Before we commence testing of the waste materials, the pilot team undertook the following activities:

• desktop research
• literature reviews
• professional consultations
• planning project stages, deliverables and investment of funding.

This process was completed in September 2022. Some research and planning will continue throughout the pilot as required.

What we learned

As a result of this process, we discovered projects experimenting with a bioretention system that is partially made from recycled materials (a bioretention system is one that filters pollutants and nutrients, like a raingarden). We didn't find any projects using only recycled materials as a filtration system from only local sources. In this stage we also identified key success criteria to apply to Remix Raingardens to see how these raingardens perform against a typical raingarden.

After planning and research, the pilot team conducted a series of tests in a nursery environment comparing the effectiveness of different waste materials used for water filtration, infiltration and plant growth.

A total of 20 small-scale tests will use recovered waste materials like crushed glass, street sweepings, aggregated concrete, crushed brick, timber, food and other organic waste in layered configurations. Water will be poured over these small tests and the flow, amount and quality of the water that comes through the garden will be measured.

Baseline tests using conventional raingarden materials like gravel, sand and soil will enable the project team to make comparisons and understand the viability of different waste material compositions.

Key findings will be shared here and will inform the following stages.

What we did

In a Parkville nursery, we’ve been conducting tests on raingardens made from different materials, conventional materials and four types of recycled materials.

Raingardens have a series of layers to effectively manage and filter water. Based on research, we focused our initial testing on the filter layer, the layer that stops contaminants, rubbish and sediment from entering the water systems.

In the control gardens we used standard materials (conventional, white-washed sand and soil) for the filter layer (commonly called the ‘filter media’). In the other gardens, we used a variety of recycled materials to mimic typical sand and soil ingredients. A total of 20 gardens with different mixtures were created to ensure quality testing.

Recycled materials can be categorised into sand and soil materials based on their characteristics.

Recycled ‘sand’ materials:

• crushed brick, aggregate and recycled sand - we will call this ‘Construction sand’ in testing
• 100 per cent crushed glass – we will call this ‘Glass sand’ in testing.

Recycled 'soil’ materials:

• food waste and street sweepings with a higher concentration of food waste – we will call this 'High food soil’ in testing
• food waste and street sweeping budget ‘soils’ - we will call this 'Low food soil’ in testing.

The recycled ‘soil’ and ‘sand’ materials were combined to create four different compositions based on research and advice from industry experts.

Simulated stormwater made to mimic typical Melbourne stormwater were used on the gardens, with the filtered results sent to a lab for analysis to determine the effectiveness of the filter materials.

What we tested for

To understand the effectiveness of our recycled raingardens, we compared the test gardens with our control gardens.

We also measured the filtered water to detect the different levels of contaminants and amount of sediment remaining. We test specifically for any remaining nitrogen, phosphorous and water sediment (Total Suspended Solids – TSS) in the filtered water. Nitrogen is used by plants for lots of leaf growth and color while Phosphorous helps plants form new roots, make seeds, fruit and flowers and fight disease. Cloudy water due to sediment means there are particles in the water suggesting impurities or contaminants.

An effective raingarden reduces amounts of contaminants and sediment from stormwater.

What we learned

The results from our first round of labs tests are varied yet promising. While some raingardens incorporating waste materials perform like the control raingardens, others are not removing enough contaminants and sediments.

The graph below represents lab tests of the five different types of raingardens. By comparing the control garden results against the others, an assessment can be made on which filter materials are performing similarly. We found while the gardens that use crushed glass or 'Glass sand' seem to be removing some TSS or total suspended solids, rather than removing phosphorous and nitrogen, they’re actually adding them into the water.

Alternatively, the gardens with the ‘Construction sand' and soil made from food waste and street sweepings (High Food and Low Food) are performing similarly to the control garden – even a little better!

We have also been trialing a native plant called Carex Appressa in the gardens. This is a species commonly used in raingardens for its ability to filter nitrogen. It is growing well in all the raingardens. We will plant Carex and a variety of other plants in the gardens in Fishermans Bend to continue testing plants.

What’s next

We will continue to water the nursery gardens and do another round of lab tests. This stage is not yet complete and we’ll update this page with findings soon.

Once thorough testing has determined which waste materials are most useful in a raingarden design, a pilot design will be developed and trialed in Fishermans Bend over an extended period.

During this stage, the site for the pilot will be determined in collaboration with planners and experts, to enable both valuable testing and community collaboration.

A prototype will be made in the Parkville nursery during this stage, to ensure the design is resilient to withstand testing at Fishermans Bend.

This stage will also include planning how we will use technology to capture data about the pilot. This might include pedestrian sensors to measure community engagement, micro-climate sensors to measure temperature and humidity around the gardens, soil moisture and acidity, water flow rates and more.

Finally, development of the design will include signage and communication tools to share with the community and invite feedback.

What we’ve done

Together with students from the University of Melbourne Makerspace we have developed the design of the raingarden. The design will be tested on Turner Street in Fishermans Bend.

Key considerations made include the number of different raingardens to be tested, which sensors to be used to capture data, how stormwater is used for testing, how to best share information with the community and how we capture water for lab testing.

Below is a render of the concept design.

What’s next

We continue to develop the design and plan to install the raingardens on Turner Street, Fishermans Bend by March.

The Fishermans Bend installation will consist of a series of raingardens. Over a period of several months, sensors will capture a variety of data so we can understand the viability of using waste materials in raingardens.

Signage will describe the pilot and enable community input.

The length of this stage will depend on the performance of the raingardens. We will measure the gardens through collecting sensor data and undertaking observational analysis to understand the viability of raingardens that incorporate recycled materials.

During this stage, there will be opportunities for the community to learn about the pilot and share thoughts and feedback. A dashboard will be published and made accessible to the community to share data and insights.

The final pilot stage will involve analysing all collected data to understand the viability of this concept.

This stage may also include developing a design guide for how to use recycled materials in raingardens that successfully filter and manage rainwater while enabling resilient plant life.