PUBLISHED PAPERS
Reports

Summary
This report was prepared for the Angas Bremer Water Management Committee and forms part of the committee's strategy to mobilise a community who want best practice irrigation management in their district. FullStop Wetting Front Detectors were installed at depths of 50 cm and 100 cm on over 100 properties, and growers reported how deep their irrigation water penetrated and the salinity of the water captured by the detectors.

A total of 98 growers sent information back to the committee, detailing the amounts of water applied and when, the response of the wetting front detectors and data on salt content. 72 of these records, all from wine grape growers, were complete in every respect and were used to compile this report.

The average amount of irrigation used was 1.75 ML/ha, with 10 growers applying more than 2.5 ML/ha and 12 applying less than 1 ML/ha. The average number of irrigation events was 29 (the range was 8 to 78). The average application at one time was 7 mm, but this also ranged from an average of 1.2 to 16 mm per irrigation.

The shallower detector at 50 cm responded one or more times on 82% of properties (average of 10 times). The 100 cm deep detector tripped on one or more occasions on 44% of properties (average of 4 times). There was no relationship between the total amount of water applied in the season and the number of times the detectors responded. The detector response was much more influenced by the amount of water applied at one time. Simple modelling studies showed that the width of wetting patterns had a major influence on depth of water penetration and supported the data provided from the Wetting Front Detectors.

Salinity measurements were made by 26 of the 34 growers who activated a FullStop at 100 cm depth. In addition 10 growers who did not get the deep detector to respond returned salinity samples from the 50 cm depth FullStop. A total of 85 samples were measured at 100 cm depth having an average salinity of 2208 ppm (3.7 dS/m). A total of 23 samples were measured at 50 cm depth having an average salinity of 6500 ppm (10.8 dS/m). This data confirms that the FullStops are working as expected; if wetting fronts do not pass detectors, then salt must be accumulating above them. When they are activated, high concentrations of salt are measured.

Since the FullStop is a very new method for evaluating irrigation practice, the results above need to be supported by soil coring and salt measurements at different positions in the soil profile. Nevertheless some preliminary recommendations can be given. The proportion of growers who regularly activate the deep detectors could reduce their leaching fractions, but the majority need to factor leaching fractions into their irrigation management strategy. An increase or a decrease in the leaching fraction may involve changing only the timing and amounts of irrigation given at one time, not the annual total. Done correctly, salt can be moved to just below the active root zone where it can be stored, since water table levels are generally falling in the district.

The most fascinating insight from this research is the idea of irrigating to an upper and lower salt concentration as opposed to managing soil water content. For crops that are intentionally under-irrigated, precision in soil water monitoring is not required. For crops intentionally stressed at certain times, the soil water content data can be misleading, because the influence of salt increases as the water content declines.

Background
In the 1980s, the then Technical Services Unit (TSU) of the ACT Parks and Conservation Service embarked on a comprehensive program to save water on ovals and public places in the ACT. Their most notable achievement was to translate their considerable research experience into a computer based irrigation control system, called Comtrol. When Comtrol was launched in the early 90s it brought about large savings in water.

Ten years later Canberra Urban Parks and Places decided to review the operation of Comtrol, to see if it was still delivering up to expectations. The project was carried out in two phases. During Phase One, detailed soil water monitoring was carried out at OÕConnor oval for a six month period to establish if and how further water saving could be made. Monitoring was extended to two further ovals during Phase Two, whilst the work at OÕConnor was continued. The data were compared with precision measurements of cool season turf water use in Canberra conducted by CSIRO Land and Water.

Introduction
Efficient use of water in the agricultural sector is an issue of national importance in South Africa and the science required to achieve this goal is relatively mature. However, a recent survey among commercial farmers showed that they ranked irrigation scheduling as priority number four or five amongst their major concerns. Most farmers are prepared to admit that their system is not perfect, but at least it works. After a period of trial and error they have settled on management system that satisfied them and they need a good reason to re-evaluate it. Small-scale farmers were preoccupied with issues such as their access to land, water, credit and markets and showed little awareness of the importance of water use efficiency.

This project introduced a Wetting Front Detector to farmers with the purpose of stimulating a re-think about irrigation management on their farms. The Wetting Front Detector (WFD) was designed to be the simplest tool that could assist farmers to improve their understanding of irrigation. To achieve this aim, the wetting front detector must pass two tests. First, the device itself and how it works must make intuitive sense to farmers. It should be relatively easy to install and give "believable" results that challenge the farmer's perceptions. Second it must pass the accuracy test. We have to demonstrate that crops irrigated according to the principles of the Wetting Front Detector perform adequately against standard scientific procedures. A combination of research and extension was employed to satisfy these objectives.

Summary
Before this study, CSIRO Land & Water developed an irrigation scheduling tool called the FullStopTM. The device is buried in the soil and tells a farmer when a wetting front has reached the required depth in the soil and therefore when to stop irrigating. The device also collects and stores a sample from the wetting front that can be used to measure the concentration of solutes that are moving in the soil.

This study describes a one-year pilot project to determine if the water sample from the FullStop could be used to manage nitrate and salt in irrigated agriculture. The emphasis was on looking at the simplicity, robustness and potential usefulness to farmers. We examined the performance of the FullStop in several different crops and soil types under both drip and sprinkler irrigation. Case studies are presented from sugarcane in Bundaberg, capsicums in Gosford and tomatoes and turf in Canberra.

The FullStop responded to wetting fronts generated by irrigation and rainfall at all sites and almost always stored a soil solution for analysis. At Bundaberg the major leaching events were forced by rainfall, and since these events are not predictable it is important to keep nitrate levels as low as practicable. The high N levels recorded after fertigation at 0.5 m suggests to the manager that fertiliser application would be better in smaller, probably more frequent doses. Results from Gosford and Canberra showed how easily nitrate is lost from the topsoil early in the season. In both cases there was evidence that the plants receiving less water early in the season performed best because nitrate leaching was reduced.

The FullStop method highlighted the critical interaction between water and fertiliser management and provided useful feedback to the manager. This study concludes that the FullStop can be used to save water and improve the management of nitrate and salt in irrigated agriculture.