provide a means of building better green infrastructure more quickly and economically.

Photo by Sahar Coston-Hardy, 2018.

Photo by Sahar Coston-Hardy, 2018.


Soil-less Soil is positioned at the nexus of three pressing environmental issues associated with urban development, namely (1) achieving vegetative green stormwater management, (2) overconsumption of natural resources and (3) the overburden of landfills with reclaimable materials.

Green Stormwater Infrastructure:

The Green City, Clean Waters plan has made Philadelphia a national leader in green stormwater infrastructure innovation since its adoption by the Philadelphia Water Department in 2011. However, If the city fails to build and maintain enough Green Stormwater Infrastructure (GSI) to meaningfully reduce its CSO problem, it is likely to face additional regulatory action from the EPA. Not only would that be a significant financial burden for Philly, but it would greatly harm the GSI movement nationally. In order to meet the city’s ambitious 25-year goals to green one-third of the impervious area in the combined sewer system, we have to find ways to build more GSI in Philly and at a faster rate. One significant gap in achieving green acre goals is a lack of cost effective and reliable biofiltration soils. The Water Department’s soils specifications are performance-based, meaning the specific make-up of of a soil mix relies on the contractor to procure, mix and install material that will support living systems and provide the necessary stormwater infiltration.  Since natural materials like sand and mined topsoil are never identical from one source to the next, the process for testing and accepting soil blends for GSI installation is time-intensive, often requiring mixes to be recalibrated. Frequently, time and budget do not allow for this level of quality control and the installed plants and infiltration capacity suffer.

Urban soils typically do not have the properties required for sustainable stormwater infiltration, or to adequately support plant growth in an urban environment. They can be overly compacted, reducing both water and gas infiltration, root elongation, and nutrient cycling. Urban soils are often contaminated with both chemical pollutants and anthropic material such as concrete, steel, brick and other construction debris. For this reason, soil organic and mineral materials must be sourced off site, typically mined from natural and rural areas. Soil mining has impacts on watershed health that extend the urban footprint well beyond its boundaries, and include the impacts of natural material extraction (i.e., sand, gravel and topsoil mining) and those of waste management and disposal. The Soil-less Soil initiative aims to reduce threats to waterway health from raw material extraction (“upstream”) and waste disposal (“downstream”).  

Resource Extraction:

Recent scientific and media attention to the environmental, social, and human health impacts of sand and gravel over-extraction, as well as the dangers of our ever-growing reliance on the material group for modern construction (as a component of concrete, asphalt, and for grouting oil and gas exploration wells) highlight the need to identify feasible sand and gravel material substitutes. Sand and gravel represent the most heavily extracted material group on the planet, ahead of fossil fuels and biomass, and their extraction and transport produce a host of negative environmental impacts including disturbance of habitats, watersheds, ecological communities, and food webs (Torres et al. 2017). While our region does not face short-term shortages of sand or gravel, the negative environmental impacts of sand and topsoil mining do impact the areas of heavy extraction, including the Pine Barrens of New Jersey and in alluvial zones along the Delaware River.

Reclaimable Waste:

Municipalities including Philadelphia are confronted with the challenges of materials that are recyclable, in theory, but have run into market or technological barriers; glass is a textbook case. Typically, glass containers are broken or shatter on their way to a recycling facility; from there they must be sorted and sent to a glass manufacturer that can accept the material and reconstitute it as a glass product. This entails endrun hauling and freighting to remote facilities that make glass shards more expensive than raw sand to source. Additionally, glass particles that are smaller than ¼” cannot generally be reused to make new glass as they are too difficult to clean and separate. Therefore, most glass recycling processes create waste by-products in the form of glass fines that are sent to the landfill. Serendipitously, these low-value glass fines are appropriately sized for use in soils.

Philadelphia is currently experiencing a crisis in the management of recyclable waste, due in large part to far-reaching changes to recycling rules in China. These were enacted in January 2018 and have drastically reduced demand in recycling markets. This year alone, the cost to the city of Philadelphia to manage recyclables has increased nearly tenfold (from $4 to $35-$37 per ton). Because of this recent market shift, thousands of tons of recyclable material left curbside in dozens of American cities have gone to landfills (Albeck-Ripka 2018).

