Current Alerts 11 currently active, find out more »

Yarra Watch 2016–17 Summer Highlights Report

Yarra River at Healesville

Yarra River at Healesville. Source: EPA Victoria

The Yarra River is widely used for recreational activities such as boating and swimming throughout summer.*

Environment Protection Authority Victoria (EPA), in partnership with Melbourne Water, runs the Yarra Watch program, which forecasts and monitors recreational water quality at four sites along the Yarra River – Kew, Warrandyte, Healesville and Launching Place. Water quality is determined by the levels of bacteria in the water, which can pose an increased risk of disease for people coming into contact with the water.

The summer season ran from 1 December 2016 until 13 March in 2017.

* Please note that it is illegal to swim in the Yarra downstream of Gipps Street, Abbotsford. Kew at Chandler Highway Bridge is the only Yarra Watch site in the Lower Yarra where it is legal to swim.

How did swimming spots along the Yarra perform?

Table 1: Results of water quality monitoring at Yarra River swimming spots.



Launching Place




Most Yarra River swimming spots met the water quality objective for swimming this summer (Table 1).

Warrandyte comfortably met our water quality objective for swimming and had the best overall water quality of all Yarra swimming spots this season.

The only Yarra site that did not meet the objective was Launching Place. This does not mean Launching Place was unsafe for swimming the whole season, but it did experience a higher number of days with poorer water quality, which can increase the risk of swimmers being exposed to disease-causing bacteria. Bacterial levels at Launching Place were variable over the summer. Most days of poorer water quality occurred after recent rain and stormwater runoff. Poorer water quality during dry weather was less frequent. Poor water quality may be due to a range of pollution sources such as leaking septic tanks and animal faeces (e.g. agricultural livestock and birds).

All sites meet objectives for secondary contact recreation (i.e., boating, rowing, canoeing).

Yarra water quality has been variable in recent summers 

Table 2: Water quality trends for swimming at Yarra River sites.

Swimming objectives met?

2013–14 2014–15 2015–16 2016–17 Water quality trend
Launching Place  ✗ No change
Healesville  ✓ Variable
Warrandyte  ✓ No change
Kew  ✓ Variable

Water quality in the Yarra River has varied over the last four years. Stormwater pollution after rain has been the most common factor determining whether a Yarra River site meets the objective for swimming (Table 2). 

Warrandyte has been consistently the best location for swimming in the Yarra River in terms of water quality for recreation.

Healesville varies between summers, while Launching Place has continually not met overall season objectives. This means bacterial levels in water at these sites, have from time to time, posed an increased risk to swimmers. Bacterial pollution was most often due to stormwater runoff entering the river. We note that our monitoring results may pick up more bacteria if the scheduled sampling happened directly following a rain event. However, bacteria may also be present during dry weather as a result of faecal contamination from a range of animal and human sources in the Yarra’s rural and upper catchment (see ‘Water quality and health for swimmers’ section for more information on faecal sources). EPA Victoria will be working with Melbourne Water to better understand and track pollution sources in the mid to upper Yarra River (e.g. Launching Place and Healesville) where dry-weather pollution is known to occur.

EPA only provided a forecast of either ‘Fair’ or ‘Poor’ at Healesville (that is, there were no ‘Good’ forecasts) due to its higher background bacterial levels in the water. EPA will review recent water quality at Healesville after the 2016-17 season to assess whether it can be issued with ‘Good’ forecasts for next summer.

What information do we gather and report about the Yarra River?

The information we report on this website and through Twitter includes the following:

  • Water quality forecasts: 
    • These are reported twice daily during the summer season.
    • Forecasts of Good, Fair or Poor are given for each location to inform decisions about recreational use.
    • They are based on weather forecasts, observations and warnings, water quality history, and weekly water sampling results.
  • Weekly water quality monitoring:
    • Samples are analysed for the presence of E. coli bacteria, which indicates that water may not be safe for recreational use due to an increased risk of exposure to disease-causing bacteria and other pathogens.
    • If it’s not considered safe to swim, the public are notified on the Yarra and Bay website and Twitter.
    • EPA works with local councils and water authorities to investigate the source of water pollution.
    • These samples are used to assess whether the site has met the end-of-season water quality objectives for swimming.
  • Alerts about pollution incidents, fish deaths and algal blooms:
    • Water quality alerts are given all year round.
    • Alerts are based on reports from the community, businesses and government, and EPA or other responding agency.

