Going underground to beat dam evaporation

By Mary O'CallaghanOctober 26th, 2016

Holding wet-season water underground for use in the dry season is a technique that holds promise for parts of northern Australia.
aerial view of dammed underground water and agricultural land

The Burdekin Delta scheme, near Townsville, is Australia’s oldest MAR scheme. Image: Keith Bristow/CSIRO

When the rivers of northern Australia are raging torrents and the floodplains are full and lush, it seems like a no-brainer to store some water for use when the dry season inevitably arrives.

But, of course, it’s not that simple. Capturing and storing the water is only one part of it. High rates of evaporation and vast distances are just some of the reasons we need to assess the economic and environmental viability of any water storage solution in the region.

An option that warrants further investigation is to store the water underground—not only can it eliminate evaporation and provide long-term storage, it can also replenish groundwater supply, prevent seawater from intruding, and protect ecosystems that depend on groundwater for their very survival.

There are various techniques for storing water underground for use later. The concept of using human intervention to do so, known as ‘managed aquifer recharge’, or MAR, has two general approaches:

  • infiltration, whereby the water is gravity fed down into the aquifer via seepage
  • injection, whereby the water is gravity fed or pumped into the aquifer via a well.

Schemes that allow the water to filter down into a shallow aquifer are generally much cheaper than those where you need injection wells to replenish, or recharge, the aquifer.

Mapping the opportunities – where supply meets demand

Dr Joanne Vanderzalm, a senior research scientist at CSIRO, is on a mission to find out where MAR might be viable in northern Australia. She and her team aim to map the opportunities for three regions:

  • the Fitzroy catchment in Western Australia
  • the Mitchell catchment in Queensland
  • the Darwin catchments (Adelaide, Finniss, Mary and Wildman) in the Northern Territory.

“You need a suitable aquifer for managed aquifer recharge to be feasible. So first we need to understand the subsurface and the aquifer properties,” says Dr Vanderzalm. “Then we need to investigate how much water can be stored before we can decide which technique is best suited.”

The real opportunities lie in those places where a suitable aquifer, water source and the demand for water intersect, she says.

“It’s a screening process. Once we have identified the potential storage capacity, we will then overlay that with proximity to available water for recharge and demand. We’re not advocating any particular end use of the water, other than economic development. But, to be cost-effective, the water needs to be close to where you use it to minimise infrastructure costs.”

With at least 16 million hectares of soil potentially suitable for irrigated agriculture in northern Australia, but only enough water to irrigate 10% of that area, access to water is a major constraint for development.

Dr Vanderzalm’s MAR project is part of a broader groundwater research program which itself is part of the $15 million Northern Australia Water Resources Assessment (NAWRA), an initiative of the Australian Government’s Agricultural Competiveness White Paper. The CSIRO-led assessment is tasked with investigating opportunities to develop water and agriculture in the three regions by June 2018.

The Assessment, says CSIRO’s Research Leader for Northern Australia Chris Chilcott, “will enable government, industry and communities to make informed decisions about sustainable development in the north”.

A familiar concept in Australia

The oldest MAR scheme in Australia, and one of the largest, is in northern Queensland’s Burdekin Delta, near Townsville, where up to 45 gigalitres (GL) a year is stored underground through infiltration basins established in the 1960s, sustaining groundwater levels for agriculture and keeping seawater at bay.

The Northern Territory has Australia’s first ‘soil aquifer treatment’ scheme which stores treated wastewater underground and augments groundwater resources at Alice Springs. MAR has also been used to store shallow groundwater in a deeper aquifer to supplement drinking water supply on Goulburn Island in the NT.

The Cloudbreak iron ore mine in the Pilbara uses an injection technique to recharge more than 20 GL/year of brackish and saline water for mine water management and to protect groundwater-dependent ecosystems.

managed aquifer recharge diagram

managed aquifer recharge schematic

There are two main types of aquifers – those that are confined by a low permeability layer, which for MAR requires injecting water via a well, and those that are unconfined and allow water to infiltrate through permeable soils, where recharge can be enhanced by basins and galleries.

In the southern states, operational MAR schemes abound, particularly in urban centres where they are used to store stormwater and treated wastewater. Perth’s groundwater replenishment scheme is Australia’s first MAR scheme to recycle wastewater for drinking water. The 14 GL/year scheme is scheduled for operation later this year, with a planned expansion to 28 GL/year in 2017.

Adelaide has a number of schemes that inject stormwater into deep limestone aquifers. A total capacity of about 20 GL/year is available for non-potable use such as irrigating public space. Additional work is underway to support wastewater recycling via aquifers to potentially double this capacity and support an expanded horticulture industry.

Underground dams – a promising technique for the north?

A MAR technique commonly used overseas, especially in Japan, Korea, China and Brazil, is the ‘underground dam’. A barrier is constructed to intercept groundwater flow and store water for use during dry periods. It’s a technique that looks promising for some parts of northern Australia, says Dr Vanderzalm.

“Picture a nice sandy riverbed and several metres beneath it is hard rock,” she explains. “You build a barrier that extends from, say, a metre below the riverbed down to the rock and laterally to the riverbanks. Water can still flow over the top of the barrier, but the remainder is stored underground for use. Think of a weir in a river, but buried, banking up groundwater instead of surface water.”

On the Ashburton River in the Pilbara, mining magnate Andrew Forrest has built an underground dam on his family’s property, Minderoo Station, to store wet-season flows. The feat of engineering, which Mr Forrest refers to as an ‘upside-down weir’, is now supplying water year round for cattle and fodder.

“This MAR scheme is the first of its kind in Australia,” says Dr Vanderzalm. “Others were trialled but have not met with success.”