Additional Opportunity

Soil media designed for plant growth typically contain an organic component in addition to the mineral components (e.g., sand, silt, and clay). In the context of urban GSI, a means of further reducing the environmental footprint would be to incorporate class A biosolids (i.e., processed sewage approved for use in any setting) as the organic component. The use of biosolids as an organic amendment in public works is underdeveloped, as many municipalities simply use it as a landfill cover. However, the technology to generate safe biosolids is well established and numerous fertilizer products derived from biosolids are on the market. Using biosolids in GSI settings could redirect a large volume away from landfills and provide a high-quality and more locally-sourced alternative to current options.

LCA_Soil_Boundary Diagram.jpg


With regard to a glass-for-sand substitute in planting soil, several studies indicate ground glass would perform as well or better than a typical sand-based soil mix in terms of infiltration, environmental impact, and plant vitality. A recent study found that utilizing recycled glass waste as a partial sand replacement in sand drains improves permeability performance and accelerates drainage consolidation (Wang et al. 2017). This study did not consider the glass-sand soil as a planting medium. Moreover, a life-cycle assessment conducted in Hong Kong showed that recycled aggregates from construction and demolition waste as well as waste glass have environmental benefits, such as reduced greenhouse gas emissions and reduced non-renewable energy consumption when compared to virgin sand and gravel aggregate (Hossain et al. 2016). No similar study has been published in the United States. Finally, a study of the use of recycled glass cullet as artificial dune fill found that dune-stabilizing vegetation in a recycled glass medium outperformed those specimens growing in the natural beach sediment controls (Makowski et al. 2016).


A thoroughly tested and proven use for low-market-value, post-consumer waste as a high-performance soil medium will create powerful synergies among several local initiatives in Philadelphia.

While Philadelphia’s Green City, Clean Waters initiative is now well-established, the program has enough years remaining for this to have a very large impact. In their 2017 Stormwater Program Annual Report, PWD listed 608 different GSI installations as being maintained, of which 321 were tree trenches and stormwater trees. Assuming a standard tree-pit size of 3x4x10 ft, the cumulative amount of soil utilized for the tree-containing structures contains approximately 27,000 cubic feet (or 1,500 tons) of sand. Over the next 19 years, with a similar rate of project implementation, this quantity may quadruple. In order to overcome the quality control barriers described above and achieve economies of scale, the Soil-Less Soil project proposes to source primary soil mix components for the City from a ready municipal supply of waste glass. This inert material would offer a uniform supply of component material, eliminating chemical and physical variability. Similarly, Class A biosolids represent a highly uniform and consistent source of organic matter, eliminating variability in nutrient supply, decomposition rate, weed seed contamination, etc. In addition, these materials are price-competitive due to their common availability throughout the US. By developing an open specification for municipal waste sourced top soil, the Philadelphia Water Department could ensure quality, efficacy, and cost effectiveness in the implementation of green acre projects.

As part of the Philadelphia Greenworks initiative, the City published the 2016-2025 Municipal Waste Management Plan (MWMP), which established an ambitious goal to divert 70% of solid waste from landfill disposal. With the current crisis in US recycling markets described above, the difficulty of meeting this goal is amplified and innovative ways to reduce waste are sorely needed. Until recently, the City of Philadelphia was able to make a profit on the sale of recycled materials. In recent months, the cost to the city of exporting these same materials has skyrocketed. Keeping some portion of the city’s waste glass within city limits, for direct local use, would represent substantial cost savings. Some of these savings could be passed on to city agencies, such as the Water Department or Parks and Recreation, that use this recycled material in place of virgin sand.

In 2017, the Office of Sustainability at the University of Pennsylvania documented over 15 tons per month of pre-consumer composting, which came from Penn Dining Cafes and local restaurants serviced by Penn, being diverted from landfills. The office has taken an advisory role in the Soil-less Soil project in the hopes of finding a way to close the glass waste loop on campus. OLIN Labs is currently working with four Eco-Rep students at UPenn, and one research assistant with the Ian McHarg Center, to write a feasibility study describing the potential use of campus glass waste in new landscape installations and in ongoing maintenance and operations. This study will be completed during the 2018-19 academic year.

In a recent talk at UPenn, Philadelphia’s Managing Director Michael DiBerardinis expressed the need to work across city departments to address pressing social and environmental challenges. The Soil-less Soil project team has assembled a steering committee that can help to cross departmental and institutional boundaries for maximum applicability and scalability of study outcomes. Committee members include Howard Neukrug, Director of the Water Center at Penn and former CEO of the Philadelphia Water Department; Dan Garofalo, Director of Sustainability at Penn; Nic Esposito, Director of the Zero Waste Cabinet for the City of Philadelphia; Jason Grabosky, Director of the Center for Urban and Community Forestry at Rutgers; and Jason Lubar, Urban Forester at the Morris Arboretum.