Pollution can be reported at any time to EPA on 1300 EPA VIC (1300 372 842).

What influenced water quality and forecasts this summer?

Yarra River at Kew

Yarra River at Kew. Source: EPA Victoria.

During rain events, pollutants wash down stormwater channels and into the Yarra River, which increases the risk of swimming at nearby swimming spots.

When rain occurred this season, Yarra Watch informed the public by issuing a water quality forecast of Poor, advising people to avoid swimming near stormwater outlets after rain.

Extended periods of Poor forecasts were mainly issued during periods of heavy rain; these occurred around late December 2016, late January 2017 and early February 2017.

Swim advisories and other water quality alerts issued over summer

Swim advisories are issued if there are high bacterial levels found during dry weather, when people are more likely to use the river for swimming. This summer, there were no swim advisories issued for the Yarra based on the weekly sampling at the four swimming spots.

There was one high bacterial result at Healesville on 22 February, however water quality returned to low levels in the resample the next day. Two high results are needed (both the weekly sample and the resample the next day) before a swim advisory is issued. A desktop investigation could not find an obvious pollution source, however likely sources are thought to be animals in the agricultural catchment and potential leakage or discharges from unsewered properties.

There were four water quality alerts issued for the Yarra River based on reports to EPA from the public and businesses. Three of these were in the lower Yarra where it is illegal to swim:

  • A pollution alert was issued on 16 December 2016 for the Yarra after a white substance was reported coming from a stormwater drain adjacent to Evan Walker Bridge in the Melbourne CBD. The most likely source was paint or a similar substance, but the source could not be located due to difficulties in accessing the stormwater drain. The substance was left to disperse in the river.
  • On 29 December 2016 a pollution alert was issued for the Yarra following two wet-weather sewer spills in Kew. These were due to discharges from Melbourne Water and Yarra Valley Water emergency relief structures during a large storm. Discharges from emergency relief structures can occur during large rainfall events, when heavy rain enters sewerage systems and increase flows in sewer pipes beyond the capacity of the sewer. If this occurs, the excess flow (a mix of stormwater and sewage) is discharged directly into the stormwater system through emergency relief structures.
  • On 3 January 2017, a pollution alert was issued for an oil film in the water at jetties near Docklands. Investigation determined that it was most likely washed from roads into stormwater drains during heavy rain in the previous week, and had not dispersed in the water. The oil film was left to evaporate and degrade during the following days.

Oil film at Docklands on 3 January 2017

Oil film at Docklands on 3 January 2017. Source: EPA Victoria

  • On 18 January 2017, a pollution alert was issued after a sediment discharge from a stormwater drain was reported in the Yarra River near Flinders Street Station. After investigation, the source of the discharge could not be located.

Alerts were removed after the pollution could not be detected or was no longer visible. Pollution of rivers and creeks can be difficult to investigate as EPA relies on timely public reports of pollution and for the pollution to still be occurring when EPA officers arrive at the scene. Pollution should be reported to the EPA hotline on 1300 EPA VIC (1300 372 842).

Water quality and health for swimmers

Increased risk of illness primarily occurs when water is contaminated with human or animal faeces. Yarra Watch monitors E. coli in the River, which is a bacteria that is an indicator of faecal pollution. Faeces may carry any number of disease-causing organisms (pathogens), with gastroenteris being the most common illness experienced by swimmers.

The level of human faecal sources in stormwater or other pollution can vary according to the duration and intensity of rain, and depends on land use or activities in the Yarra’s large catchment. Human faecal contamination was found during wet and dry weather in a 2005 EPA study at 38 different tributaries and drains entering the Yarra. Overall around 27% of total samples returned high amounts of human faecal contamination. Human faecal contamination in the River still occurs today with dry weather sewer spills and sewage overflows during storms (e.g. late December 2016) entering the Yarra. The extent of human faecal contamination in the Yarra is comparable to studies in other parts of the world. For example, human faecal contamination was present in between 21-44% of samples from a study in Ochlockonee and St Lucie River in Florida, and 25-69% for Dungeness River catchment in the state of Washington (United States).

