Harvesting Smaller Rains
Droughts are not new. We have known about the flood and drought cycle for years, and the storage capacity of our dams per person far exceeds most other countries. Global warming, population growth, and the expansion of irrigation are not abrupt changes. They are trends that have been building for years and will continue.
In community action on water, we saw that we are largely harvesting only the larger rains, yet the total volume falling in smaller rains is more than adequate to meet many needs. The problem is to harvest these smaller rains and avoid losing them again to evaporation.
When small rains fall on soil, they are absorbed and quickly evaporate. There are three practical solutions: capture water from hard surfaces, work the soil to funnel water into percolation holes, and use anticipatory scheduling so irrigation happens when losses are lowest.
Tank Water: Clean and Simple, but Limited
Catching roof water in household tanks is the simplest approach. Over the years, water tanks have not always been welcomed by authorities; in many areas they were even banned. Even today, there are spurious arguments about cost and purity.
The fact is that water tanks are highly cost effective, and water quality can be very high if a few simple precautions are taken. Water authorities that deride tanks rarely emphasise that their own supply often begins as runoff and is initially contaminated by animal droppings. It is basically processed cow shit. Maybe well processed cow shit, but still cow shit.
Roof water, by comparison, is generally much cleaner at the source. Contamination mainly comes from leaves in gutters and some bird droppings. A simple first-flush system and basic maintenance resolves most of this. If people want a second layer of protection, household filters can provide a high level of filtration.
Tank water can be made safe and is often of higher quality than processed dam water. The only problem is volume. Tanks can cover normal household use—drinking, cooking, washing, toilets—but there is nowhere near enough water for food production or for the green environment most people aspire to.
Roads and Hard Surfaces: Lots of Water, Low Quality
The area of hard surfaces in urban regions is huge, and enough water lands on those surfaces to meet large demands. Road water is the opposite of roof water: plentiful, but dirty.
It is not only animal droppings. There is street rubbish, oil residues, exhaust particulates, brake lining dust, and everything traffic leaves behind. It wins at least the bronze gong for yukky water. But it is available in large quantities, and that matters.
Segregation: Why Utility Water Should Not Be Treated Like Drinking Water
One of the “holy cows” of the water industry is resistance to segregation. Traditionally, all public water has to be potable water. The twin-pipe idea is often raised and rejected, usually on cost, with the assumption that a second system must duplicate the existing premium reticulation network.
The arguments for segregation are overwhelming. We use a small volume of drinking-quality water and very large volumes where some contamination is acceptable—or even useful, such as irrigation water carrying nutrients. The key point is that segregation does not require duplicating the whole system.
Instead, we capture roadside runoff locally and store it in the ground, then distribute it as utility water for irrigation and general use.
Percolation Holes: Putting Dirty Water Where It Becomes Useful
Capturing roadside runoff can be as simple as boring holes through existing drainage pipes so water enters the subsoil. Where it goes depends on local hydrology: it may feed a local aquifer or emerge lower down a slope as a wet patch.
That water can then be captured in a small local dam or lake (which can become a community asset), or accessed via a bore to fill a local header tank when needed. Water filtered through the ground will not be drinking standard, but it can be perfectly adequate for irrigation and yard duties such as cleaning cars. It can be distributed locally using low-cost poly pipe under gravity.
Catching Water in Vegetated Lands
Percolation holes can be used not only beside roads but also in vegetated land. This applies to catchments and to general land areas. Many arid regions globally rely on water harvested in mountain regions. Australia has few serious mountains, but a lot of land. Some land will need to function as water catchment—not only via conventional dams (which mainly harvest large rains), but by storing water in the ground.
Globally, the water stored in soil exceeds the fresh water stored in dams and natural lakes. Even if only about 10% of soil volume is available for water storage, the sheer volume of soil makes total storage capacity enormous.
The problem is that most rain is captured in the surface layer and then evaporates away without running off or penetrating beyond the evaporation and transpiration layers (the effective root zone).
What You See After Rain: Small, Medium, and “Baby Streams”
After a small rain—only a few millimetres, say under 10 mm—water is absorbed and there is no surface water. It looks impossible to harvest, unless you use an impervious seal such as a road or plastic film that can direct water to storage.
With a medium rain—around 10 to 50 mm—you often see scattered puddles and depressions holding surface water. At the lower end of this band, these patches do not connect into flow. With more rain they can converge into what is not quite a stream, more a “tribulette”, which fills local hollows but may still not overflow into runoff.
It is common to see water run off slopes as a baby stream that stops flowing on flatter ground as water is progressively absorbed by soil. You can watch the same behaviour beside bush roads: water runs off the hard road into the drain, flows along until it finds a depression, veers off, and then simply stops as the soil drinks it up. Australia is full of these ephemeral creeks that start and stop. Shortly after rain stops, what remains evaporates.