Because Minderoo Station is operational and is located in a geological setting that is more representative of northern Australia than some of the international examples, Dr Vanderzalm has been granted access to go onsite to understand the site characteristics.

“Our interest is in research”, she says. “We want to know what it is about the sediment properties and the river flow dynamics that make it successful. We want to understand the physical characteristics of the riverbed and surrounding geology, the impact on the groundwater level and the maintenance requirements. This kind of information will help us gain a general understanding of suitable environments for this technique.

“We think MAR techniques, such as underground dams, have potential promise for use in some areas of northern Australia, but it’s crucial to have demonstration schemes that people can learn from. There is naturally some uncertainty for potential investors in schemes that are below ground. Having a demonstration scheme reduces that uncertainty and builds confidence.”

A question of economics for irrigated agriculture

Dr Richard Evans, Principal Hydrogeologist with Jacobs Engineering Group, was involved in the Minderoo Station weir and is now investigating opportunities for MAR in the Pilbara and in Katherine in the NT, under the National Water Infrastructure Development Fund.

“We are confident that underground dams will work in a technical sense in the right hydrogeological environment. It still has to be proven and we need to demonstrate it. But the challenge is to demonstrate that MAR schemes will work in an economic sense.

“For irrigation in northern Australia, if it’s not cheap it’s just not going to fly. The challenge is what does it cost relative to the alternatives? If the alternative is a big dam, that could cost many millions. Can irrigated agriculture pay that back? Usually not. Our thinking is in line with CSIRO’s—it’s all about cashflow and the ability to pay back the loan on the capital cost of the infrastructure.”

The main advantage of some MAR schemes, says Dr Evans, is that they are relatively cheap to run. Then there’s the obvious advantage of evaporation being “practically nil” underground.

Scalability is another advantage of MAR schemes, he says. It’s possible, for example, to progressively add underground dams along the same river.

“These are significant hurdles”, says Dr Evans. “But if the hydrogeology is right and other factors, we believe it can be economic.”

The potential in the 3 regions

While it’s very early days in her research, Dr Vanderzalm is already getting a sense of how viable MAR might be in the three study areas.

“The Darwin region is more suited to injection techniques because of the geology. MAR has the potential to replenish declining groundwater resources in two of the four catchments that are already developed,” she says.

The Mitchell catchment, on the other hand, has potential, she says, on some of the major tributaries of the Mitchell River. “We’re looking for that intersection between water we can detain for recharge and an aquifer that is suitable for storing water.”

For the Fitzroy catchment, it’s too soon to say: “There may be some potential in the lower Fitzroy, but the possible impact on groundwater-dependent ecosystems must be assessed.”

Internationally renowned MAR expert Dr Peter Dillon has experience with MAR in monsoonal climates in India and Africa. He sees potential for it in northern Australia, but doesn’t envisage it supporting broad scale irrigated agriculture.

“I expect MAR would support mosaic-type irrigated farming, as it does in India’s upland catchments and in parts of Africa,” he says. “It won’t transform the landscape into an Ord Valley—we’re not talking about that scale of irrigation. It is more likely to be the livestock industry that will benefit, and the mining industry, where water can be very valuable for enabling extraction and processing.”

Listen to Dr Peter Stone’s presentation at the 2016 World Water Congress held in Brisbane on Unlocking the Potential of Northern Australia.


  1. This could be a brilliant scheme if properly researched and applied, but it’s got to be remembered that these streams and rivers support a large food chain from the plankton, crustacean, fish, mollusc, reptile and bird which are also used by our Aboriginal first people who definitely should be involved in the planning and implementation.

    1. Thanks for the comment. This work is part of a broader program, the North Australian Water Resource Assessment. In that CSIRO is looking at quantifying the resources (water, land) and the risks of potential developments including the ecological and social risks. That will include working with indigenous communities and understanding the ground and surface water dependent ecosystems (including the food chains).

  2. My family and I plan to implement a MAR on a small tributary of the Nambucca River, mid-north coast of NSW. The property consists of a catchment of c.1,000H, gravel alluvial valley floor about 6 metres deep, dry 99% of the time but high short term creek flow. Local council and DPI Water NSW expect all sorts of studies and engineering plans before approval of such a project. Does anyone have relevant experience, information to satisfy the bureaucrats?

  3. We understand there have been completed MAR studies in three areas – Fitzroy Catchment WA, Mitchell Catchmnt, Qld and Darwin. Has there ever been similar studies done in the Mount Isa region generally ; what might provoke a study in that area, or has that area been discounted as not viable ? Our background has been, among other things, in farming, building and property development. We hear so much about the potential for development of Northern Australia so much so that we would like to make you aware that a syndicate of us stand ready should properly researched opportunities present themselves in this location.

    1. Hi Geoff, thank you for your question and interest in MAR.
      Our scientists have provided the following response. There has been a series of water resource assessments undertaken over the past decade focusing on development in northern Australia, many of which are precursors to the one discussed in the ECOS article, the Northern Australia Water Resource Assessment (Fitzroy, Mitchell and Darwin assessment areas).
      Two reports that may be of interest are:
      1. Flinders-Gilbert Agricultural Water Resource Assessment – not encompassing Mount Isa, but in the adjacent Flinders catchment. The Flinders catchment report can be accessed here https://www.csiro.au/en/Research/LWF/Areas/Water/Assessing-water-resources/Flinders-Gilbert/Flinders-report

      2. An earlier study (2009), the Northern Australia Sustainable Yield project which included the Leichhardt river basin (and Mount Isa). The report for the Flinders-Leichhardt region can be accessed here https://publications.csiro.au/publications/#publication/PIprocite:5810c620-913e-440b-8ffd-cb4567c052cc


Commenting on this post has been disabled.