High bacterial levels in stormwater run-off to the river can be caused by sewerage and septic tank leakages, cross-connections between sewerage and drainage systems, or litter and animal faecal waste (for example, bird and dog droppings) entering drains. Additionally, when there have been heavy rain events, such as in late December 2016, sewage overflows from emergency relief strictures can also increase bacterial levels in the river.

In dry weather, potential sources of faecal contamination in the river include:

  • leaking sewer infrastructure (particularly where it is above the stormwater drainage system)
  • cross-connections between sewerage or septic tanks with stormwater drainage systems
  • poorly-operating septic tanks and onsite sewage treatment plants
  • animal faeces from:
    • birds
    • agricultural livestock accessing the river or its tributaries, or runoff of livestock faeces during periods of irrigation
  • Sediment that contains bacteria resuspended during environmental-flow releases.

Increased bacterial levels can be expected as a result of these inputs, however, they are usually short-lived.

How accurate were Yarra Watch water quality forecasts?

The accuracy of Yarra Watch forecasts was measured by comparing morning forecasts (10am) to weekly bacterial water quality sampling results in the River. Note that forecasts are produced daily, but water sampling only occurs once a week on a Wednesday. So, the accuracy test only compares the Wednesday sample to the corresponding forecast on the same day. Because of this, the accuracy assessment below cannot fully represent the accuracy of all forecasts over the summer period, but gives a general indication of forecast accuracy.

Most Yarra Watch forecasts provided appropriate advice to protect public health (Tables 3 and 4). Most days of poor water quality were issued with forecasts advising against swimming or to check for signs of pollution before entering the water.

Table 3: Results of overall forecast accuracy for 2016-17


Overall accuracy



 What does it mean?

What will be our next action?

Appropriate advice


This is the most common measure of forecast accuracy used by other forecast services worldwide. 93 per cent of forecasts provided appropriate advice about whether it’s safe to swim. In other words 52 out of the 56 forecasts issued on Tuesdays this summer provided appropriate advice.

For example, when actual water quality was Good EPA issued either Good or Fair forecasts.

When water quality was Poor EPA issued either Poor or Fair forecasts.

Fair forecast ratings advise the river user that water quality may not be suitable, to check for signs of pollution before swimming. This rating is therefore appropriate for either Good or Poor water quality, as caution is advised.

Continue to monitor accuracy to ensure we provide appropriate advice.

Forecast accuracy


50 per cent of forecasts correctly predicted the actual water quality. In other words 28 out of the 56 forecasts issued on Wednesdays this summer were correct.

This means that 50 per cent of forecasts accurately predicted the actual water quality i.e. Good forecasts for Good water quality, Fair for Fair water quality days, and Poor forecasts for Poor water quality. 

Improve our forecast model to increase our accuracy.

Table 4: Results of missed and false alarms in forecasting 2016-17


Missed and false alarms



 What does it mean?

What will be our next action?

Missed alarms on poor water quality days 


There were no Good forecasts issued on days with Poor water quality.

Continue to monitor accuracy to ensure we don’t issue too many of these types of alarms.

False alarms on Good water quality days


Out of 56 forecasts we issued on Wednesdays over summer, only four Poor forecasts were issued on days with Good water quality. This means that on some occasions we advised people not to swim when actual water quality was good.  These types of alarms may have economic and social impacts if issued often.

Continue to monitor accuracy to ensure we don’t issue too many of these types of alarms.

Yarra Watch Fast Facts

  • Of the total water forecasts issued over the summer, 52 per cent were Good, 30 per cent Fair and 18 per cent Poor.
  • During the summer, there were four water quality alerts issued for the Yarra River.
  • The most water quality alerts were issued for the Yarra River, especially in the lower reaches through the CBD.  Some tributaries to the Yarra River, such as Scotchmans, Edgars, Merri and Stony Creek, were also issued with more alerts than other waterways.
  • Forecasts were provided to over 6,500 followers on Twitter 

Future directions

EPA Victoria will continue to work with Melbourne Water to better understand and track pollution sources that cause water quality to be unsafe for swimming. The focus of this work will be on the mid to upper Yarra River (e.g. Launching Place and Healesville) where dry-weather pollution is known to occur.

For more information on how we generate forecasts and the accuracy of our forecasting go to Generating forecasts for Beach Report and Yarra Watch.

Program Partners