Grains Research and Development

Date: 01.08.2004

Western Plains project seeks to explain yield differences

Soil map of the district showing the locations of the project"s sites.

By Alan Palmer

The role of Precision Agriculture in the Western Plains of NSW is to be examined in a NSW Agriculture project, supported by GRDC, AGnVET Services and Charles Sturt University. A team of 14 professional staff, including agronomists, a biometrician, an economist, engineers, a mathematician, managers and spatial analysts, will work on the project.

These skills are augmented by the team"s interaction with other teams involved in GRDC"s Initiative SIP09: Precision Agriculture - Managing Variability.

Many farmers in the Western Plains region - an area bounded by Dubbo, Nyngan and Coonamble - have been collecting yield maps for several years but few, if any, have made practical use of them. This is hardly surprising due to the spatially variable soils and seasonally and spatially variable rainfall that characterise the region. This variability makes it difficult to understand the yield differences appearing in the maps even before the effects of disease, frost and other, sometimes undetected, events are added.

|If a grower cannot adequately explain why the variability is in last year"s yield map, how can that map be a basis for planning for the next crop? It is with this question in mind that the project seeks to provide an improved basis for precision agriculture. We are starting from the premise that in the absence of limitations due to plant population, nutrition or disease, yield is determined primarily by available water. The project will focus on the estimation of plant available water in the coming crop season so that seed and fertiliser rates can be optimised.

The amount of water available to a crop depends on soil type and condition, previous paddock history and rainfall. The widespread adoption of conservation farming means that run-off and run-on due to within-paddock differences in elevation are mostly negligible on the low slopes found in the region.

The major soil types in the area have traditionally been grouped into red soils and grey soils and managed accordingly. Most of the soil in the red soil group is hard setting, has low plant available water capacity (PAWC), and can be sodic.

However, interspersed within this group is a range of soils with higher PAWC, non-hard setting and non-sodic. Compounding this variability is the presence of grey soils that range from the physically intractable with severe production constraints to some of the most productive soils in the region.

We have sites at Claremont near Nyngan, Waverly, south of Trangie and the Trangie Agricultural Research Centre (TARC). The district soil map above shows that these sites are in quite different soil zones.

Soil variability in the region occurs at district, farm and often within-paddock scale. We are therefore looking for ways in which growers can identify zones of each soil type and estimate the useful depth of soil within paddocks. This will effectively measure the PAWC of each zone and tell the grower how big the “water bucket” is in each part of the paddock.

One of the tools we are testing to map soil type and condition is a tine dynamometer. This measures the force required to push a point through the soil. This force depends on soil type, compaction and moisture content - all of which have some bearing on the size of the soil water bucket, how much water is in the bucket or how quickly the bucket can be filled. We have used the dynamometer to map our sites this year.

The map of the Claremont site indicates that the eastern side of the paddock is different to the western side. The stripes in the map of draught force are in the same direction as stripes in the yield map of the paddock. This gives us some confidence that mapping draught force will be useful in predicting potential crop performance. In the near future, we will look at mapping draught force at depths greater than the 75mm used this year. Force at greater depths may give useful information on the presence of plough pans and other impediments to water and root entry.

The next step is to find out how much water is actually in the bucket. At present, the simple push-probe is probably the most useful tool for this before sowing. Available water will be directly measured at planting, anthesis and harvest by push coring to 1.5m.

This will be correlated with remote sensing tools such as EM31, dynamometer and satellite images to see how well these technologies can predict plant available water. In cooperation with other SIP09 teams, we are also looking for ways in which this might be automated so that plant available water might be measured on the go at sowing time and eventually used to control variable seed and fertiliser rates.

The other major focus of the project is monitoring the crop to provide a basis for in-crop treatments such as post-emergence fertiliser application or chemical spray to control weeds or disease. We will look at the economics and practicalities of using airborne and satellite images for this monitoring.

A unique aspect of this project will be the use of an array of rain gauges at each site to quantify rainfall variability. This, coupled with extensive use of neutron probes, will provide the basis for evaluating the usefulness of the remotely acquired images. Work on monitoring will start during this year"s winter crops.

Most of the techniques that we are studying will eventually be available to growers through contractors or consultants. Final analysis of the project will look at the economic worth of techniques. We expect that this analysis will tell growers the maximum price they should pay for these services.

For more information:
Alan Palmer, 02 6880 8058,

GRDC research code: DAN 00054