Rainfall Per Person: Why the Waste Is So Large
If you divide the total rain that falls on Australia by the population, you get a staggering figure—around a million litres per person per day. Most of that arrives as the smaller and medium rains and is lost to evaporation. Distribution varies: larger rains dominate in the far north and far south, the dry centre relies on occasional big storms, and the coastal fringe (where most people live) receives a large share of medium rains that remain reliable even in drought and are expected to stay reliable under global warming. Some argue they may even increase with warmer seas, yet we largely waste them.
Micro Dams and Percolation Holes
We can capture useful amounts of water using micro dams and percolation holes. A micro dam is not designed to store water on the surface long-term. It will spend most of its time dry. Its role is to catch and hold water briefly and give it time to soak deep into the ground.
Some people argue that percolation holes “steal” water from others by reducing river flows. There is little substance to this. Percolation is slow, so during major rainfall there will still be runoff. Percolation holes mainly capture water from smaller rains that would otherwise evaporate away.
In much of Australia, the top layer of soil forms an insulating crust that protects deeper water from evaporation. Desert plants survive partly because of this. That crust has to be wetted out before there is runoff or deep recharge, and the water used to wet it is usually lost by evaporation. Percolation holes are simply a way of making better use of water that would otherwise be lost.
Wicking Beds: Local Storage Turned Into Subsurface Irrigation
Wicking bed technology started in a modest way, and only now are the implications being appreciated. I became involved through a project in Ethiopia, looking for ways to provide water to grow enough food in a country that can be racked by starvation.
Before I went, I had the wrong image. Ethiopia is not just dried desert and sand. The climate is not that different to Australia, with reasonable rainfall and agricultural production that can support a significant population on average. The problem is erratic rainfall. Crops can look fertile, then a break in rain at the critical time when seed heads should fill can leave people without food. Just a couple of weeks without rain at the wrong time can cause untold suffering.
The core problem is short-term local storage. The original solution was to increase the water holding capacity of soil by creating what is essentially an underground pond. In the earliest version, a channel was dug, lined with plastic film, and the soil replaced.
It was soon realised this could be extended into efficient subsurface irrigation by laying a pipe along the base. In Ethiopia, flood irrigation is often the only method, but flow rates are low, so water is lost soaking into the ground before it reaches the end of a furrow. With a wicking bed, it is like filling an underground bathtub. Water can trickle along the pipe without soaking away. The high flow rates needed for efficient flood irrigation are not required.
Wicking Beds With Rain Harvesting: Water Amplification
Wicking beds can also catch water locally by extending the plastic sheet so rainfall is funnelled into the base of the bed—water amplification. This is more than just increasing catchment area. The plastic film catches even small rains and directs water down to where it is protected from evaporation by the soil crust.
In conventional agriculture, small rains often only wet the surface and are lost quickly. Amplifying “wings” can multiply the effect of rainfall. At the other extreme, heavy rain can pass through soil beyond the root zone and be lost. A wicking bed captures much of that water and stores it for longer. With both small and large rains, wicking beds make more effective use of water.
Row crops could use this concept where there is no irrigation, or where irrigation is limited.
Wicking Beds and Grey Water
Most wicking beds built so far in Australia still rely on some external water source, even if only for top-ups. In the simplest case, people connect roof downpipes directly into a bed, removing the need for a tank, which is useful where space is limited. The disadvantage is there is no control over scheduling.
Grey water can provide a fairly constant supply, but demand from plants is not constant. Relying only on grey water can cause oversupply during rainy periods. Preventing grey water from escaping into the general environment is critical.
A better approach is to have enough wicking bed area so grey water provides only a small proportion of total water, with additional water added as needed. This also dilutes grey water, which is generally alkaline.
Why Productivity Often Increases
One fascinating feature of wicking beds is that they often produce higher productivity than conventional beds. This is largely an observation first; the theory comes later. There are several plausible reasons.
First, the wetting and drying cycle acts like a natural flood-and-drain system. Roots are wetted from underneath, and as water is used, air is pulled in from above, giving a breathing action to the soil.
Second, a wicking bed naturally creates a moisture gradient: dry near the top (apart from germination) and saturated lower down. Somewhere in that gradient is often an ideal water-to-air ratio for roots.
Third, successful wicking beds usually require better soil construction. If heavy clay is simply dug out and put back, it can waterlog and plants may die from anaerobic conditions. Many beds therefore include a layer of organic material and, ideally, a starter inoculant—worm capsules with food and a microbe mix—to keep the soil open and biologically active.
It is probably a combination of these factors that produces the increased productivity.
Large Scale Use
Wicking beds began as a small-scale storage idea, with irrigation efficiency as a bonus. It was once assumed they would be limited to home gardens and hobby farms. It is now clearer that simple, cheap subsurface irrigation could have much wider application in commercial agriculture.