The study compares the performance of Soil-less Soil to that of a typical GSI soil mix through three lenses:

  • As a stormwater management system: infiltration rates, structural capacity, leachate quality.

  • As a planting medium: plant growth, root mass, mortality.

  • As a sustainable material: life cycle assessment of material production in terms of energy, greenhouse gas emissions, resource use, pollution. The life cycle assessment (LCA) will be the first published in the US comparing crushed glass to virgin sand as a construction material.

The goal of the research effort is to establish the feasibility, safety, and functionality of a renewable soil mix, following a three-step process:

(1) The first phase of the study, which is currently underway, is a laboratory-based analysis whose goal is to determine the optimal mixture of fine ground glass, biosolid compost, and mineral soil to achieve the physical and chemical properties required for use in GSI tree trenches. We will prepare three replicate batches of each of a broad array of soil mixes; these will include substitutions of glass cullet and biosolid compost for sand and [typical organic matter] at several mixing ratios. In all cases, the remaining component of the mix will be a mineral soil with silt and clay fractions, as this will be needed to ensure appropriate values of some chemical properties. We will also vary the size distribution of the glass cullet. We will then evaluate a suite of chemical and physical properties using samples from each batch, including, for example, cation exchange capacity, base saturation, nutrient concentrations (both adsorbed to the soil and in leachate), pH, water retention, saturated hydraulic conductivity, and infiltration. The resulting data will allow us to identify the mixes that have the most promise for performing well in GSI settings.

(2) The mixes that perform best in Phase I will be evaluated for their ability to support plant growth in a greenhouse setting. The study will take place in Temple University’s climate-controlled greenhouse through the 2019 growing season. Tree saplings from several species will be grown individually in containers that are arranged in a randomized complete block design to minimize the effects of environmental variation on the results. After a period of growth under well-watered conditions, plants will be subjected to a modified water regime in which half of the individuals experience near-drought conditions. Metrics of plant performance will include overall assessments of growth (caliper size through time and final leaf area and biomass) as well as more specific indicators of the physiological and chemical constraints and tradeoffs the trees experience aboveground (e.g., stomatal conductance, carbon assimilation, chlorophyll fluorescence, specific leaf area, leaf C:N ratio) and below (specific root length, root diameter distribution, root C:N ratio).  

(3) We will evaluate the performance of the soil-less soil mix deemed most appropriate for use in GSI tree trenches in the context of a full-scale outdoor installation. Replicate trees (with approx. 2” diameter stems) representing species with contrasting water use strategies will be planted in pits filled with the selected mix, while an additional set of trees will be planted in the native soil at the site. Trees will be arranged in a grid in which half of the rows are comprised of each type of soil, thus yielding a randomized complete block design. The study will continue for two years, with a similar set of measurements collected as described for the greenhouse study above.

In parallel with the above, a literature review and comparative life cycle assessment (LCA) will be performed on the materials used in the study to quantify environmental impacts of soil component substitution, including non-renewable energy use, greenhouse gas emissions, and impacts to water and air quality. This analysis will serve as the Capstone project for a Masters of Environmental Studies student at the University of Pennsylvania during the 2018-2019 academic year.



The first three phases of research will yield datasets highly amenable to analysis with common statistical and multivariate techniques (e.g., linear models and principal components analysis). In the lab phase specifically, we expect to identify 2-3 mixes that have properties statistically non-differentiable from sand-based mixes (P values > 0.05) but that are similar to (e.g., within 20% of) established recommendations. With the greenhouse study, we expect to demonstrate that growth and physiological metrics are likewise similar between plants grown in the soil-less soil vs. in the sand-based mix, and that they are in ranges previously documented for healthy individuals of the species. The greenhouse phase will also identify the soil-less mix likely to be most suitable for use in GI systems containing trees, i.e., the mix to be used in the subsequent full-scale evaluation. Once we conclude the two year tree pit evaluation we expect trees to be growing at rates comparable to those already in Philadelphia’s GI systems, thus demonstrating the viability of soil-less soil in a real-world setting. Studies will provide also hands-on research experience and educational opportunities for graduate and undergraduate students from each of our institutions through installation, monitoring, and analysis.

