Water harvesting


A rangeland in good condition

When rain falls in the rangeland some will run-off. Some will be absorbed and will germinate annual plants.

The rain water will run-off exposed rock. It will run-off the soil where there is a lack of plant cover and the soil structure is poor. Even in good rangeland areas there will be some such areas.

The water will then flow down the slope. If the rainfall is heavy it will carry silt with it.

If the rangeland is in good condition the water will move only a few metres before it meets an obstacle.

This is usually a perennial plant - for example a tree, fodder shrub or a grass tussock. The plant will be surrounded by dead litter in the form of branches and leaves.

If this is the second or further rain it will run-off an even shorter distances as there will be a cover of annual plants that germinated after the first rain.

The water and silt will be slowed down and absorbed into the soil.

Depending on the amount of rainfall the water may continue further down the slope into a wadi bed.

In the wadi bed there will be more trees and litter to slow the water down.

If there is a great deal of rain the water will flow down the wadi.

The end point of the wadi can be a delta. When a wadi comes out of some hills for example. There is a change of slope that slows the water. The water is absorbed into the soil and recharges the aquifer.

Alternatively the water may flow into a salt pan where it evaporates.

Degraded rangeland

Here the run-off is much greater.

The soil over large areas is poorly structured due to a lack of organic matter.

It caps over and the rain runs-off . Little is absorbed.

Once the run-off begins there are few obstacles to slow it down.

The wadis are no longer full of trees and bushes and dead material. The water rushes on down the slope.

Can it be reversed?

It is extremely difficult to completely reverse the effects of rangeland degradation. In many cases the soil has washed away exposing bare rock.

Even the replacement of organic matter and soil structure is a difficult process.

The erosion of the soil has also removed many plant nutrients and these can be expensive to replace in the form of fertilisers.

Physical action

Physical action is aimed at replacing the run-on areas with new man-made barriers rather than the naturally obstacles that have been destroyed by over-exploitation (grazing and firewood collection).

In the longer term it is hoped that these artificial run-on areas will be colonised with plants that will form more permanent obstacles to the flow of water.


Ripping was one of the first means of breaking the hard crust of degraded rangeland soils in an attempt to reduce run-off and create areas where seed could germinate.

The problem with ripping was that the rips formed channels for the water to flow. These channels could become gullies and create further erosion.

One solution was to break the rips into pits (see below) and the other was to rip on the contour.

Ripping on the contour on the vast areas of the rangeland required a level to be fitted to the tractor.

To survey and mark contours is too slow and too expensive.

The level is simple. A plastic hose is fitted with two clear plastic tube. They are mounted front and back on the tractor. In fact it is better to mount one in front of the driver and one on a frame extending a few metres in front of the tractor.

The tube is filled with coloured water.

The tractor is parked in a level place and the level of the water is marked.

When the tractor is driven across the rangeland it is possible steer it up or down the slope to keep it level.


Early machines

The first pitting machines that I have discovered were built in the Adelaide Hills by a small machinery company at Woodside about fifty years ago. The owner worked with the owner of a large rangeland property to develop a disc and tined pitter.

As with so many farming inventions they were too busy doing the work to record their invention for the historical record. Perhaps there were earlier machines but I have not been able to find any record of them.

The principle of the machine is extremely simple. A single disc or tined implement (mounted on wheels and drawn by a tractor - not mounted on the linkage) is modified by mount the wheels in an eccentric manner on the axle.

This means that the axle and turning point point of the wheel is not at the centre of the wheel.

When the implement is drawn forward by the tractor the frame of the implement moves up and down as the wheels turn.

To dig pits in the rangeland the depth of the disc or tine is adjusted until it is just touching the soil when the wheels have lifted the implement to the UP position.

The tractor now moves forward.

As the tractor moves forward the wheel turns and the implement is lowered into the ground. When the wheel has completed half a rotation the implement is in the fully DOWN position.

The tractor continues forward. The implement now rises.

Each complete turn of the wheels raises and lowers the implement.

If the implement were permanently in the ground it would rip a trench.

The UP and DOWN movement means that the trench is not continuous but is a series of pits.

As well as the disc and tine versions of the pitting plough there are those with mouldboards and some with double mouldboards like a potato ridger.

Linkage mounted machines

There are now linkage versions of the pitter that work with complex hydraulic (also known as a Vallerani plough). I do not know whether there are any significant improvement in the linkage version that justify its considerably higher cost.

This is the tractor mounted pitter with a large ICARDA audience. The pits can be seen in the front of the photo. There on attempt to seed the pits. Given the degraded nature of the surrounding rangeland seeding is essential. It can be done by hand but a seeder/pitter would be more economical.

Seeder and pitter

This is a great advance. The original pitting carried out in Australia 50 years ago relied on wind borne seed to lodge in the pits.

In WANA the rangeland is so degraded that there is little seed. The pits must be seeded and preferably fertilised with a small amount of phosphate.

These pitter/seeders are built by a company in Western Australia.

The best form of seeding is with medic pods and I am not sure that the seeder can handle pods.

Contour banks

Contour banks have been used in the Marginal Zone and Cereal Zone. The difficulty with their use in the rangeland is their cost.

They are surveyed and then built using a bulldozer.

There are interception banks that hold the water in a large pond to soak in and there are banks built on a slight slope that lead the water slowly down to a water way (usually a wadi).

Given the high cost of contour banks they are usually linked with pitting or contour ripping.

Hand made barriers.

In the WANA region a number of projects have employed local farmers and their families to build various barriers of stone and earth.

Hand made water harvesting areas.

Wadi beds

The natural barriers (trees and dead material) in wadi beds have disappeared. Artificial barriers of stone have been built to slow down the flow of water and cause it to spread on the surrounding land.

These are most successful when the wadi leaves the hills and forms a delta.

In Libya the Wadi Karouba project was a most successful example of spreading the water over the delta through a series of barriers and the filling in of deep wadis.

A wadi barrier.


The physical action on the rangeland is only to create an environment for pasture.

Medic pods can be sown in the pits.

On the wadi delta land the Libyans found that medics from the Cereal Zone grew well. They needed cultivars with plenty of hardseed.

The delta areas are prone to "false breaks." That is the medic will germinate with a single flooding but not necessarily produce seed unless there is another flooding.

Hard seed allows the medic to survive these false breaks.


The addition of phosphate is an essential ingredient of rangeland pasture development.