Design note 3 - Permeable pavements, stormwater harvesting and reuse

For permeable pavement information, please see Melbourne Water's MUSIC Guidelines, Chapter 13.

Stormwater Harvesting and Reuse

Stormwater harvesting schemes should be modelled using the 20 year climate template. Results for this period should be used to predict the likely long-term reliability and yield of the scheme, and for economic assessment. The 20 year templates are based on the long-term average of multiple stations using all available data and are considered the best available estimate for likely future behaviour.

Climate change projections indicate that temperatures are likely to rise and mean annual rainfall is likely to reduce in future. This may affect stormwater harvesting schemes by increasing irrigation demands and reducing inflow volumes. While the effects are likely to be minimal in the short term (20-30 years), they will increase over time. There is also potential for a ‘step’ change in climate resulting in more severe impacts being felt sooner.

Sensitivity testing should ideally be undertaken using a selected historical period with lower rainfall or a climate change adjusted time series to indicate potential future performance during drought or long-term climate conditions. A period of good quality data (minimal missing or accumulated) should be selected with a mean annual rainfall based on consideration of likely average conditions for the location and future climate projections for the period of time of interest.

It is recognised that rainfall during the recent period including the millennium drought (1996-2009) has been significantly lower than the long-term average. This is illustrated in Figure 11, with a snapshot of selected data for the City’s Anakie Road depot (2004-2017) and North Geelong (1980-1989) rainfall stations.



Figure 11 : Monthly rainfall totals.

  • Geelong North BOM station 1980-1989 (box plots) and
  • Anakie Road depot 2004-2017 (blue dots depicting median values).

Basic design parameters for stormwater reuse systems

Stormwater reuse systems should aim to provide an alternative to potable water as the primary source with a minimum reliability of 75 percent, minimum total demand of 22 ML/year and yield of 16.5 ML/year. Smaller systems may be considered on a case by case basis subject to agreement and approval in writing by the City.

Any proposed re-use system must be discussed with the City’s Irrigation Manager (Parks and Gardens Department) to determine project specific requirements as part of the detailed design and approval process.

For reference, typical reuse demand figures are provided in Table 5, below. However, specific re-use demand must be discussed with the City’s Irrigation Manager (Parks and Gardens Department) to determine project specific requirements as part of the detailed design and approval process. Values for monthly distribution of irrigation demand are also provided for reference in Table 6, below.

Table 5 : Typical reuse demand.

Type Typical Reuse Demand (ML/Ha/yr)
Warm Season Turf 3.2
Cool Season Turf 4.5


Table 6 : Monthly distribution of reuse demand.

Month Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec
percent of annual demand 29 19 13 1 0 0 0 0 0 1 13 24


Total and directly connected impervious

It is recognised that flow volumes from existing areas may be over-estimated if the total impervious (TI) fraction, rather than directly connected impervious (DCI) fraction is used. Calibration studies to observed flows suggest that the ratio of DCI:TI may average around 66 percent but vary from as low as 30 percent up to 90 percent for established urban areas. Where a stormwater reuse scheme is treating and reusing water from a well-established existing area, impervious fractions for the catchment should be based on estimated DCI using a DCI:TI ratio of 66 percent.

Note: TI must be used for the design of treatment assets for all new developments as they typically have high levels of imperviousness and drainage connection and a higher proportion of connectivity should conservatively be assumed.

Where diversions and pumps are used to divert flows for a stormwater reuse scheme, the modelling should reflect anticipated real conditions. This includes, bypass of low flows to preserve baseflows where needed and bypass of high flows above the design. This is important, even at concept level, to avoid over-stating potential reuse volumes captured.

Where diversions occur, these should not adversely impact upon baseflows for receiving waterways. Baseflows should be separated and bypassed with a diversion for catchments greater than 100 hectares unless otherwise agreed in writing by the City.
The low flow bypass rate should be calculated such that the volume of water bypassed is not less than the total baseflow volume expected from the upstream catchment. This may be estimated using the node water balance for the upstream source nodes in MUSIC or more sophisticated analysis considering observed flows for the waterway for larger schemes.

The design flow rate for treatment and diversion should be estimated using rational method or a hydrologic model such as RORB. This should be the 4 EY (1 in 3 month ARI).

Weir overflows from an upstream sediment basin or wetland inlet pond should be diverted around the stormwater storage to ensure only fully treated flows are directed to reuse.


Next sections: Imported data nodes, generic treatment nodes, use of secondary links and exporting results

Previous section: Bioretention

Back to table of contents





Page last updated: Wednesday, 18 December 2019

Print