Ultimately, the project aims to create a framework for turning solid waste into urban infrastructure in Philadelphia, but one that can be replicated in other cities as well. To this end, we will prepare a broadly replicable soil specification for implementation in professional practice and public works applications using municipal waste material recovery. This will be made non-proprietary and publicly available in an effort to drive industry-wide transformation and accessibility to renewable soils for green infrastructure.


Richard Roark is a practicing partner at OLIN, an award winning and innovation based design firm in the practice of landscape architecture. Richard’s work focuses on the application of improving the resilience of communities through civic design and the application of living and renewable materials in building practice. Richard has developed project specifications for the Philadelphia Water Department to improve green infrastructure specifications for infiltrating soils.  He is currently responsible for the implementation of manufactured soils with renewable Class A Biosolids on the University of Washington campus. Over its 40 year history, OLIN has pioneered the development of living systems for urban landscapes. OLIN is a research-driven landscape architecture and urban design practice with a dedicated research and development department, OLIN LABS. Rebecca Popowsky is the OLIN LABS External Research Coordinator and will provide project management for the Soil-Less Soil project.

The Ian L. McHarg Center at the University of Pennsylvania School of Design builds on PennDesign’s position as a global leader in urban ecological design. Billy Fleming is the Center’s full-time Research Coordinator.

Sasha Eisenman is an associate professor and Joshua Caplan is a research associate in the Department of Landscape Architecture and Horticulture at Temple University. They have a number of active research projects investigating plant health and soil quality in green infrastructure systems in Philadelphia; their research is funded by the US EPA and the Pennsylvania Department of Transportation. The studies evaluate biochar-based green roof media, PWD’s urban tree trench design, and the performance of PennDOTs’ bioswales along I-95. They have published dozens of peer-reviewed papers, many of which evaluated plant responses to soil conditions and soil properties themselves.

Tim Craul, President of Craul Land Scientists, Inc., is a Certified Professional Soil Scientist with extensive experience in soil design for urban planting and stormwater infrastructure. He is a guest lecturer at Penn State University, the University of Pennsylvania and Harvard University School of Design. Tim’s 2006 book Soil Design Protocols for Landscape Architects and Contractors is widely referenced by landscape practitioners. Tim has also been Ridge and Valley MLRA lead Soil Scientist for USDA-NRCS and has developed all the digital soil spatial data for Pennsylvania where both jobs required significant quality control, quality assurance and oversight to meet goals with quality data.


The Soil-Less Soil Research Initiative Steering Committee includes representatives of the Morris Arboretum of the University of Pennsylvania, the Office of Sustainability of the University of Pennsylvania, the Water Center of the University of Pennsylvania, the Zero Waste and Litter Cabinet of the City of Philadelphia, the Center City District of Philadelphia, and the Center for Urban and Community Forestry at Rutgers University.


Albeck-Ripka L (2018). Your recycling gets recycled, right? Maybe, or maybe not. The New York Times. 29 May 2018.

Brooks AL, Wang S. Jambeck JR (2018). The Chinese import ban and its impact on global plastic waste trade. Science Advances 4:eaat0131

[EPA] U.S. Environmental Protection Agency (2018). Advancing Sustainable Materials Management: 2015 Fact Sheet Assessing Trends in Material Generation, Recycling, Composting, Combustion with Energy Recovery and Landfilling in the United States.

Hossain MU, Poon CS, Lo IMC, Cheng JCP (2016). Comparative environmental evaluation of aggregate production from recycled waste materials and virgin sources by LCA. Resources Conservation and Recycling 109: 67–77.

Owen AG, Hammond LKF, Baker SW (2005). Examination of the physical properties of recycled glass-derived sands for use in golf green rootzones. International Turfgrass Society Research Journal 10: 1131-1137.

Makowski C, Finkl CW, Rusenko K (2013.) Suitability of recycled glass cullet as artificial dune fill along coastal environments. Journal of Coastal Research 29: 772-782.

Peduzzi P (2014). Sand, rarer than one thinks. Environmental Development 11:208-218.

[PWD] Philadelphia Water Department (2016). Green City, Clean Waters Evaluation and Adaptation Plan. Report to the Pennsylvania Department of Environmental Protection.

Torres A, Brandt J, Lear K, Liu J (2017). A looming tragedy of the sand commons. Science 357: 970-971.

Wang FC, Feng XN, Gong H, Zhao HY (2017).  Study of Permeability of Glass-Sand Soil.” Archives of Civil Engineering 63: 175–190.