Beds can be linked by cascading water from one to the next, or by fitting each bed with a simple cut-off valve. When one bed fills, water is diverted to the next, watering beds in sequence. In sloping country, beds must be level and aligned along contour lines, but cascade watering with cut-off valves can make them well suited to hills and may offer an alternative to large areas of flood irrigation.
Anticipatory Irrigation: Timing Water to Reduce Evaporation Losses
Anticipatory irrigation is a simple way of achieving the twin aims of making use of smaller rainfalls and minimising evaporation losses. The aim is to get water deep into the soil, protected from evaporation.
Just as there is a threshold for runoff in dams, there is a threshold volume of irrigation that must be applied before water penetrates into deeper soil. There is always an insulating crust that must be wetted first, and the water used to wet out this crust is often lost by evaporation within hours. Irrigators know they must apply enough water to fill the profile, extending the time between irrigations and reducing repeated threshold losses.
What is less obvious is that the best time to irrigate is often just after rainfall. The surface is already wet, so a smaller volume of water is needed to fill the profile. Sometimes rain is expected but plants need water now; then the aim is to apply only enough water to satisfy short-term needs. At other times extreme heat is forecast; it can be better to irrigate ahead of time rather than under high evaporation conditions.
Why Probes Alone Do Not Solve the Scheduling Problem
Soil moisture probes are widely used, but there are two intrinsic problems. First, they measure moisture close to the probe only. Moisture varies widely across the root zone, so readings depend heavily on probe position. Experts try to choose an “average” location, but this is far harder than it sounds.
The bigger problem is wetted volume. Irrigation systems rarely apply water uniformly; they wet only part of the root zone. This makes it difficult to calculate total soil water from a few sample points.
The Jar of Stones: The Simple Measurement Trick
The answer can be explained with a jar of stones that is already partly filled with water. A water expert might try to measure the current water and calculate the empty spaces between stones. The solution is almost childlike: simply measure how much water is needed to fill the jar. That tells you exactly how much was missing.
Applying the Same Idea to Irrigation Scheduling
We fill the soil profile and use probes to identify when the soil is “full”. More specifically, we measure how much water must be applied for moisture to reach the bottom of the root zone. We do not care how much water is in the soil at that point; we define that condition as full.
Then we let the plant use some water. We may not know exactly how much it used, but we can measure it by refilling the profile. The snag is that it takes time for water to soak down to the base of the root zone, so we cannot keep applying water until it is full without introducing errors.
The practical method is to make a best estimate of how much water has been used—guess a crop factor and multiply by evaporation—then apply that amount. We do not even have to start with a full profile. We guess, apply, and measure irrigation depth. We adjust the crop factor until, after applying the estimated water, the profile is full. Then, at any point, we know from evaporation how much water is needed to fill (or partially fill) the profile.
Guessing can be hit and miss, so we make it efficient using a predictor–corrector method built into simple software.
Basic Theory (In Plain Steps)
- We need the plant water use (crop factor) and the maximum allowable deficit in the soil.
- These are site specific; we cannot measure them directly, but we can learn them by monitoring.
- Start with cautious estimates of crop factor and allowable deficit.
- Measure evaporation, estimate current deficit, compare with allowable deficit, and decide whether to irrigate.
- After irrigating, measure soil moisture or irrigation depth and use this feedback to adjust crop factor.
- When crop factor stabilises, observe onset of plant stress to determine allowable deficit.
Software Routine (Overview)
In mid-cycle, you enter evaporation and rainfall, confirm the last irrigation is recorded, and review each block’s water status. A full profile is taken as zero; negative numbers indicate how much water is needed to refill. A threshold can flag when a block is ready to irrigate.
You then plan irrigation by selecting the blocks that need water and printing a schedule. After irrigating, you measure irrigation depth (or soil moisture) after allowing time for stabilisation, record the irrigation, and then calculate and record the revised crop factor. The crop factor trend graph helps show whether the system is settling.
If irrigation water is saline, the method of applying only enough water to refill the profile can lead to salt accumulation, so periodic flushing may be required. Finally, you review annual water use, because irrigation is not only about plant needs; it is also about juggling the amount of water available.
Conclusion
Smaller rains are not a nuisance. They are a valuable resource that we mostly lose to evaporation. Tanks harvest clean roof water but do not supply enough volume for food production. Road runoff is plentiful but dirty, and becomes useful when captured locally and filtered through soil as utility water. Micro dams and percolation holes turn brief surface water into deep storage. Wicking beds extend the same logic by creating practical underground storage for crops. Anticipatory irrigation completes the picture by timing irrigation so more water reaches deep soil when evaporation losses are lowest.
This is an overview of the basic routine now we have to do this in the real world learning the correct crop factor and water holding capacity of the soil without damaging the plants or wasting water in the process. This is shown in the video on anticipatory irrigation:
