In the mid-1990s, Colin Austin asked a simple question: are there better ways of applying water? Horticulture had become more efficient, but it used only a small fraction of irrigation water. The real opportunity lay in flood irrigation, the largest water user. This chapter follows a series of speculative experiments—subsurface pipes, a “froth machine”, improved subsurface drip, and finally a rethink of flood itself—leading to Microflood, the tilt valve, and ultimately the wicking bed.
Chapter 3 — Subsurface and Improved Flood Irrigation
Are There Better Ways of Applying Water?
This was the question I faced in the mid nineties.
Over the years there have been dramatic improvements in horticultural irrigation: drip, subsurface drip, micro-sprinklers, and similar systems. However the reality is that horticulture only uses about 5% of irrigation water and is already reasonably efficient. Even a 40% gain in efficiency—an ambitious target—would only produce a saving of about 2%.
The largest user of water by far is flood irrigation, largely used for feeding animals. Flood, even at its best, is highly inefficient. Here was the opportunity to generate significant water savings. That was the logic I was working with at the time.
Cost and Scale: The Core Challenge
The major challenge was cost and scale. Irrigated pasture is far less profitable than horticulture (on an area basis), so the question was obvious: could we develop a low cost alternative to flood irrigation?
At that time I had become somewhat mesmerised by the benefits of subsurface irrigation. It avoids one of the biggest problems we face in Australia—excessive evaporation— and it leaves a dry surface for animals to graze.
The challenge I accepted was to see if I could develop an inexpensive subsurface irrigation system suitable for pastures.
Failure as the Cost of Innovation
I knew this was going to be a monster challenge. I have to admit upfront that I failed in my primary objective. But failures lay the groundwork for ultimate success.
Readers who want to move straight to the crunch line may prefer to skip ahead to the next chapter. But if you believe, as I do, in speculative research—and accept failure as the cost of innovation—then this chapter is an example of a project which, while appearing to be a total disaster, led to technologies such as the wicking bed and soil carbon capture which may be of profound importance to all of us.
The Inside–Outside Pipe
Subsurface irrigation seemed to offer the most potential. But with the (then) price of water and the cost of installing conventional subsurface pipe, it did not look feasible. A major new innovation was required.
The first idea was the inside–outside pipe—certainly different, and potentially viable at large scale at low cost.
How the Inside–Outside Pipe Was Meant to Work
The idea was simple:
- Plough a length of lay-flat pipe into the soil.
- Inflate it to form a tunnel through the ground.
- Release the pressure and feed water into the tunnel to wet the ground from underneath.
The mechanics were also simple. A conventional pump sucked water from a small reservoir at the bottom of the slope to pressurise the pipe and form the tunnel. The top end of the pipe exited the ground into a miniature dam filled by the pump. No water could escape from the dam because the pipe, while pressurised, pushed tightly against the soil.
The pump was then turned off. Water in the pipe would run backwards through the pump, so the pipe progressively collapsed. As it collapsed, it allowed water in the top dam to flow down the tunnel, wetting the soil from underneath.
It was a simple and ingenious system, and it worked well—most times.
The Problem: Uneven Ground, Uneven Water
The problem emerged when the ground was uneven or slope varied along the bay. Low points received more water than they should, creating poor distribution.
No doubt this could have been addressed with laser grading and better control of the return flow through the pump. But another idea had already emerged.
The Froth Machine
One of the major problems with irrigation is that the soil, at least locally, becomes wetter than the ideal for plant production.
With flood irrigation it is widely recognised that growth often diminishes immediately after flooding, then rises to a maximum, then drops again as the soil dries out.
Proponents of modern systems like drip and micro-sprinklers argue that the soil is wetted more uniformly. In reality, water does not magically distribute itself evenly through soil. Moisture gradients are required to drive water movement. After the initial movement, when soil is saturated, the rate of movement drops rapidly, leaving a wide variation in moisture content.
Maximum Growth Needs a Water–Air Balance
Plants grow best when there is a balance between water and air in the soil. Conventional irrigation rarely achieves that balance consistently.
So the idea came: why not inject a mixture of air and water directly into the soil?
This would also reduce some engineering problems. Drip and micro-sprinkler systems use very small orifices to regulate flow, but those orifices block. Even with good filtration, growers regularly inspect for blockages. This is one reason subsurface irrigation has not been adopted as widely as its theoretical benefits suggest.
Froth as a “Rugged” Solution
A froth—air mixed with water—should improve growth by improving the air–water ratio, while also enabling a simple and rugged delivery system.
And so a froth machine was developed to inject this mixture into the soil, with the hope that the air would help distribute the water and produce a uniformly moist (but not wet) root zone. The world of innovation is full of ideas with overwhelming theoretical benefits and a wide gulf between theory and practice. The froth machine was one of those innovations.
My strongest memory from the first trial in a farmer’s paddock is the farm dogs going wild as thousands of emitters whistled as the air rushed out. I would sooner forget the divergence between theory and practice.
It was time for a major rethink.
Rethinking Subsurface Irrigation
In innovation and technical development there is a universal trend called convergence. In early stages there is wide variation, but as technology develops the variation disappears and solutions become similar.
You can see this in the evolution of cars. Over time, vehicles become almost indistinguishable until a major new development— like front wheel drive or fuel injection—creates a new wave of variation before convergence resumes. Perhaps it was time to stop chasing “way out” ideas and look at improving conventional subsurface irrigation. Commercial subsurface systems were essentially above-ground dripper systems buried in the soil.
So I asked: if we were designing subsurface irrigation from scratch, aiming for a low-cost system rugged enough to replace flood, what would it look like?
Water Movement in Soil: The Hidden Constraint
One big problem with drip irrigation—often not well understood—is the movement of water through soil. Several mechanisms exist, but in conventional irrigation the weak surface tension forces are usually dominated by gravity. That means little sideways movement and lots of downward movement.
Applying water at a faster rate makes pressure flow more dominant—water is literally forced sideways. Tests confirmed this. Water could be moved sideways by a couple of metres, compared to roughly 300 mm relying on surface tension alone. This meant the number of irrigation lines could be reduced to about a third, helping the cost target needed to compete with flood irrigation.
Manufacturing and Pressure Drop
Larger pipes were required, but that brought a benefit: manufacturing could shift to a blown film line, which is intrinsically cheap.
The next problem was pressure variation along the line at higher flow rates. This was solved by developing a computer program that drove a punch machine, allowing emitter spacing to vary along the line to compensate for pressure drop.
Trials and the Rugged Reality of Farming
Trials were run on various farms with different soil types. The results were a mishmash. Some systems worked well, others suffered a string of technical difficulties.
Reviewing the results led to a blunt conclusion: the system could be made to work with careful, sophisticated installation and management, but it did not fit the rugged farming scene.
Farming is a tough business. Products must be simple and reliable. Technologies that work under controlled conditions are often not viable in harsh, variable environments.
By this point, significant money—measured in the millions of dollars—had been spent. It was time for another major rethink.
Reworking Flood Irrigation
My ambition had been to develop a system to replace inefficient flood irrigation, and so far we had failed miserably. There is an old saying: if you cannot beat them, join them.
So the question became: if we cannot replace flood irrigation, what can we do to make it more efficient?
Simulating Flood Irrigation
With my background in computer simulation, my first approach was to build a simulation of flood irrigation. Once the code was working, I could “play” with variables quickly and see what happened. In one afternoon I learned more than could be learned in a lifetime of real-world trials.
Some variables were “soft”—for example, slope and surface roughness had little effect on final outcomes. Other variables dominated, especially the speed of application.
The target was to achieve a uniform depth of infiltration along the paddock. A slow application rate saturated the top to a great depth. A very fast rate produced shallow penetration.
A Two-Stage Process for High Efficiency
The simulation revealed that a two-stage process could achieve very high efficiencies.
- Stage one: a very rapid flow rate, cut off before the flow front reaches the end of the paddock, but timed so there is sufficient water “in transit” to reach the end of the bay.
- Stage two: a much slower application rate so the entire paddock carries a film of water on the surface, which then soaks in to the required depth.
In simulation, practical constraints do not matter—you simply type in numbers. But the result was powerful: efficiencies well over 90% were possible simply by managing flow rates.
Turning Simulation Into a Practical System
The challenge was turning this understanding into something farmers could actually use. It was never expected that simulation software would be widely adopted by farmers. That would require a new skill set, and the low-tech farmers—who often waste the most water—would have little interest.
Traditional flood systems use large channels to deliver high flow rates. At the farm level there is pressure to make paddocks as large as possible, which reduces efficiency and worsens distribution.
Ideally water would be distributed by pipes, but the required flow rates demand very large, expensive pipes. Many attempts have been made to replace channels with pipes, but the economics rarely work. A new approach was needed—one that treated the distribution system and paddock as an integrated system.
The “Loose–Loose” Trap
Simple calculations show the trap. A small pipe is adequate if it runs continuously, but large pipes are needed to deliver the large slugs of water required for efficient flood irrigation.
It becomes a classic loose–loose situation:
- Big pipes or channels can improve efficiency, but they waste capital and encourage even bigger paddocks.
- Small paddocks improve efficiency, but they waste land and increase evaporation from the extra channel area.
Microflood
Once these realities were accepted, the solution became obvious: irrigate small areas in sequence. That allows efficient distribution through small pipes and increases irrigation efficiency across the farm.
This could be done at high capital cost using conventional valves and controllers. What was needed was a simple valve that would divert water from one area to the next once a certain volume had passed.
This led to the development of Microflood and the tilt valve. The tilt valve is a gravity-operated valve where water is bled into a hinged pipe which tilts back to allow full flow. When sufficient water fills the pipe, it overbalances and shuts off flow into that section and automatically diverts flow to the next section.
Simple and effective—but virtually impossible to sell.
When One Failure Leads to a Better Idea
That was not the end of the story. By chance I was to find a system far more efficient— virtually eliminating water lost past the root zone and reducing evaporation from the surface,
so that nearly all water applied was used for plant growth.
It also improved plant growth and productivity, and reduced the need for frequent irrigation. Even more importantly, it had the ability to capture significant carbon in soil, which could have major implications in the battle against global warming.
I sometimes wonder whether it is a good thing when ideas fail to catch on, because they can lead to the development of an even better idea.
That system was the wicking bed, which I will talk about in the next chapter.
Colin Austin — © Creative Commons. Reproduction allowed with source acknowledgment; commercial use requires a license.
